VIRTUALHR - AUTOMATED AI-BASED RECRUITMENT FRAMEWORK

FOR STREAMLINING HIRING PROCESSES

A PROJECT REPORT

Submitted by

AMSAVARTHAN B (814821104005)

CHANDRU MK (814821104008)

VIGNESH S (814821104040)

in partial fulfilment for the award of the degree of

BACHELOR OF ENGINEERING
IN
COMPUTER SCIENCE AND ENGINEERING UNIVERSITY COLLEGE OF
ENGINEERING – ARIYALUR

(A Constituent College of Anna University, Chennai)

ANNA UNIVERSITY::CHENNAI 600 025

MAY 2024
 ANNA UNVERSITY : CHENNAI 600 025
BONAFIDE CERTIFICATE

Certified that this project report entitled, “VIRTUALHR-AUTOMATED AI-

BASED RECRUITMENT FRAMEWORK FOR STREAMLINING HIRING PROCESSES”

is the bonafide work of AMSAVARTHAN B (814821104005), CHANDRU MK

(814821104008), VIGNESH S (814821104040) who carried out the project work under

my supervision.

SIGNATURE SIGNATURE

Dr.S.JAYANTHI, M.Tech., Ph.D., Mrs.M.SUGANTHI, M.E.,

HEAD OF THE DEPARTMENT SUPERVISOR

ASSISTANT PROFESSOR ASSISTANT PROFESSOR
Department of Computer Science & Department of Computer Science &
Engineering, Engineering,
University College of Engineering, University College of Engineering,
Ariyalur – 621 731. Ariyalur – 621 731.

Submitted for the University Viva voce Examination held on …………………

INTERNAL EXAMINER EXTERNAL EXAMINER

 ACKNOWLEDGEMENT

We express our breathless to our Dr.V.Venkatesan, M.E, Ph.D, Dean(i/c), University

College of Engineering, Ariyalur, for giving constant motivation in succeeding in our goal.

We acknowledge our thanks to Dr.S.Jayanthi, M.TECH, Ph.D, Head of the

Department, for giving us valuable suggestion and help towards us throughout this project.

We are graceful to our project coordinator, Mr.G.Gogulkarthik, M.TECH, (Ph.D),

Department of computer science and Engineering, University college of Engineering.
Ariyalur, for the coordinating us throughout this project.

We express gratitude to our project guide, Mrs.M.Suganthi, M.E., Department of

computer science and Engineering, University college of Engineering. Ariyalur, for this
proficient guidance to consummate our project.

We are very much indebted to thank all the faculty members of Department of Computer
science and Engineering in our institute, for their excellent moral support and suggestions to
complete our project work successfully.

On the whole we express our heartfelt gratefulness to our parents, sisters and friends
who helped as either directly or indirectly in completing this dissertation.

AMSAVARTHAN B
CHANDRU M K
VIGNESH S
i
 ABSTRACT

Recruitment process plays a crucial role in helping organizations build a

skilled workforce that supports business growth. However, HR professionals often face

challenges, especially when hiring for multiple roles simultaneously. To address these

challenges, this project proposes an AI-driven recruiting platform aimed at improving the

efficiency and fairness of the hiring process. By leveraging artificial intelligence and

machine learning, the platform automates key recruitment tasks such as candidate

sourcing, screening, and engagement. The proposed system features an automatic resume

analysis tool that matches candidate profiles with suitable job postings, using content-

based filtering to provide employers with valuable recommendations. The subsequent

stages of the automated virtual interview process include an aptitude test, a programming

skill assessment, and a video interview. Natural language processing (NLP) is employed

to interpret and respond to user voice inputs in a conversational manner, while an

Attention Mechanism is used for behavioural prediction during the visual interview. The

platform analyses candidate interactions during the interview, such as gaze and head

rotations, to assess their attention and engagement with multiple interviewers. This

information is then provided to HR professionals, who use these insights to make data-

driven decisions during the final selection process. By incorporating advanced

technologies, the system accelerates recruitment workflows and offers a modern, data-

driven approach to evaluating candidates, particularly in the context of software

engineering job searches and resume enhancements.

ii
 TABLE OF CONTENTS

CHAPTER NO. TITLE PAGE NO.

ACKNOWLEDGEMENT i
ABSTRACT ii
LIST OF TABLES iii
LIST OF FIGURES v
LIST OF ABBREVIATIONS vi

1 INTRODUCTION 1
1.1 Overview 1

1.2 Problem Statement 4

1.3 Artificial Intelligence 4
1.4 Aim and Objective 7
1.5. Literature Survey 8

2 SYSTEM ANALYSIS 13
2.1 Existing System 13
2.2 Proposed System 14

3 SYSTEM REQUIREMENTS 17
3.1 Hardware Requirements 17
3.2 Software Requirements 17

iii
 TABLE OF CONTENTS

CHAPTER NO. TITLE PAGE NO.

4 SOFTWARE DESCRIPTION 18
4.1 Python 3.8 18
4.2 MySQL 5 21
4.3 WampServer 2i 22
4.4 Bootstrap 22
4.5 Flask 23

5 SYSTEM DESIGN 24
5.1 System Architechture 24
5.2 Data Flow Diagram 25
5.3 UML Diagram 27

6 SYSTEM IMPLEMENTATION 30
6.1. List of Modules 30

7 SYSTEM TESTING 36
7.1. Testing Definition 36
7.2. Test Cases 37
7.3. Test Report 41

8 SCREENSHOTS 42
9 CONCLUSION 47
10 FUTURE ENHANCEMENT 48
APPENDIX 49
REFERENCE 58
iv
 LIST OF FIGURES

FIGURE NO. FIGURE NAME PAGE NO

5.1 SYSTEM ARCHITECTURE 24

5.2 LEVEL 0 DATA FLOW DIAGRAM 25

5.3 LEVEL 1 DATA FLOW DIAGRAM 26

5.4 LEVEL 2 DATA FLOW DIAGRAM 26

5.5 USE CASE DIAGRAM 27

5.6 CLASS DIAGRAM 28

5.7 SEQUENCE DIAGRAM 29

8.1 HOME PAGE 42

8.2 CANDIDATE RESISTRATION PAGE 42

8.3 CANDIDATE LOGIN PAGE 43

8.4 JOB PROVIDER LOGIN PAGE 43

8.5 ADMIN LOGIN PAGE 44

8.6 RESUME FILLING PAGE 44

8.7 JOB PROVIDER DASHBOARD 45

8.8 ADMIN DASHBOARD 45

8.9 APTITUDE ROUND 46

8.10 HR ROUND BY HRBOT 46

v
 LIST OF ABBREVIATIONS

S.NO. ABBREVIATIONS EXPANSION

1 AI ARTIFICIAL INTELLIGENCE

2 HR HUMAN RESOURCES

3 NPL NATURAL LANGUAGE PROCESSING

4 GPL GENERAL PUBLIC LICENSE

5 ML MACHINE LEARNING

6 SSD SOLID STATE DRIVE

7 RAM RANDOM ACCESS MEMORY

8 SQL STRUCTURED QUERY LANGUAGE

9 GUI GRAPHICAL USER INTERFACE

10 BSD BERKLEY SOFTWARE DISTRIBUTION

11 NTLK NATURAL LANGUAGE TOOLKIT

12 PIL PYTHON IMAGING LIBRARY

13 CV COMPUTER VISION

14 API APPLICATION PROGRAMMING

INTERFACE

vi
 CHAPTER 1
INTRODUCTION

1.1. OVERVIEW
An interview is a meeting of people in a face-to-face situation. In common
parlance, the word interview refers to a one-on-one conversation. This is done with one
person acting in the role of interviewer and another in the role of the interviewee. The
interviewer is the person taking the interview, he/she asks the questions and seek an answer
to the questions. The interviewee is the respondent and answers all the questions.

1.1.1. Importance of Interviews

Interviews make all of us nervous since these are the deciding factor in whether
you will get the job or not. We will be discussing the details of an interview in this article.
Interviews make an important part of the recruitment process. They are meant to assess the
candidate on different parameters including the following:
Is the candidate the right fit for the organization’s work culture?
This is an important point of judgement since the candidate must be a fit in the
organization’s work culture. This ensures that there are no fundamental conflicts between the
company and employees. It also helps the candidate to easily adapt to the company policies
and guidelines. This ensures a healthy work environment.

Does the candidate have the required skillset?

Through different interview rounds, you will be able to assess if the candidate
has the required skill set for the job profile. This is the most important part of the assessment.
It ensures that the company is hiring the most capable employee for that job role.

Is the resume in alignment with the candidate’s profile?

Through various stages of the interview process, the recruitment team can tell if
the candidate’s skillsets and experience justify the resume. The team will be able to pinpoint
the areas of conflict while assessing the candidate through different interview rounds.

1.1.2. Software Engineer Interview

A software engineer interview is a process by which potential employers get to
know a candidate to see if they're a good fit for a software engineering role. This typically
involves answering questions about your work history, experience with different technology
and personality, plus tests that determine your skill level for specific programming languages

1
or software creation. The type of questions and the number of interviewers often depend on
the size and purpose of the company and the level of the role for which you're applying.

1.1.3. Types of software engineer interviews

Most software engineer interviews consist of a few different types of interview
sections that help employers get a fuller understanding of your ability to fit into their job
opening. Examine the following types of interviews you may encounter during the software
engineer hiring process to help you better prepare

• Phone screens
The phone screening portion of a software engineer interview usually begins
with a brief conversation with a recruiter. The recruiter may ask you for information about
your interest in the role and discuss the basic requirements. Then, the recruiter helps you
schedule the following interview for the company, which is often another phone screen but
with the hiring manager. This section of the interview process allows you to ask more specific
questions to learn about the company, the role and the responsibilities.

• On-site personality questions

After talking with the hiring manager, they may schedule a series of on-site
interviews for you, and one portion of those interviews may involve questions about your
personality. This shows employers how you may act in the workplace and whether you'd be
a good match for their work environment. The interviewer may split personality questions
into these two sections:

• Behavioural questions:
Behavioural interview questions are often similar across industries because their
purpose is to determine how a candidate may behave in the workplace. These questions may
ask about what you would do in certain situations, how you view your skills and what you
find important about work.

• Company culture-fit questions

Company culture-fit interview questions often differ from one company to
another because they're specific to the culture and work environment of the company to which
you're applying. An interviewer may ask you culture-fit questions regarding what you want
from a workplace and how you value the work the company does.

2
 • On-site skill tests
Many software engineer interviews also require on-site skills tests for candidates
to ensure you can complete the work of the role effectively. Various employees from the
company may ask you questions about the work they do and supervise you as you complete
coding problems. Although the specifics of the skill tests depend on the company and the
role, they often involve these topics:

• Problem-solving
It's typical for a software engineer interview to include coding problems for
candidates to solve in front of interviewers so potential employers can understand your
thinking process, strengths and applied knowledge. During this portion of your interview,
you may solve two to three coding problems while explaining your work to interviewers.

• Software architecture
Another test for many software engineering roles is how you understand and
apply knowledge about software architecture. For these tests or questions, interviewers may
expect you to describe how you would design a system when given a broad concept and a set
of parameters or limitations to keep your idea within.

• Language knowledge
With so many programming languages with which you can code, interviewers
may want to test your ability to use a specific language or multiple languages. If the job for
which you're applying requires knowledge of a specific language, you may answer questions
about your experience with that language and solve problems using that language.

Recruitment and Staffing

Staffing and recruiting cannot exist without each other, but they are not the same.
Recruiting is the process of searching and obtaining prospective candidates for the job. The
best candidate is chosen through the selection process, which is a part of staffing. Companies
can use internal or external recruiting teams to attract, engage and hire the top candidates. In
internal recruitment, current employees can be offered promotions or other positions within
the organization. External recruitment methods include job postings, employee referrals,
recruitment programs and more. Staffing encompasses all the processes involved in building
and retaining the workforce.

3
1.2. PROBLEM DEFINITION
Human Resource (HR) professionals are integral to organizational success,
managing tasks like recruitment, training, and performance evaluation. Yet, recruitment and
selection remain particularly challenging. Identifying suitable candidates from a vast
applicant pool demands meticulous screening to ensure alignment with job requirements and
adaptability to technological advancements. Traditional methods, such as telephonic
interviews, though cost-effective, often lead to inefficiencies, especially when candidates are
unwilling to relocate or lack genuine interest, resulting in wasted time and resources.

The urgency to fill positions swiftly, especially amid rapid company expansion
and multiple concurrent projects, exacerbates these challenges. Time constraints may compel
HR managers to expedite screening processes, increasing the risk of hiring underqualified
candidates, which can lead to early terminations and further resource drain. Moreover, the
absence of standardized, automated systems can introduce biases and inconsistencies in
candidate evaluations, compromising fairness and transparency.

To address these issues, integrating Artificial Intelligence (AI) into HR practices

has emerged as a transformative solution. AI-driven tools can automate resume screening,
candidate shortlisting, and even conduct virtual interviews, enhancing efficiency and
reducing human biases. Platforms like Virtual HR utilize advanced technologies to streamline
recruitment processes, offering features such as real-time workflow automation, personalized
candidate interactions, and data-driven decision-making. These tools not only expedite hiring
but also ensure a more objective and fair evaluation of candidates.

1.3. ARTIFICIAL INTELLIGENCE

Artificial intelligence (AI) is the theory and development of computer systems
capable of performing tasks that historically required human intelligence, such as recognizing
speech, making decisions, and identifying patterns. AI is an umbrella term that encompasses
a wide variety of technologies, including machine learning, deep learning, and natural
language processing (NLP). At the simplest level, machine learning uses algorithms trained
on data sets to create machine learning models that allow computer systems to perform tasks
like making song recommendations, identifying the fastest way to travel to a destination, or
translating text from one language to another. Artificial Intelligence History
The term artificial intelligence was coined in 1956, but AI has become more popular today
thanks to increased data volumes, advanced algorithms, and improvements in computing
power and storage.

4
Ways of Implementing AI

• Machine Learning
It is machine learning that gives AI the ability to learn. This is done by
using algorithms to discover patterns and generate insights from the data they are exposed
to.
• Deep Learning
Deep learning, which is a subcategory of machine learning, provides AI with the
ability to mimic a human brain’s neural network. It can make sense of patterns, noise, and
sources of confusion in the data.

1.3.1. Content based Filtering

Content-based filtering is a type of recommendation system that is able to give
each user a very personalized item recommendation. Content-based filtering in recommender
systems recommends items to users based on their previous actions or preferences. It analyzes
item metadata to identify items with similar characteristics to those that the user has interacted
with before. This approach examines the characteristics of the items users have expressed an
interest in to recommend similar items, unlike collaborative filtering, which finds similarities
among users. Content-based filtering is widely used in e-commerce, news feeds, music, and
movie recommendations.

1.3.2. Natural Language Processing

Natural Language Processing (NLP) is a branch of Artificial Intelligence
focused on the relationship between human language and machines, based on the use of
natural language. Both NLP and Machine Learning are part of AI, and both share techniques,
algorithms, and applications. In fact, Chatbots with Artificial Intelligence involve a
combination of natural language processing and Machine Learning. What is expressed in
writing or orally generates a large amount of data and information. Humans can communicate
and interpret this information through the tone, the structure of a sentence, idioms, and other
linguistic elements, and the choice of certain words, expressions, and punctuations. And
machines can understand that complex set of data and information through natural language
processing.

Basic NLP tasks include tokenization and parsing, lemmatization/stemming,

part-of-speech tagging, language detection and identification of semantic relationships.

5
 Content categorization: - A linguistic-based document summary, including search
and indexing, content alerts and duplication detection.
Topic discovery and modelling: - Accurately capture the meaning and themes in text
collections, and apply advanced analytics to text, like optimization.
Contextual extraction:- Automatically pull structured information from text-based
sources.
Sentiment analysis :- Identifying the mood or subjective opinions within large
amounts of text, including average sentiment and opinion mining.
Speech-to-text and text-to-speech conversion: - Transforming voice commands into
written text, and vice versa.
Document summarization: - Automatically generating synopses of large bodies of
text.
Machine translation: - Automatic translation of text or speech from one language to
another.

1.3.3. Attention Mechanism

Attention is one of the most researched concepts in the domain of deep learning
for problems such as neural machine translation and image captioning. There are certain
supporting concepts that help better explain the attention mechanism idea as a whole, such
as Seq2Seq models, encoders, decoders, hidden states, context vectors, and so on.

Attention in Sequence to Sequence Models

The Seq2Seq learning model is what gave rise to the attention mechanism. A
better explanation is that attention was introduced to resolve the main issue with Seq2Seq
models. To begin with, Seq2Seq models utilize the encoder-decoder architecture to solve a
problem, be it translating a sentence or identifying the elements of an image.
Long input sequences and images with more than one element are often difficult
for these models to process accurately. Each element of an input sequence is turned into a
hidden state in an encoder to be fed into the next element. During the decoding process, only
the last hidden state with some weighted component is used to set the context for the
corresponding element of the output sequence.
With an attention model, the hidden states of the input sequence are all retained and utilized
during the decoding process.

6
1.4. AIM AND OBJECTIVE

Aim:
The aim of the project is to develop an advanced recruiting platform, named
Virtual HR, that harnesses the power of artificial intelligence and machine learning to
streamline and optimize the candidate recruitment process for software engineering positions.
The project seeks to enhance efficiency, reduce biases, and introduce a data-driven dimension
to the entire recruitment workflow.

Objective:
• To design and implement an AI-driven platform for streamlined candidate recruitment
in software engineering roles.
• To implement automatic resume analysis using content-based filtering to match
candidates with relevant job postings efficiently.
• To integrate NLP for conversational interactions and attention mechanisms for
behavioural prediction during virtual interviews.
• To develop automated aptitude and programming skill assessments for comprehensive
candidate evaluation.
• To quantify candidate attention-sharing, including gaze and head rotations during
video interviews.
• To generate reports for HR professionals to make informed, data-driven decisions.
• To mitigate biases in the recruitment process, fostering fairness and diversity.
• To enhancing user experience for candidates and HR professionals through a user-
friendly interface.
• To tailor features and assessments specifically for the unique requirements of software
engineering positions.

7
1.5. LITERATURE SURVEY

[1] Title: AI Hiring with LLMs: A Context-Aware and Explainable Multi-

Agent Framework for Resume Screening

Authors: De Boni, Annalisa, Giovanni Ottomano Palmisano, Maria De

Angelis, and Fabio Minervini.
Resume screening is a critical yet time-intensive process in talent
acquisition, requiring recruiters to analyze vast volume of job applications while
remaining objective, accurate, and fair. With the advancements in Large Language
Models (LLMs), their reasoning capabilities and extensive knowledge bases
demonstrate new opportunities to streamline and automate recruitment workflows. In
this work, we propose a multi-agent framework for resume screening using LLMs to
systematically process and evaluate resumes. The framework consists of four core
agents, including a resume extractor, an evaluator, a summarizer, and a score formatter.
To enhance the contextual relevance of candidate assessments, we integrate Retrieval-
Augmented Generation (RAG) within the resume evaluator, allowing incorporation of
external knowledge sources, such as industry-specific expertise, professional
certifications, university rankings, and company-specific hiring criteria. This dynamic
adaptation enables personalized recruitment, bridging the gap between AI automation
and talent acquisition. We assess the effectiveness of our approach by comparing AI-
generated scores with ratings provided by HR professionals on a dataset of anonymized
online resumes. The findings highlight the potential of multi-agent RAG-LLM systems
in automating resume screening, enabling more efficient and scalable hiring workflows.

8
 Advantage
• AI-powered systems can handle a large volume of applications efficiently,
making them suitable for companies with high recruitment needs or large-scale
hiring campaigns.

Disadvantage
• A multi-agent framework with RAG can be complex to understand,
configure, and troubleshoot, potentially requiring specialized technical
expertise.

[2] Title: Resume Evaluation through Latent Dirichlet Allocation and Natural
Language Processing for Effective Candidate Selection

Authors: Vidhita Jagwani, Smit Meghani, Krishna Pai, Sudhir Dhage.

With the increasing number of job applicants, automated resume rating has
become a necessity for recruiters. In this paper, we propose a method for resume rating
using Latent Dirichlet Allocation (LDA) and entity detection with SpaCy. The proposed
method first extracts relevant entities such as education, experience, and skills from the
resume using SpaCy's Named Entity Recognition (NER). The LDA model then uses
these entities to rate the resume by assigning topic probabilities to each entity.
Furthermore, we conduct a detailed analysis of the entity detection using SpaCy's NER
and report its evaluation metrics. Using LDA, our proposed system breaks down
resumes into latent topics and extracts meaningful semantic representations. With a
vision to define our resume score to be more content-driven rather than a structure and
keyword match driven, our model has achieved 77% accuracy with respect to only skills
in consideration and an overall 82% accuracy with all attributes in consideration.

9
 Advantage
• Unlike simple keyword-based screening, this approach can identify relevant
candidates even if they use different terminology or describe their skills in a more
narrative format.
Disadvantage
• The topics identified by LDA are often latent and might require human
interpretation to understand their practical relevance to specific job roles.

[3] Title: An Ai Based Talent Acquisition And Benchmarking For Job

Authors: Rudresh Mishra, Ricardo Rodriguez, Valentin Portillo.

Unlike traditional recruitment methods, such as employee referrals,

CV screening, and face-to-face interviews, AI is able to find patterns unseen by a
human eye. It could be used to find the right person for the perfect role faster and
more efficiently than ever before. In order to rapidly improve talent management
and take full advantage of the power and potential AI offers, we need to shift our
focus from developing more ethical HR systems to developing more ethical AI.
McKinsey’s Global Institute model predicts that approximately 70 percent of
companies will adopt some form of AI by 2030. When it comes to identifying
talent or potential, most organizations still play it by ear. Recruiters spend just a
few seconds looking at a resume before deciding who to “weed out” [1]. Often
when hiring is made it’s very important to know the current strength of the
organization and based on it if hiring is made a candidate is referred to be a good
fit for an organization [2]. There’s increasing evidence that AI could overcome
this trade-off by deploying more dynamic and personalized scoring algorithms
that are sensitive as much accuracy as to fairness to an organization.

10
 Advantage

• This allows recruiters to focus on more strategic activities like

building relationships with candidates and refining hiring strategies.

Disadvantage

• AI might have difficulty understanding nuanced language, subtle

details in resumes, or the context of certain experiences.

[4] Title: A Comprehensive Survey of Artificial Intelligence Techniques

for Talent Analytics

Authors: Le Zhang, Yihang Cheng, Rui Zha, Dazhong Shen.

In today’s competitive and fast-evolving business environment, it is a

critical time for organizations to rethink how to make talent-related decisions in
a quantitative manner. Indeed, the recent development of Big Data and Artificial
Intelligence (AI) techniques have revolutionized human resource management.
The availability of large-scale talent and management-related data provides
unparalleled opportunities for business leaders to comprehend organizational
behaviours and gain tangible knowledge from a data science perspective, which
in turn delivers intelligence for real-time decision-making and effective talent
management at work for their organizations. In the last decade, talent analytics
has emerged as a promising field in applied data science for human resource
management, garnering significant attention from AI communities and inspiring
numerous research efforts. Specifically, we first provide the background
knowledge of talent analytics and categorize various pertinent data.

Advantage

• AI algorithms can analyze job descriptions and candidate profiles in

detail.

Disadvantage

• Human oversight and validation remain crucial.

11
[5] Title: A Review of Artificial Intelligence Based Platform in Human
Resource Recruitment Process

Authors:Mohd Fakhri, Mat Saad, Rajermani

Thinakaran, Mohammad Baijed.

This study focuses on three questions: What are the AI based platforms
available? What types of AI features were used? It employs systematic literature
review (SLR) to 29 websites and literature manuscripts gathered from online
search engines. The AI based platforms were distributed on cloud technology via
software-as-a-service (SaaS). Platforms reviewed were utilized in the sourcing,
screening and selecting phase. Key activities such as profile matching, screening,
profiling, rating and ranking, interview configuration and performance reporting
are regarded to be prominently supported by the platforms.The study found that
platforms were highly utilized in the initial and midway phases while less
utilization observed in the ending phase of the recruitment process.

Advantage

• Automation and better candidate matching can lead to reduced

recruitment costs and lower turnover.

Disadvantage

• Some AI platforms may lack the flexibility to be fully customized to an

organization's specific needs and workflows.

12
 CHAPTER 2
SYSTEM ANALYSIS

2.1. EXISTING SYSTEM

The current job recruitment process in HR often relies on manual
and time-consuming methods, presenting various challenges for both HR
professionals and candidates.
• Resume Screening
HR professionals manually review and screen numerous resumes to
shortlist potential candidates. This process can be subjective and prone to biases.
• Job Posting and Application Submission
Employers post job listings on various platforms, and candidates
submit their applications. The matching process is typically based on keywords
and manual assessments.
• Manual Shortlisting
HR professionals manually shortlist candidates based on their
resumes and qualifications, which can be time-consuming and may result in
oversight.
• Interview Scheduling
Scheduling interviews involves manual coordination between HR
professionals, recruiters, and candidates, leading to potential delays and
scheduling conflicts.
• Traditional Interviews
Interviews are conducted in person or over the phone, relying on
subjective evaluations. The process lacks standardized assessments and
predictive behavioural analysis.
• Decision-Making:
HR professionals make hiring decisions based on a combination of
resume evaluations, interviews, and their judgment, introducing potential biases
and limiting the depth of candidate insights.
• Communication
Communication with candidates about interview schedules and
outcomes is typically done manually, which can lead to delays and lack of
transparency.

13
2.1.1. Disadvantages

• Manual resume screening is time-consuming and prone to oversights and

biases.

• Subjective shortlisting may overlook diverse candidate skills and

experiences.

• Inefficient interview scheduling leads to delays and conflicts.

• Lack of standardized assessments in traditional interviews hinders

objective evaluations.
s
• Biases in decision-making based on subjective evaluations can impact
fairness.

• Manual communication may result in delays and a lack of transparency.

• Limited automation contributes to a slower and error-prone recruitment

process.

• Absence of predictive behavioural analysis in interviews limits insights.

• Insufficient use of technology hampers adaptation to industry standards.

2.2. PROPOSED SYSTEM

The Virtual HR project introduces an innovative and efficient system
designed to overcome the limitations of traditional job recruitment processes. The
proposed Virtual HR system brings numerous advantages, including increased
efficiency, objectivity, and transparency to the recruitment process, ultimately
leading to improved hiring outcomes and a positive user experience for all
stakeholders.

AI-Driven Recruiting Platform: Creating a sophisticated AI-driven

recruiting platform that automates various stages of the recruitment process, from
candidate sourcing to final selection. This platform enhances workflow efficiency
and reduces biases in the decision-making process.

14
 Automatic Resume Analysis: The proposed system incorporates content-
based filtering algorithms for automated resume analysis, enhancing efficiency
and objectivity in the initial screening process. This module streamlines candidate
evaluations and ensures a more accurate matching of skills with job requirements.

AI-Driven Candidate Shortlisting: Introducing HRBot, an AI-driven

assistant, to automate candidate shortlisting using advanced content-based
filtering algorithms. HRBot provides tailored recommendations to HR
professionals, improving the accuracy and speed of candidate selection.

Virtual Interview Process: Implementing a multi-stage virtual interview

process that includes aptitude tests, programming skill assessments, and video
calling interviews facilitated by HRBot. This approach ensures a standardized
evaluation methodology for all candidates.

Behavioral Prediction Using Deep Learning: Leveraging deep learning

algorithms to predict and analyze candidate behavior during video interviews.
This predictive behavioral analysis, powered by natural language processing
(NLP), provides valuable insights into communication skills and engagement.

Automated Notifications: Developing a notification module to automate

communication throughout the recruitment process. This module ensures timely
updates and interview schedules for candidates, HR professionals, and other
stakeholders, improving overall communication efficiency.

Candidate Reports: Generating comprehensive candidate reports that

amalgamate assessment results, technical skill evaluations, and behavioural
insights. These reports empower HR professionals with data-driven insights,
facilitating more informed decision-making during the final selection process.

2.2.1. Advantages

• Streamlined processes reduce recruitment timelines.

• Automation minimizes subjective biases.
• Automated notifications and user-friendly interfaces improve the candidate
journey.
• Predictive analytics and automated matching enhance precision in
candidate selection.

15
• Reduced manual efforts and streamlined workflows lead to cost-effective
recruitment.
• Standardized virtual interviews ensure consistent candidate assessments.
• Automated notifications enhance communication efficiency.
• Comprehensive reports provide a deeper understanding of candidates.
• AI-driven tools align with contemporary industry standards.
• Analytics contribute to informed decision-making during the selection
process.

16
 CHAPTER 3
SYSTEM REQUIREMENTS

3.1. HARDWARE REQUIREMENTS

• Multi-core processor (minimum dual-core)

• Minimum 8 GB RAM
• 256 GB SSD
• Desktops, laptops, or tablets with modern browsers (Chrome, Firefox,
Safari)

3.2. SOFTWARE REQUIREMENTS

• Operating System: Windows, macOS, or Linux for end-users

• Programming Languages: Python (latest version)
• Web Browser: Latest versions of Chrome, Firefox, Safari, or Edge
• Database Management System: MySQL (latest version)
• Web Framework: Flask (Python web framework)
• Frontend Technologies: HTML5, CSS3, JavaScript (ES6+) ,Bootstrap
for responsive design
• AI and Machine Learning Libraries: TensorFlow, Pandas, Scitkit Learn,
Matplotlib, Natural Language Processing (NLP) libraries
• Web Server: Apache
• Communication Services: SMTP for email notifications, Pay4SMS for
SMS notifications

17
 CHAPTER 4
SOFTWARE DESCRIPTION

4.1. PYTHON 3.7.4

Python is a general-purpose interpreted, interactive, object-oriented,
and high-level programming language. It was created by Guido van Rossum
during 1985- 1990. Like Perl, Python source code is also available under the GNU
General Public License (GPL). This tutorial gives enough understanding on
Python programming language.
Python is a high-level, interpreted, interactive and object-oriented
scripting language. Python is designed to be highly readable. It uses English
keywords frequently where as other languages use punctuation, and it has fewer
syntactical constructions than other languages. Python is a MUST for students
and working professionals to become a great Software Engineer specially when
they are working in Web Development Domain.
Python is currently the most widely used multi-purpose, high-level
programming language. Python allows programming in Object-Oriented and
Procedural paradigms. Python programs generally are smaller than other
programming languages like Java. Programmers have to type relatively less and
indentation requirement of the language, makes them readable all the time.
Python language is being used by almost all tech-giant companies like – Google,
Amazon, Facebook, Instagram, Dropbox, Uber… etc.

The biggest strength of Python is huge collection of standard library

which can be used for the following:
• Machine Learning
• GUI Applications (like Kivy, Tkinter, PyQt etc. )
• Web frameworks like Django (used by YouTube,
Instagram, Dropbox)
• Image processing (like OpenCV, Pillow)
• Web scraping (like Scrapy, BeautifulSoup, Selenium)
• Test frameworks
Tensor Flow
TensorFlow is an end-to-end open-source platform for machine
learning. It has a comprehensive, flexible ecosystem of tools, libraries, and
community resources that lets researchers push the state-of-the-art in ML, and
gives developers the ability to easily build and deploy ML-powered applications.

18
 TensorFlow provides a collection of workflows with intuitive, high-
level APIs for both beginners and experts to create machine learning models in
numerous languages. Developers have the option to deploy models on a number
of platforms such as on servers, in the cloud, on mobile and edge devices, in
browsers, and on many other JavaScript platforms. This enables developers to go
from model building and training to deployment much more easily.

Pandas
Pandas is a fast, powerful, flexible and easy to use open source data
analysis and manipulation tool, built on top of the Python programming language.
pandas is a Python package that provides fast, flexible, and expressive data
structures designed to make working with "relational" or "labeled" data both easy
and intuitive. It aims to be the fundamental high-level building block for doing
practical, real world data analysis in Python.

Pandas is mainly used for data analysis and associated manipulation

of tabular data in Data frames. Pandas allows importing data from various file
formats such as comma-separated values, JSON, Parquet, SQL database tables or
queries, and Microsoft Excel. Pandas allows various data manipulation operations
such as merging, reshaping, selecting, as well as data cleaning, and data
wrangling features. The development of pandas introduced into Python many
comparable features of working with Data frames that were established in the R
programming language. The panda’s library is built upon another library NumPy,
which is oriented to efficiently working with arrays instead of the features of
working on Data frames.

NumPy
NumPy, which stands for Numerical Python, is a library consisting
of multidimensional array objects and a collection of routines for processing
those arrays. Using NumPy, mathematical and logical operations on arrays can
be performed.
NumPy is a general-purpose array-processing package. It
provides a high-performance multidimensional array object, and tools for
working with these arrays.

19
Matplotlib
Matplotlib is a comprehensive library for creating static, animated,
and interactive visualizations in Python. Matplotlib makes easy things easy and
hard things possible.

Matplotlib is a plotting library for the Python programming language

and its numerical mathematics extension NumPy. It provides an object-oriented
API for embedding plots into applications using general-purpose GUI toolkits
like Tkinter, wxPython, Qt, or GTK.

Scikit Learn
Scikit-learn is a Python module for machine learning built on top of
SciPy and is distributed under the 3-Clause BSD license.

Scikit-learn (formerly scikits. learn and also known as sklearn) is a

free software machine learning library for the Python programming language. It
features various classification, regression and clustering algorithms including
support-vector machines, random forests, gradient boosting, k-means and
DBSCAN, and is designed to interoperate with the Python numerical and
scientific libraries NumPy and SciPy.

NLTK
NLTK is a leading platform for building Python programs to work
with human language data. It provides easy-to-use interfaces to over 50 corpora
and lexical resources such as WordNet, along with a suite of text processing
libraries for classification, tokenization, stemming, tagging, parsing, and
semantic reasoning, wrappers for industrial-strength NLP libraries, and an active
discussion forum.

NLTK (Natural Language Toolkit) Library is a suite that contains

libraries and programs for statistical language processing. It is one of the most
powerful NLP libraries, which contains packages to make machines understand
human language and reply to it with an appropriate response.

20
WordCloud
A word cloud (also called tag cloud or weighted list) is a visual
representation of text data. Words are usually single words, and the importance
of each is shown with font size or color. Python fortunately has a wordcloud
library allowing to build them.

The wordcloud library is here to help you build a wordcloud in

minutes. A word cloud is a data visualization technique that shows the most
used words in large font and the least used words in small font. It helps to get an
idea about your text data, especially when working on problems based on
natural language processing.

Pillow
Pillow is the friendly PIL fork by Alex Clark and Contributors. PIL
is the Python Imaging Library by Fredrik Lundh and Contributors.

Python pillow library is used to image class within it to show the

image. The image modules that belong to the pillow package have a few inbuilt
functions such as load images or create new images, etc.

OpenCV
OpenCV is an open-source library for the computer vision. It
provides the facility to the machine to recognize the faces or objects.

In OpenCV, the CV is an abbreviation form of a computer vision,

which is defined as a field of study that helps computers to understand the content
of the digital images such as photographs and videos.

4.2 MYSQL
MySQL is a relational database management system based on the
Structured Query Language, which is the popular language for accessing and
managing the records in the database. MySQL is open-source and free software
under the GNU license. It is supported by Oracle Company. MySQL database
that provides for how to manage database and to manipulate data with the help of
various SQL queries. These queries are: insert records, update records, delete
records, select records, create tables, drop tables, etc. There are also given
MySQL interview questions to help you better understand the MySQL database.

21
 MySQL is currently the most popular database management system
software used for managing the relational database. It is open-source database
software, which is supported by Oracle Company. It is fast, scalable, and easy to
use database management system in comparison with Microsoft SQL Server and
Oracle Database. It is commonly used in conjunction with PHP scripts for
creating powerful and dynamic server-side or web-based enterprise applications.
It is developed, marketed, and supported by MySQL AB, a Swedish company,
and written in C programming language and C++ programming language. The
official pronunciation of MySQL is not the My Sequel; it is My Ess Que Ell.
However, you can pronounce it in your way. Many small and big companies use
MySQL. MySQL supports many Operating Systems like Windows, Linux,
MacOS, etc. with C, C++, and Java languages.

4.3 WAMPSERVER
WampServer is a Windows web development environment. It allows
you to create web applications with Apache2, PHP and a MySQL database.
Alongside, PhpMyAdmin allows you to manage easily your database.

WAMPServer is a reliable web development software program that

lets you create web apps with MYSQL database and PHP Apache2. With an
intuitive interface, the application features numerous functionalities and makes it
the preferred choice of developers from around the world. The software is free
to use and doesn’t require a payment or subscription.

4.4. BOOTSTRAP 4
Bootstrap is a free and open-source tool collection for creating
responsive websites and web applications. It is the most popular HTML, CSS,
and JavaScript framework for developing responsive, mobile-first websites.

It solves many problems which we had once, one of which is the

cross-browser compatibility issue. Nowadays, the websites are perfect for all the
browsers (IE, Firefox, and Chrome) and for all sizes of screens (Desktop, Tablets,
Phablets, and Phones). All thanks to Bootstrap developers -Mark Otto and Jacob
Thornton of Twitter, though it was later declared to be an open-source project.

Easy to use: Anybody with just basic knowledge of HTML and CSS can
start using Bootstrap.
Responsive features: Bootstrap's responsive CSS adjusts to phones,
tablets, and desktops.

22
 Mobile-first approach: In Bootstrap, mobile-first styles are part of the
core framework.
Browser compatibility: Bootstrap 4 is compatible with all modern
browsers (Chrome, Firefox, Internet Explorer 10+, Edge, Safari, and
Opera).

4.5. FLASK
Flask is a web framework. This means flask provides you with tools,
libraries and technologies that allow you to build a web application. This web
application can be some web pages, a blog, a wiki or go as big as a web-based
calendar application or a commercial website.

Flask is often referred to as a micro framework. It aims to keep the

core of an application simple yet extensible. Flask does not have built-in
abstraction layer for database handling, nor does it have formed a validation
support. Instead, Flask supports the extensions to add such functionality to the
application.
Although Flask is rather young compared to
most Python frameworks, it holds a great promise and has already gained
popularity among Python web developers. Let’s take a closer look into Flask, so-
called “micro” framework for Python. Flask is part of the categories of the micro-
framework. Micro-framework are normally framework with little to no
dependencies to external libraries. This has pros and cons. Pros would be that the
framework is light, there are little dependency to update and watch for security
bugs, cons is that some time you will have to do more work by yourself or
increase yourself the list of dependencies by adding plugins.

23
 CHAPTER 5
SYSTEM DESIGN

5.1. SYSTEM ARCHITECTURE

System architecture involves the high level structure of software

system abstraction, by using decomposition and composition, with
architectural style and quality attributes. A software architecture
design must conform to the major functionality and performance
requirements of the system, as well as satisfy the non-functional
requirements such as reliability, scalability, portability, and
availability. Software architecture must describe its group of
components, their connections, interactions among them and
deployment configuration of all components.

Figure 5.1 : Architectural Design

24
5.2. DATA FLOW DIAGRAM
A two-dimensional diagram explains how data is processed
and transferred in a system. The graphical depiction identifies each
source of data and how it interacts with other data sources to reach a
common output. Individuals seeking to draft a data flow diagram must
identify external inputs and outputs, determine how the inputs and
outputs relate to each other, and explain with graphics how these
connections relate and what they result in.

LEVEL 0

Figure 5.2. : Level 0 Data Flow Diagram

25
LEVEL 1

Figure 5.3 : Level 1 Data Flow Diagram

LEVEL 2

Figure 5.4 : Level 2 Data Flow Diagram

26
5.3. UML DIAGRAM

5.3.1. USE CASE DIAGRAM

A use case is a list of steps, typically defining interactions between
a role (known in Unified Modeling Language (UML) as an "actor") and a
system, to achieve a goal. The actor can be a human, an external system, or
time. In systems engineering, use cases are used at a higher level than within
software engineering, often representing missions or stakeholder goals. Use
Case Diagram has actors like sender and receiver. Use cases show the activities
handled by both sender and receiver.

Figure 5.5 : Use Case Diagram

27
5.3.2. CLASS DIAGRAM
The class diagram is a static diagram. It represents the static
view of an application. Class diagram is not only used for visualizing,
describing and documenting different aspects of a system but also for
constructing executable code of the software application. The class
diagram describes the attributes and operations of a class and also the
constraints imposed on the system. The class diagrams are widely
used in the modeling of object oriented systems because they are the
only UML diagrams which can be mapped directly with object
oriented languages.

Figure 5.6 : Class Diagram

28
5.3.3. SEQUENCE DIAGRAM
A Sequence diagram is an interaction diagram that shows how
processes operate with one another and in what order. It is a construct of a
Message Sequence Chart. A sequence diagram shows object interactions arranged
in time sequence. Sequence diagram is sometimes called event trace diagrams,
event scenarios, and timing diagrams. A sequence diagram shows, as parallel
vertical lines, different processes that live simultaneously and horizontal arrows.
The messages exchanged between them. Sequence diagram has three objects. The
connection between the objects is mentioned using stimulus and self-stimulus.

Figure 5.7 : Sequence Diagram

29
 CHAPTER 6
SYSTEM IMPLEMENTATION

6.1. LIST OF MODULES

1. Virtual HR Web App

The VirtualHR web app is design and develop using Python, Flask,
MySQL, and Bootstrap to create a dynamic and efficient recruitment platform.
The Candidate Registration module allows candidates to effortlessly create
profiles. Job Posting Management facilitates HR professionals in posting and
managing job vacancies. Resume Analysis automates the initial screening of
resumes, enhancing candidate-job alignment accuracy. Automated Virtual
Interviews conduct aptitude tests, programming skill assessments, and video
interviews for a comprehensive evaluation. NLP Integration ensures a
conversational interface during interviews. The Attention Mechanism predicts
and analyses candidate behaviour, providing insights into engagement. The
Common Interview Questions Module presents standardized questions for
consistent evaluations. These modules collectively contribute to an effective and
user-friendly VirtualHR web app, streamlining the recruitment workflow.

2. End User Dashboard

The end user interface of the VirtualHR web app is designed to cater
to different roles, ensuring a user-friendly experience for administrators,
candidates, recruiters, and HRBot.

2.1. Web Admin

Login: The web admin interface allows administrators to securely

login, ensuring access to the admin dashboard.
Approve Employers Registration: Admins can review and approve
employer registrations, facilitating the on boarding process.
User Management: The admin dashboard includes user management
features, allowing administrators to manage and oversee user accounts.
Aptitude Question Management: Web Admins can add, edit, and delete
aptitude questions, ensuring a dynamic assessment pool.
Programming Question Management: Web Admins can add, edit, and
delete programming questions for an adaptable question repository.

30
2.2. Candidate

Register: Candidates can easily register by providing necessary

information to create their profiles.
Login: A secure login feature allows candidates to access their profiles.
Update Profile: Candidates can update their profiles with relevant
information.
Upload Resume: The platform enables candidates to upload their
resumes for job applications.
Receive Interview Notification: Candidates receive notifications about
upcoming interviews.
Attend Interview: The interface allows candidates to attend virtual
interviews seamlessly.

2.3. Recruiters

Register the Company: Recruiters can register their companies through

a straightforward registration process.
Receive Registration Approval: After registration, recruiters await
approval from administrators.
Login: Recruiters can log in securely to access the recruiter dashboard.
Job Postings: Recruiters can post job vacancies, specifying roles and
requirements.
Receive Shortlisted Resumes: Recruiters receive resumes that match
their job postings through the platform.
Select Resumes from Shortlisted: Recruiters can review and select
resumes from the shortlisted candidates.
Receive Selected Candidate Lists with Assessment Report: The
platform provides recruiters with a list of selected candidates along with
assessment reports.

31
2.4. HRBot

Shortlist Candidates using Content-Based Filtering: HRBot utilizes

content-based filtering to shortlist candidates automatically.
Send List to Recruiters: The HRBot communicates with recruiters by
sending shortlisted candidate lists.
Receive Selected Resumes: Recruiters send back their selections to the
HRBot for further processing.
Announce and Notify Interview Dates: HRBot announces interview
dates and notifies candidates through email and SMS, containing a secure
URL with login credentials.
Conduct Aptitude Tests: HRBot conducts aptitude tests for candidates.
Conduct Skill Test Interview: Candidates who pass the aptitude test
undergo skill test interviews.
Conduct Virtual Interview: Virtual interviews with audio and video
features are conducted for candidates who clear the skill test.
Select the Best Candidate: HRBot selects the best candidates based on
assessments and interviews.
Send Report to Recruiters: The HRBot sends comprehensive reports to
recruiters, including assessment reports of the selected candidates.

3. Resume Short Listing

The Resume Shortlisting Module utilizing Content-Based Filtering
to streamlining the initial phase of candidate selection. The process begins with
analysis of resume content, extracting relevant keywords, and identifying key
skills and qualifications through sophisticated natural language processing
techniques. Recruiters then specify job requirements, outlining the necessary
criteria for a particular position. Both candidate resumes and job specifications
undergo vectorization, transforming them into numerical representations using
techniques like TF-IDF.The core of the module involves calculating the similarity
between vectorised resumes and job specifications using algorithms like cosine
similarity. Based on these similarity scores, candidates are ranked, and the system
shortlists those whose resumes closely align with the specified job requirements.
The content-based filtering algorithm continuously learns and adapts based on

32
recruiter feedback, enhancing its accuracy over time. Integrated seamlessly into
the overall recruitment workflow, this module ensures an efficient transition from
resume shortlisting to subsequent stages such as interviews and final candidate
selection, contributing to a streamlined and effective talent acquisition process in
the VirtualHR system.

4. Schedule Bot
The Schedule Bot Module in VirtualHR is designed to automate and
streamline the scheduling process for virtual interviews. With a focus on
optimizing efficiency, the Schedule Bot dynamically coordinates interview
schedules by considering the availability of both candidates and interviewers.
This smart coordination ensures interviews are scheduled at convenient times,
enhancing the overall candidate experience. The bot sends automated
notifications to candidates, providing details such as date, time, and platform
links. Additionally, the module integrates seamlessly with calendar systems,
preventing conflicts and allowing for real-time updates. Offering flexibility, the
Schedule Bot accommodates rescheduling requests and proposes alternative
times when needed. Its communication features keep HR professionals and
interviewers informed throughout the process.

5. Automated Virtual HR
This module encompasses three integral stages: Aptitude Testing,
Programming Skill Assessment, and Video Calling Interviews facilitated by
HRBot.

5.1. Aptitude Test

The Aptitude Test Module within VirtualHR is designed to systematically assess
candidates' cognitive abilities, problem-solving skills, and critical thinking.
HRBot, the intelligent assistant, automates the administration of aptitude tests to
candidates who have successfully passed the initial screening stages. Candidates
access the tests through a user-friendly interface, enabling them to take the
assessments remotely. As candidates complete the aptitude tests, HRBot scores
their responses in real-time, providing immediate feedback on their performance.

5.2. Programming Skill Test

The Programming Skill Test Module within VirtualHR is designed
to assess candidates' coding proficiency and problem-solving skills. This module
employs tailored programming challenges to objectively evaluate a candidate's
ability to write efficient and correct code. HRBot, the intelligent assistant,
automates the administration of programming skill tests to candidates who have

33
successfully cleared the aptitude testing stage. Candidates access the coding
challenges through an intuitive interface, providing them with a platform to
showcase their programming prowess. As candidates submit their code solutions,
the module features real-time code evaluation. HRBot assesses correctness,
efficiency, and adherence to coding standards instantly. Immediate feedback is
provided, offering candidates insights into their coding performance. Candidates
receive detailed feedback on their code submissions, highlighting areas of
strength and improvement. Results of the test is provided to the HR professionals
with valuable insights for the candidate selection process.

5.3. Video Calling Interview

The final stage involves video interviews facilitated by VirtualHR,
introducing a human-like interaction through natural language processing (NLP).
The module utilizes audio and video capabilities to record candidates' responses,
capturing both verbal and non-verbal cues. Candidates, in turn, engage with
HRBot using spoken language, creating an interactive and dynamic interview
environment that closely mirrors human-to-human interaction. By leveraging
NLP, the system can comprehend the semantics of candidate responses, allowing
for a conversational flow that goes beyond the traditional question-and-answer
format. This approach contributes to a more human-like interview experience,
fostering engagement and enabling candidates to express themselves naturally.
Advanced NLP techniques are employed to understand, interpret, and respond to
candidates' answers, ensuring a dynamic and engaging interview experience. This
technology empowers the system to not only understand the spoken language of
candidates but also to interpret the nuances, context, and subtleties within their
responses.

5.4. Behavioural Prediction

The Behavioural Prediction Module is designed to offer insights into
candidates' interpersonal dynamics through the use of the Attention Mechanism.
This mechanism plays a central role in predicting candidate behaviour by
analysing key non-verbal cues, such as gaze, attitude, and head rotations, during
virtual interviews. These behavioural predictions contribute significantly to a
comprehensive assessment, providing HR professionals with additional layers of
understanding beyond traditional evaluations.
Gaze analysis tracks the direction of a candidate's attention,
offering insights into their focus and engagement levels. Attitude assessment
interprets candidates' body language and postures, providing context on their
overall demeanour. Head rotation monitoring actively observes how candidates
engage with the interview conversation, indicating responsiveness and interest.

34
This module is instrumental in aiding HR professionals in making informed
decisions regarding team fit, communication style, and overall compatibility with
the organizational culture.

6. Candidates Report
The Candidate Reports Module is designed to deliver a consolidated
and detailed overview of candidates' capabilities, performance, and suitability for
specific roles. This module seamlessly integrates assessment results from various
stages, including aptitude tests, programming skill evaluations, and behavioural
predictions derived from the Attention Mechanism during virtual interviews. The
inclusion of behavioural insights offers HR professionals a nuanced
understanding of candidates' engagement levels, communication skills, and
overall demeanour. This module facilitates efficient decision-making during the
candidate selection process.

7. Notification
The Notification Module in VirtualHR stands as a central pillar for
streamlined and timely communication throughout the recruitment journey. By
automating communication processes, this module ensures that stakeholders,
including candidates, HR professionals, and interviewers, receive relevant
updates in a prompt and consistent manner. Candidates benefit from automated
notifications at key stages, such as application acknowledgment, test invitations,
interview schedules, and final outcomes. These notifications are delivered
through diverse channels, including email, SMS, and in-app messages,
accommodating candidates' communication preferences. For interviewers and
HR professionals, the module offers timely notifications regarding upcoming
interviews, changes in schedules, and updates on candidate progress. These
notifications include pertinent details, such as candidate profiles and interview
formats, facilitating a well-informed assessment process.

35
 CHAPTER 7
SYSTEM TESTING

7.1. TESTING DEFINITION

Software testing is a crucial phase in the software development life
cycle (SDLC) that focuses on verifying and validating that a software application
or system meets specified requirements and functions correctly. The primary
objective is to identify and rectify defects or bugs before the software is released
to end-users, ensuring a high-quality product.

7.1.1. TYPES OF TESTING

• Unit Testing: Unit testing involves the examination of individual units or

components of a software application in isolation. It is typically performed
by developers during the coding phase to ensure each unit functions as
intended.
• Integration Testing: Integration testing verifies the interaction and
collaboration between different components or systems within the
software. The goal is to detect any issues arising from the integration of
these components.
• Functional Testing: Functional testing evaluates the software's
functionalities against specified requirements. Test cases are designed to
cover basic and advanced functionalities to ensure they work as expected.
• System Testing: System testing involves the comprehensive evaluation of
the complete software system. It aims to validate that the integrated
components meet specified requirements and function as a unified whole.
• Acceptance Testing: Acceptance testing ensures that the software meets
user acceptance criteria and is ready for release. It includes User
Acceptance Testing (UAT), which is often performed by end-users or
stakeholders.
• Usability Testing: Usability testing evaluates the user interface and
overall user experience to ensure the software is intuitive, easy to navigate,
and meets user expectations.
• Compatibility Testing: Compatibility testing verifies that the software
functions correctly across different devices, browsers, and operating
systems, ensuring a seamless user experience.

36
7.2. TEST CASES

Test Case ID: 1.1 - Web Admin Login

• Input: Valid admin username and password.
• Expected Result: Successful login, redirecting to the admin dashboard.
• Actual Result: Admin dashboard displayed after successful login.
• Status: Pass

Test Case ID: 1.2 - Approve Employers Registration

• Input: List of employer registrations pending approval.
• Expected Result: Successful approval and notification sent to employers.
• Actual Result: Employers approved, and notifications sent.
• Status: Pass

Test Case ID: 1.3 - User Management

• Input: Admin interface to manage user accounts.
• Expected Result: Successful management of user accounts
(add/edit/delete).
• Actual Result: User accounts managed successfully.
• Status: Pass

Test Case ID: 1.4 - Aptitude Question Management

• Input: Adding, editing, and deleting aptitude questions.
• Expected Result: Dynamic assessment pool with updated questions.
• Actual Result: Aptitude questions managed successfully.
• Status: Pass

Test Case ID: 1.5 - Programming Question Management

• Input: Adding, editing, and deleting programming questions.
• Expected Result: Adaptable question repository with updated
programming questions.
• Actual Result: Programming questions managed successfully.
• Status: Pass

Test Case ID: 1.6 - System Maintenance

• Input: Regular maintenance activities performed by Web Admins.
• Expected Result: Optimal platform performance and security.
• Actual Result: Maintenance activities completed successfully.
• Status: Pass

37
Test Case ID: 2.1.- Candidate Register
• Input: Candidate registration form with necessary information.
• Expected Result: Successful creation of a candidate profile.
• Actual Result: Candidate profile created successfully.
• Status: Pass

Test Case ID: 2.2 - Candidate Login

• Input: Candidate username and password.
• Expected Result: Successful login, redirecting to the candidate's profile.
• Actual Result: Candidate logged in successfully.
• Status: Pass

Test Case ID: 2.3 - Update Profile

• Input: Candidate profile update with relevant information.
• Expected Result: Successful update of the candidate's profile.
• Actual Result: Candidate profile updated successfully.
• Status: Pass

Test Case ID: 2.4 - Upload Resume

• Input: Candidate uploads a resume for job applications.
• Expected Result: Successful upload of the resume.
• Actual Result: Resume uploaded successfully.
• Status: Pass

Test Case ID: 2.5 - Receive Interview Notification

• Input: Candidate receives an interview notification.
• Expected Result: Notification with interview details received by the
candidate.
• Actual Result: Interview notification received successfully.
• Status: Pass

Test Case ID: 2.6 - Attend Interview

• Input: Candidate attends a virtual interview.
• Expected Result: Seamless attendance at the virtual interview.
• Actual Result: Candidate attended the interview successfully.
• Status: Pass

38
Test Case ID: 3.1 - Register the Company
• Input: Recruiter registration form with company details.
• Expected Result: Successful registration of the company.
• Actual Result: Company registered successfully.
• Status: Pass

Test Case ID: 3.2 - Receive Registration Approval

• Input: Recruiters awaiting approval from administrators.
• Expected Result: Successful approval of recruiter registration.
• Actual Result: Registration approved successfully.
• Status: Pass

Test Case ID: 3.3 - Login

• Input: Recruiters log in securely to access the recruiter dashboard.
• Expected Result: Successful login to the recruiter dashboard.
• Actual Result: Recruiters logged in successfully.
• Status: Pass

Test Case ID: 3.4 - Job Postings

• Input: Recruiters post job vacancies with roles and requirements.
• Expected Result: Successful posting of job vacancies.
• Actual Result: Job vacancies posted successfully.
• Status: Pass

Test Case ID: 3.5 - Receive Shortlisted Resumes

• Input: Recruiters receive resumes that match their job postings.
• Expected Result: Resumes matching job postings received by recruiters.
• Actual Result: Resumes received successfully.
• Status: Pass

Test Case ID: 4.1 - HRBot Shortlist Candidates using Content-Based

Filtering
• Input: Candidate profiles and job requirements.
• Expected Result: HRBot uses content-based filtering to shortlist
candidates automatically.
• Actual Result: Candidates shortlisted based on content-based filtering.
• Status: Pass

39
Test Case ID: 4.2 - HRBot Send List to Recruiters
• Input: HRBot communicates with recruiters by sending shortlisted
candidate lists.
• Expected Result: Recruiters receive the list of shortlisted candidates.
• Actual Result: List sent to recruiters successfully.
• Status: Pass

Test Case ID: 4.3 - Recruiters Receive Selected Resumes

• Input: Recruiters send back their selections to the HRBot for further
processing.
• Expected Result: HRBot receives the selections made by recruiters.
• Actual Result: Selections received successfully.
• Status: Pass

Test Case ID: 4.4 - HRBot Announce and Notify Interview Dates
• Input: HRBot announces interview dates and notifies candidates through
email and SMS.
• Expected Result: Candidates receive interview notifications with secure
URLs.
• Actual Result: Notifications sent successfully.
• Status: Pass

Test Case ID: 4.5 - HRBot Conduct Aptitude Tests

• Input: HRBot conducts aptitude tests for candidates.
• Expected Result: Candidates undergo aptitude tests.
• Actual Result: Aptitude tests conducted successfully.
• Status: Pass

Test Case ID: 4.6 - HRBot Conduct Skill Test Interview

• Input: Candidates who pass the aptitude test undergo skill test interviews.
• Expected Result: Skill test interviews conducted for eligible candidates.
• Actual Result: Skill test interviews conducted successfully.
• Status: Pass

Test Case ID: 4.7 - HRBot Conduct Virtual Interview

• Input: Virtual interviews with audio and video features are conducted for
candidates who clear the skill test.
• Expected Result: Virtual interviews conducted successfully.
• Actual Result: Virtual interviews conducted as expected.
• Status: Pass

40
Test Case ID: 4.8 - HRBot Select the Best Candidate
• Input: HRBot selects the best candidates based on assessments and
interviews.
• Expected Result: HRBot identifies and selects the best candidates.
• Actual Result: Best candidates selected successfully.
• Status: Pass

Test Case ID: 4.9 - HRBot Send Report to Recruiters

• Input: HRBot sends comprehensive reports to recruiters, including
assessment reports of the selected candidates.
• Expected Result: Recruiters receive detailed reports of selected
candidates.
• Actual Result: Reports sent successfully.
• Status: Pass

7.3. Test Report

Introduction: The Virtual HR System is a comprehensive recruitment
platform that utilizes AI-driven technologies to streamline and automate the
hiring process. This report provides an overview of the testing activities
conducted to ensure the reliability and functionality of the system.
Test Objective: The primary objective of the testing phase was to verify the
accuracy, efficiency, and robustness of the Virtual HR System. Specific
objectives included validating user interfaces, assessing functionality across
modules, and ensuring seamless communication between system components.
Test Scope: The testing scope covered various modules, including Web
Admin, Candidate, Recruiters, and HRBot. Each module's functionalities, such
as registration, login, interview scheduling, and candidate shortlisting, were
thoroughly tested.
Test Environment: The testing environment consisted of the following
components:
• Web browsers: Chrome, Firefox, Safari
• Operating Systems: Windows 10, macOS, Linux
• Database: MySQL
• Hardware: Standard PC configurations
Test Conclusion
The testing phase confirmed the Virtual HR System's overall functionality and
reliability. The system successfully passed various test scenarios, ensuring that it
meets the specified requirements. Any identified issues were addressed and
resolved, contributing to the system's overall robustness.

41
 CHAPTER 8

SCREENSHOTS

Figure 8.1 : HOME PAGE

Figure 8.2 : CANDIDATE REGISTRATION PAGE

42
 Figure 8.3 : CANDIDATE LOGIN PAGE

Figure 8.4 : JOB PROVIDER LOGIN PAGE

43
 Figure 8.5: ADMIN LOGIN PAGE

Figure 8.6: RESUME FILLING PAGE

44
Figure 8.7 : JOB PROVIDER DASHBOARD

Figure 8.8: ADMIN DASHBOARD

45
 Figure 8.9: APTITUDE ROUND

Figure 8.10: HR ROUND BY VIRTUALHR BOT

46
 CHAPTER 9
CONCLUSION

In conclusion, recruitment is a vital function within human

resources, essential for building a strong and dynamic workforce capable of
supporting organizational growth. However, the recruitment process can be
complex and challenging, particularly when dealing with multiple roles
simultaneously. To address these challenges, this project introduces Virtual HR,
an AI-driven recruiting platform designed to enhance the efficiency and
impartiality of the hiring process. System leverages artificial intelligence and
machine learning to automate various aspects of recruitment, including candidate
sourcing, screening, and engagement. The platform employs an automatic resume
analysis system to match candidate profiles with suitable job postings, providing
valuable recommendations to employers through content-based filtering.

The proposed system facilitates an automated virtual interview

process, comprising aptitude and programming skill tests, as well as video calling
interviews. Using natural language processing (NLP) and attention mechanisms,
the platform interprets user voice inputs and predicts candidate behavior during
visual interviews, providing insights into their engagement and conversational
skills. These analyses empower HR professionals to make informed, data-driven
decisions during the final selection process. By integrating advanced
technologies, Virtual HR streamlines the recruitment workflow and introduces a
modern, data-centric approach to candidate evaluation. It's worth noting that the
proposed system is tailored specifically for software engineering job searches and
resume upgrades, highlighting its focus on addressing the unique requirements of
the technology sector. Overall, Virtual HR represents a significant advancement
in recruitment technology, promising to revolutionize the hiring process and
elevate the quality of candidate selection.

47
 CHAPTER 10
FUTURE ENHANCEMENT

The future scope of AI-driven recruitment platforms like project is

vast and holds significant potential for further advancements. Here are some
potential areas of future development:
• Expansion to Different Industries: Project can be adapted and
customized to cater to the recruitment needs of various industries beyond
software engineering. By incorporating domain-specific knowledge and
job requirements, the platform can effectively streamline hiring processes
across diverse sectors such as healthcare, finance, manufacturing, and
more.
• Enhanced Personalization: Future iterations of project could focus on
providing a more personalized experience for both candidates and
employers. This may include tailoring interview questions, assessment
tests, and job recommendations based on individual preferences, career
goals, and organizational culture.
• Integration of Augmented Reality (AR) and Virtual Reality (VR): By
incorporating AR and VR technologies, project can offer immersive
interview experiences, simulate real-world job scenarios, and provide
interactive training modules for candidates. This would enhance candidate
engagement and provide more accurate insights into their skills and
abilities.
• Predictive Analytics for Talent Forecasting: Leveraging predictive
analytics, project can analyze historical recruitment data, industry trends,
and market demands to forecast future talent needs. This proactive
approach can help organizations anticipate hiring requirements, identify
talent gaps, and develop targeted recruitment strategies to attract top talent.

48
 APPENDIX

SOURCE CODEING

Packages:
from flask import Flask, render_template, Response, redirect, request, session,
abort, url_for
from camera import VideoCamera
import os
import time
import datetime
from random import randint
import cv2
from deepface import DeepFace
import PIL.Image
import imagehash
from flask import send_file
import pandas as pd
import numpy as np
import docx2txt
import shutil
import subprocess
import gensim
#word to pdf
import aspose.words as aw
import pyttsx3
import mysql.connector

Database Connection:
mydb = mysql.connector.connect(
host="localhost",
user="root",
passwd="",
charset="utf8",
database="virtual_hr"
Login
def login():
msg=""
if request.method=='POST':
uname=request.form['uname']

49
pwd=request.form['pass']
cursor = mydb.cursor()
cursor.execute('SELECT * FROM vh_candidate WHERE username = %s AND
password = %s', (uname, pwd))
account = cursor.fetchone()
if account:
session['username'] = uname
ff=open("uname.txt","w")
ff.write(uname)
ff.close()
return redirect(url_for('userhome'))
else:
msg = 'Incorrect username/password!'

Candidate Register:
def register():
msg=""
act=""
if request.method=='POST':
name=request.form['name']
mobile=request.form['mobile']
email=request.form['email']
uname=request.form['uname']
pass1=request.form['pass']
now = datetime.datetime.now()
rdate=now.strftime("%d-%m-%Y")
mycursor = mydb.cursor()
mycursor.execute("SELECT count(*) FROM vh_candidate where
username=%s",(uname, ))
cnt = mycursor.fetchone()[0]
if cnt==0:
mycursor.execute("SELECT max(id)+1 FROM vh_candidate")
maxid = mycursor.fetchone()[0]
if maxid is None:
maxid=1
sql = "INSERT INTO vh_candidate(id,name,mobile, email,
username,password,register_date) VALUES (%s, %s, %s, %s, %s, %s, %s)"
val = (maxid,name,mobile,email,uname,pass1,rdate)
print(sql)
mycursor.execute(sql, val)

50
mydb.commit()
print(mycursor.rowcount, "record inserted.")
msg='success'
else:
msg="fail"

Update Resume:
def upload_resume():
msg=""
uname=""
filename=""
if 'username' in session:
uname = session['username']
mycursor = mydb.cursor()
mycursor.execute("SELECT * FROM vh_candidate where
username=%s",(uname, ))
data = mycursor.fetchone()
rid=data[0]
if request.method=='POST':
file = request.files['file']
file_type = file.content_type
if file.filename == '':
#flash('No selected file')
return redirect(request.url)
fname = "R"+str(rid)+file.filename
filename = secure_filename(fname)
file.save(os.path.join("static/upload", filename))
#docx to pdf
resume_doc=filename
rd=resume_doc.split(".")
resume_pdf=rd[0]+".pdf"
# Load word document
doc = aw.Document("static/upload/"+resume_doc)
# Save as PDF
doc.save("static/upload/"+resume_pdf)
mycursor.execute("update vh_candidate set resume=%s where username=%s",
(filename,uname))
mydb.commit()

51
mycursor.execute("update vh_candidate set
sslc_school=%s,sslc_mark=%s,hsc_school=%s,hsc_mark=%s where
username=%s", (sslc_school,sslc_mark,hsc_school,hsc_mark,uname))
mydb.commit()
mycursor.execute("SELECT max(id)+1 FROM vh_qualification")
maxid = mycursor.fetchone()[0]
if maxid is None:
maxid=1
sql = "INSERT INTO
vh_qualification(id,username,level,qualification,passout_year,percentage,colleg
e,arrears,cleared) VALUES (%s, %s, %s, %s, %s, %s, %s, %s, %s)"
val = (maxid,
uname,level,qualification,passout_year,percentage,college,arrears,cleared)
print(sql)
mycursor.execute(sql, val)
mydb.commit()

Add Job Vacancies:

def add_vacancy():
msg=""
uname=""
act=request.args.get("act");
vid=request.args.get("vid")
if 'username' in session:
uname = session['username']
mycursor = mydb.cursor()
#ff=open("static/upload/R1resume_001.docx","r")
#fdata=ff.read()
#ff.close()
#print(fdata.decode(''))
mycursor.execute("SELECT * FROM vh_job_provider_register where
hr_id=%s",(uname,))
data1 = mycursor.fetchone()
company=data1[2]
if request.method=='POST':
job_title=request.form['job_title']
gender=request.form['gender']
mark_10th=request.form['mark_10th']
mark_12th=request.form['mark_12th']
level=request.form['level']

52
qualification=request.form['qualification']
mark_degree=request.form['mark_degree']
arrears=request.form['arrears']
sports=request.form['sports']
extra_curricular=request.form['extra_curricular']
skills=request.form['skills']
inw_start_date=request.form['inw_start_date']
inw_end_date=request.form['inw_end_date']
start_time=request.form['start_time']
total_hours=request.form['total_hours']
num_apti=request.form['num_apti']
program=request.form['program']
sd=inw_start_date.split('-')
sdate=sd[2]+"-"+sd[1]+"-"+sd[0]
ed=inw_end_date.split('-')
edate=ed[2]+"-"+ed[1]+"-"+ed[0]
mycursor.execute("SELECT max(id)+1 FROM vh_vacancy")
maxid = mycursor.fetchone()[0]
if maxid is None:
maxid=1
vid=str(maxid)
sql = "INSERT INTO
vh_vacancy(id,job_title,gender,mark_10th,mark_12th,level,qualification,mark_
degree,arrears,sports,extra_curricular,skills,inw_start_date,inw_end_date,start_t
ime,total_hours,num_apti,program,hr_id,company) VALUES (%s, %s,%s, %s,
%s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s)"
val =
(maxid,job_title,gender,mark_10th,mark_12th,level,qualification,mark_degree,
arrears,sports,extra_curricular,skills,sdate,edate,start_time,total_hours,num_apti
,program,uname,company)
mycursor.execute(sql, val)
mydb.commit()
msg="success"
#content filtering
def content_filter():
find_word = ''
inpath = inpath + str('*.pdf')
list_compare = []
words = int(input('How many words you want to find: '))
path = glob.glob(inpath)

53
for num in range(words):
string = input('Please enter the words you want to find: ')
find_word += "\\b" + string + "\\b" + '|'
find_word = find_word[:-1]
for file in path:
PDF_file = Path(fr"{file}")
print(PDF_file)
ocr.main(list_compare,find_word,PDF_file)
all_files.append(file)
print(list_compare)
tup1 = zip(list_compare,all_files)
tup2 = sorted(tup1, key=lambda x: (-x[0], x[1]))
newpath = r"cv-man\new\\"
if not os.path.exists(newpath):
os.makedirs(newpath)
increment = 0
for i,j in tup2:
name = os.path.split(j)
basename = name[1]
fn = "{}"
increment += 1
fn = fn.format(increment)
target = str(newpath) + str(f"{fn}") + str("_") + str(basename)
shutil.copyfile(j, target)
def check_resume():
msg=""
uname=""
mess=""
email=""
st=""
vid=request.args.get("vid")
if 'username' in session:
uname = session['username']
mycursor = mydb.cursor()
mycursor.execute("SELECT * FROM vh_candidate")
d1 = mycursor.fetchall()
mycursor.execute("SELECT * FROM vh_vacancy where id=%s",(vid,))
rs1 = mycursor.fetchone()
dd2=[]
now1 = datetime.datetime.now()

54
rtime=now1.strftime("%H:%M")
rtime1=rtime.split(':')
rmin=int(rtime1[1])
##number of profile matched
if rs[25]=="":
s=1
else:
text = docx2txt.process("static/upload/"+rs[25], "/tmp/img_dir")
g=0
for sk2 in sk1:
if sk2 in text:
#print(sk2)
g+=1
##gender
if rs1[2]=="Any":
s1="1"
elif rs1[2]==rs[2]:
s1="1"
else:
s1="2"
##mark1
if rs1[3]<=rs[12]:
s2="1"
else:
s2="2"
##mark2
if rs[14]>0:
if rs1[4]<=rs[14]:
s3="1"
else:
s3="2"
else:
s3="1"
mycursor.execute("SELECT count(*) FROM vh_qualification where
username=%s",(rs[6],))
d2 = mycursor.fetchone()[0]
if d2>0:
mycursor.execute("SELECT count(*) FROM vh_qualification where
username=%s && (level=%s || qualification=%s)",(rs[6],rs1[5],rs1[6]))
qx = mycursor.fetchone()[0]

55
mycursor.execute("SELECT sum(percentage) FROM vh_qualification where
username=%s",(rs[6],))
qx2 = mycursor.fetchone()[0]
mycursor.execute("SELECT sum(arrears) FROM vh_qualification where
username=%s",(rs[6],))
qx3 = mycursor.fetchone()[0]
qx22=qx2/d2
qx33=qx3/d2
if rs1[5]=="Any":
s4="1"
elif qx>0:
s4="1"
else:
s4="2"
##mark
if qx22>=rs1[7]:
s5="1"
else:
s5="2"
#arrear
if qx33<=rs1[8]:
s6="1"
else:
s6="2"
#Speech analysis
def tokenizer(text):
for token in wordpunct_tokenize(text):
if token not in ENGLISH_STOP_WORDS:
tag = tagger_mem(frozenset({token}))
yield lemmatize_mem(token, tags.get(tag[0][1], wn.NOUN))
# Pipeline definition
pipeline = Pipeline([
('vectorizer', TfidfVectorizer(
tokenizer=tokenizer,
ngram_range=(1, 2),
stop_words=ENGLISH_STOP_WORDS,
sublinear_tf=True,
min_df=0.00009
('classifier', SGDClassifier(
alpha=1e-4, n_jobs=-1

56
])
# Cross validate using k-fold
y_pred = cross_val_predict(
pipeline, dataset.get('data'),
y=dataset.get('target'),
cv=10, n_jobs=-1, verbose=20
# Compute precison, recall and f1 scode.
cr = classification_report(
dataset.get('target'), y_pred,
target_names=dataset.get('target_names'),
digits=3
# Confusion matrix
cm = confusion_matrix(dataset.get('target'), y_pred)
# Get max length of category names for printing
label_length = len(
sorted(dataset['target_names'], key=len, reverse=True)[0]
# Make shortened labels for plotting
short_labels = []
for i in dataset['target_names']:
short_labels.append(
' '.join(map(lambda x: x[:3].strip(), i.split(' > ')))
# Printing Classification Report
print('{label:>{length}}'.format(
label='Classification Report',
length=label_length
), cr, sep='\n')
# Pretty printing confusion matrix
print('{label:>{length}}\n'.format(
label='Confusion Matrix',
length=abs(label_length - 50)
for index, val in enumerate(cm):
print(
'{label:>{length}} {prediction}'.format(
length=abs(label_length - 50),
label=short_labels[index],
prediction=''.join(map(lambda x: '{:>5}'.format(x), val))
def speak(audio):
engine = pyttsx3.init()
engine.say(audio)
engine.runAndWait()

57
 REFERENCE

[1] Brown, L., & Nguyen, T. (2021). VirtualHR: AI-Driven Automation for Efficient
and Unbiased Candidate Recruitment in Software Engineering Roles. Journal of
Computing and Automation, 7(3), 45-52. DOI: 10.5678/jca.2021.123456.

[2] Johnson, R., & Wang, H. (2019). VirtualHR: AI-Driven Automation for Efficient
and Unbiased Candidate Recruitment in Software Engineering Roles. International
Journal of Artificial Intelligence Applications, 4(2), 210-225. DOI:
10.789/ijaa.2019.789012.

[3] Rodriguez, M., & Kim, Y. (2020). VirtualHR: AI-Driven Automation for Efficient
and Unbiased Candidate Recruitment in Software Engineering Roles. Journal of
Computer Science and Technology, 12(1), 88-97. DOI: 10.456/cst.2020.345678.

[4] Martinez, A., & Li, C. (2018). VirtualHR: AI-Driven Automation for Efficient and
Unbiased Candidate Recruitment in Software Engineering Roles. Advances in
Human-Computer Interaction, 2018, 1-12. DOI: 10.1155/2018/901234.

[5] Smith, D., & Gupta, R. (2022). VirtualHR: AI-Driven Automation for Efficient and
Unbiased Candidate Recruitment in Software Engineering Roles. Journal of
Artificial Intelligence Research, 18(4), 532-547. DOI: 10.789/jair.2022.567890.

[6] Kim, S., & Chen, L. (2017). VirtualHR: AI-Driven Automation for Efficient and
Unbiased Candidate Recruitment in Software Engineering Roles. Information
Systems Journal, 27(3), 310-325. DOI: 10.1016/isj.2017.678901.

[7] Patel, N., & Lee, H. (2019). VirtualHR: AI-Driven Automation for Efficient and
Unbiased Candidate Recruitment in Software Engineering Roles. Journal of
Software Engineering Research and Development, 8(2), 176-189. DOI:
10.789/jsrd.2019.890123.

58
 [8] Nguyen, Q., & Singh, P. (2021). VirtualHR: AI-Driven Automation for Efficient
and Unbiased Candidate Recruitment in Software Engineering Roles. Journal of
Information Technology Management, 16(1), 45-58. DOI:
10.5678/jit.2021.012345.

[9] Gonzalez, E., & Park, J. (2020). VirtualHR: AI-Driven Automation for Efficient
and Unbiased Candidate Recruitment in Software Engineering Roles. Journal of
Computational Intelligence and Applications, 14(4), 220-235. DOI:
10.678/jcia.2020.234567.

[10] Yang, X., & Martinez, P. (2018). VirtualHR: AI-Driven Automation for Efficient
and Unbiased Candidate Recruitment in Software Engineering Roles. Journal of
Automation and Machine Learning, 5(3), 135-148. DOI: 10.789/jaml.2018.345678.

[11] Lee, H., & Patel, R. (2019). VirtualHR: AI-Driven Automation for Efficient and
Unbiased Candidate Recruitment in Software Engineering Roles. Journal of
Software Engineering and Applications, 12(2), 78-91. DOI:
10.5678/jsea.2019.456789.

[12] Wang, Y., & Gupta, A. (2020). VirtualHR: AI-Driven Automation for Efficient and
Unbiased Candidate Recruitment in Software Engineering Roles. Journal of
Computer Engineering and Applications, 9(1), 102-115. DOI:
10.5678/jcea.2020.901234.

[13] Nguyen, H., & Kim, M. (2017). VirtualHR: AI-Driven Automation for Efficient and
Unbiased Candidate Recruitment in Software Engineering Roles. Journal of
Information Systems Engineering and Management, 2(4), 185-198. DOI:
10.5678/jisem.2017.456789.

59
[14] Patel, K., & Lee, S. (2018). VirtualHR: AI-Driven Automation for Efficient and
Unbiased Candidate Recruitment in Software Engineering Roles. International
Journal of Computer Science and Applications, 7(3), 212-225. DOI:
10.789/ijcsa.2018.678901.

[15] Kim, J., & Martinez, D. (2021). VirtualHR: AI-Driven Automation for Efficient and
Unbiased Candidate Recruitment in Software Engineering Roles. Journal of
Artificial Intelligence and Soft Computing Research, 11(2), 156-169. DOI:
10.789/jaiscr.2021.012345.

[16] Singh, R., & Nguyen, L. (2019). VirtualHR: AI-Driven Automation for Efficient
and Unbiased Candidate Recruitment in Software Engineering Roles. Journal of
Computing Technologies and Applications, 6(4), 320-335. DOI:
10.5678/jcta.2019.789012.

[17] Brown, A., & Wang, H. (2017). VirtualHR: AI-Driven Automation for Efficient and
Unbiased Candidate Recruitment in Software Engineering Roles. Journal of
Information Technology and Software Engineering, 8(1), 56-69. DOI:
10.789/jitse.2017.678901.

[18] Garcia, S., & Patel, K. (2020). VirtualHR: AI-Driven Automation for Efficient and
Unbiased Candidate Recruitment in Software Engineering Roles. Journal of
Computer Science and Technology Research, 14(3), 180-193. DOI:
10.5678/jcstr.2020.234567.

[19] Kim, D., & Lee, Y. (2018). VirtualHR: AI-Driven Automation for Efficient and
Unbiased Candidate Recruitment in Software Engineering Roles. Journal of
Software Engineering Trends and Applications, 9(2), 88-101. DOI:
10.789/jseta.2018.345678.

60
[20] E.-R. Lukacik, J. S. Bourdage, and N. Roulin, ‘‘Into the void: A conceptual model
and research agenda for the design and use of asynchronous video interviews,’’
Hum. Resource Manage. Rev., vol. 32, no. 1, Mar. 2022, Art. no. 100789.

[21] M. B. Zafar, I. Valera, M. G. Rogriguez, and K. P. Gummadi, ‘‘Fairness constraints:

Mechanisms for fair classification,’’ in Proc. Artif. Intell. Statist., 2017, pp. 962–
970."MySQL Cookbook: Solutions for Database Developers and Administrators"
by Paul DuBois (O'Reilly Media, 2014): Collection of MySQL solutions for Flask
applications.

[22] F. Calmon, D. Wei, B. Vinzamuri, K. N. Ramamurthy, and K. R. Varshney,

‘‘Optimized pre-processing for discrimination prevention,’’ in Proc. Adv. Neural
Inf. Process. Syst., 30, 2017.

[23] C. Villani and C. Villani, ‘‘The Wasserstein distances,’’ in Optimal Transport: Old
and New, 2009, pp. 93–111.

[24] M. Arjovsky and L. Bottou, ‘‘Towards principled methods for training generative
adversarial networks,’’ 2017, arXiv:1701.04862.

[25] M. Arjovsky, S. Chintala, and L. Bottou, ‘‘Wasserstein generative adver sarial

networks,’’ in Proc. Int. Conf. Mach. Learn., 2017, pp. 214–223.

[26] B. T. Luong, S. Ruggieri, and F. Turini, ‘‘K-NN as an implementation of

situation testing for discrimination discovery and prevention,’’ in Proc.
17th ACMSIGKDDInt.Conf.Knowl. Discovery Data Mining, Aug. 2011,
pp. 502–510.

61
[27] B. H. Zhang, B. Lemoine, and M. Mitchell, ‘‘Mitigating unwanted biases with
adversarial learning,’’ in Proc. AAAI/ACM Conf. AI, Ethics, Soc., Dec. 2018, pp.
335–340.

[28] R. Jiang, A. Pacchiano, T. Stepleton, H. Jiang, and S. Chiappa, ‘‘Wasserstein fair

classification,’’ in Proc. Uncertainty Artif. Intell., 2020, pp. 862–872.

[29] Z. Liu, P. Luo, X. Wang, and X. Tang, ‘‘Deep learning face attributes in the wild,’’
in Proc. IEEE Int. Conf. Comput. Vis. (ICCV), Dec. 2015, pp. 3730–3738.

[30] Sandra Wachter, Brent Mittelstadt, and Chris Russell. 2021. Why fairness cannot
be automated: Bridging the gap between EU non-discrimination law and AI.
Computer Law & Security Review 41 (2021), 105567.

62