A REPORT ON

SUMMER INTERNSHIP

Name of the Student: Gayathri Karri

Name of the College: VIGNAN’s Institute of Information Technology

Registration Number: 21L31A0598

Period of Internship: From: 09-05-2023 to 05-07-2023

Year: 2nd year

Name and Address of the Intern Organization: Yhills Edutech Pvt.Ltd.

 An Internship Report on
ARTIFICIAL INTELLIGENCE SUMMER INTERNSHIP

Submitted in accordance with the requirement for the degree of

Bachelor of Technology

By

Gayathri Karri
(Roll No. 21L31A0598)

Under the Faculty Guideship of

Mr.V.Nagu

DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING

VIGNAN’S INSTITUTE OF INFORMATION TECHNOLOGY (A)

June 2023
 Student’s Declaration
I, K. Gayathri a student of B. Tech program,Reg.No.21L31A0598 of the Department
of Computer Science and Engineering(CSE) do hereby declare that I have
completed the mandatory internship from 09-05-2023 to 05-07-2023 in Yhills
Edutech under the Faculty Guideship of Ms.Premlatha, Department of Computer
Science and Engineering (CSE) ,(Vignan’s Institute of Information Technology).

(signature with date)

 OFFICIAL CERTIFICATION
This is to certify that K.Gayathri Reg.No.21L31A0598 has completed her internship in Yhills on
“ARTIFICIAL INTELLIGENCE” under my supervision as a part of partial fulfillment the requirement
for the Degree of Bachelor of Technology in the department of Computer Science and Engineering
(Vignan’s Institute Of Information Technology).

This is accepted for evaluation

(Signatory with Date and Seal)

Endorsements

Faculty Guide

Head of the Department

Principal
 CERTIFICATE FROM INTERN ORGANIZATION
This is to certify that K .Gayathri Reg.No 21L31A0598 of Vignan’s Institute of Information
Technology underwent internship in Yhills Edutech from 09-05-2023 to 05-07-2023. The overall
performance of the intern during her internship is found to be satisfactory.

Authorized Signatory with Date and Seal

 Acknowledgements
I would like to express my deepest gratitude and appreciation to Yhills for providing me with the
opportunity to intern at their esteemed organization. This internship experience has been invaluable in
shaping my professional growth and development.

First and foremost, I would like to thank the entire Yhills team for their guidance, support, and
encouragement throughout my internship journey. Their expertise, knowledge, and willingness to share
their experiences have been instrumental in enhancing my skills and understanding of the industry.

I am also grateful to my fellow interns at Yhills for their camaraderie and collaboration. Their enthusiasm
and shared experiences made my internship period enjoyable and fostered a positive learning
environment.

Furthermore, I would like to acknowledge the entire Yhills organization for creating a supportive and
inclusive work culture. The collaborative atmosphere, open communication, and emphasis on personal
growth have made my internship experience truly fulfilling.

Last but not least, I would like to express my heartfelt appreciation to the management team at Yhills for
providing me with this internship opportunity. I am grateful for their trust in my abilities and for entrusting
me with challenging responsibilities.

In conclusion, I am immensely grateful to Yhills for providing me with a remarkable internship experience.
The skills, knowledge, and connections I have gained during my time here will undoubtedly shape my
future endeavors. Thank you for the incredible opportunity and for being an integral part of my
professional journey.
My heartful thanks to my internship instructor Ms.Premlatha who helped me a lot to gain some
knowledge on Artificial Intelligence.
 Contents

SI.No Content Page

No
1 Personal details 1
2 Title 2
3 Declaration 3
4 Official Certification 4
5 Certificate from intern 5
organization
6 Acknowledgement 6
7 Contents 7-8
8 Chapter-1:Executive Summary 9
9 Chapter-2:overview of 10-11
organization
10 Chapter-3:internship part 12
11 Activity Log for Week-1 13
12 Weekly Report for Week-1 14
13 Activity Log for Week-2 15
14 Weekly Report for Week-2 16
15 Activity Log for Week-3 17
16 Weekly Report for Week-3 17-18
17 Activity Log for Week-4 18
18 Weekly Report for Week-4 19
19 Activity Log for Week-5 19
20 Weekly Report for Week-5 20-21
21 Activity Log for Week-6 and 21-22
weekly report
22 Outcomes Description 23
23 Daily Activity Photos 24-27
24 Internship Completion 28
Certification
CHAPTER 1: EXECUTIVE SUMMARY
Internship Report - Yhills Edutech

This internship report provides an overview of the internship experience at Yhills Edutech. The report
highlights the learning objectives and outcomes achieved during the internship period. It includes a
description of the business sector and the organization where the internship took place. Furthermore,
it summarizes all the activities undertaken by the intern throughout the internship duration.

Learning Objectives and Outcomes:

1. Gain a comprehensive understanding of the educational technology sector and its trends.
Outcome: Developed a strong knowledge base of the latest advancements and innovations in the
edutech industry.

2. Learn to apply instructional design principles in developing educational materials.

Outcome: Successfully created engaging and effective learning resources using instructional design
techniques.

3. Enhance skills in content creation and curriculum development.

Outcome: Produced high-quality educational content and contributed to the development of
curriculum materials.

4. Improve project management skills by participating in various projects.

Outcome: Gained experience in coordinating project tasks, meeting deadlines, and collaborating
with team members effectively.

5. Acquire practical experience in conducting market research and analyzing data.

Outcome: Conducted market surveys, gathered data, and contributed to market research reports,
enabling informed business decisions.
 CHAPTER 2: OVERVIEW OF THE ORGANIZATION

A. Introduction of the Organization:

Yhills Edutech is a leading company operating in the educational technology sector. Founded in [year], the
organization is dedicated to revolutionizing education by leveraging technology to create innovative learning
solutions. Yhills Edutech aims to make education more accessible, engaging, and personalized for learners
of all ages. The company offers a wide range of products and services, including online learning platforms,
interactive educational content, data-driven learning analytics tools, and curriculum development resources.

B. Vision, Mission, and Values of the Organization:

Vision: Yhills Edutech envisions a world where education transcends boundaries and empowers individuals
to reach their full potential through innovative technology-driven solutions.

Mission: The mission of Yhills Edutech is to transform education by developing and providing cutting-edge
technology solutions that enhance the learning experience and promote lifelong learning.

Values: Yhills Edutech upholds the following core values:

1. Innovation: Constantly seeking new and creative approaches to improve education.
2. Accessibility: Ensuring that education is accessible to all, irrespective of their backgrounds or
geographical location.
3. Collaboration: Fostering collaboration and partnerships to deliver high-quality educational solutions.
4. Excellence: Striving for excellence in all aspects of the organization's operations and products.
5. Learner-Centricity: Placing learners at the heart of product development and ensuring their needs are met.

C. Policy of the Organization, in relation to the intern role:

Yhills Edutech has a comprehensive policy to guide the intern's role within the organization. This policy
emphasizes providing a valuable learning experience, mentorship, and professional development
opportunities. The organization aims to create a supportive and inclusive work environment where interns
can contribute to meaningful projects, gain practical skills, and receive guidance from experienced
professionals. The policy also emphasizes open communication, respect, and adherence to ethical standards.

D. Organizational Structure:

Yhills Edutech follows a hierarchical organizational structure that supports effective decision-making and
streamlined operations. The structure consists of various departments, including:

1. Executive Leadership: Comprising the CEO, CTO, CFO, and other key executives responsible for setting
the strategic direction of the organization.
2. Department Heads: Leading different functional areas such as Content Development, Technology,
Marketing, Sales, and Operations.
3. Teams: Each department is divided into teams, led by managers or team leaders, focusing on specific
functions or projects.

E. Roles and Responsibilities of the Employees in which the intern is placed:

As an Artificial Intelligence (AI) Training Intern at Yhills Edutech, your role and responsibilities would
involve working closely with the AI team and contributing to the development and implementation of AI-
based educational solutions. Here are some key roles and responsibilities associated with the position:
1. Research and Analysis:
- Conduct research on AI technologies, algorithms, and methodologies applicable to the education sector.
- Analyze existing AI models and algorithms to identify potential applications for improving learning
outcomes.

2. Data Collection and Annotation:

- Support the collection and annotation of educational data for training and evaluating AI models.
- Collaborate with content developers and subject matter experts to curate relevant datasets.
- Ensure the quality and integrity of data through meticulous annotation and data cleaning processes.

3. Model Training and Evaluation:

- Train AI models using the collected and annotated datasets.
- Implement appropriate evaluation metrics and methodologies to assess the performance of AI models.
- Collaborate with the team to fine-tune models and optimize their accuracy and efficiency.

4. Collaboration and Learning:

- Collaborate with cross-functional teams, including developers, content creators, and instructional
designers.
- Stay updated with the latest advancements in AI technologies and educational trends.
- Actively participate in team meetings, brainstorming sessions, and knowledge-sharing activities.

F. Performance of the Organization:

Yhills Edutech has demonstrated impressive performance in terms of turnover, profits, market reach, and
market value. The organization has experienced consistent growth in revenue, driven by the increasing
demand for its innovative educational technology solutions. Yhills Edutech has expanded its market reach
globally, establishing partnerships with educational institutions, businesses, and government organizations.
The company's

commitment to innovation and customer satisfaction has contributed to its positive reputation and market
value.

G. Future Plans of the Organization:

Yhills Edutech has ambitious future plans aimed at further expanding its impact in the educational
technology sector. Some of the key future plans of the organization include:

1. Continual Innovation: Investing in research and development to introduce new products and services that
enhance learning experiences.
2. Market Expansion: Strengthening market presence in existing regions and exploring opportunities for
expansion into new markets.
3. Partnerships and Collaborations: Forging strategic partnerships with educational institutions, content
creators, and technology providers to enhance product offerings.
4. User Personalization: Developing personalized learning solutions by leveraging artificial intelligence and
data analytics.
5. Professional Development: Investing in the growth and development of employees through training
programs and career advancement opportunities.
CHAPTER 3: INTERNSHIP PART

I have worked as an Artificial Intelligence intern at Yhills Edutech. It was a great experience. During
this internship I have learnt many new things.
ACTIVITIES IN THE INTERN ORGANIZATION:
As the internship was an online internship I have performed the following tasks according to the instructions
given by my instructor. It was a 2 month internship where I need to attend the online classes weekly twice up to one
month. Each class is of two hours duration where the instructor teaches how to deal with different types of datasets.
The instructor will be explaining different types of attributes and function and built in libraries in python I used to
work with them parallelly along with the instructor. I have learnt many new things like how to import the libraries,
loading the dataset, exploratory data analysis(EDA) which is cleaning of the dataset and creating the models. In the
second month of Internship I have submitted the two projects allotted by my instructor.
 ACTIVITY LOG FOR WEEK-1

Brief description of the Learning Person In-Charge

Day Date
daily activity Outcome Signature
09/05/2023 Python (Internals, do’s and Installation of
6:00 PM - don’ts) Architecture, Data Anaconda
8:00 PM Structure Prompt
Jupyter
Notebook-An
Overview
Shorcut Lkeys
Day-1 in Jupyter
Notebook
Data Types in
Python Rules
for Naming
the Variables
11/05/2023 Python List, Tuple,
6:00 PM - Set,
8:00 PM Dictionary

Day-2
 WEEKLY REPORT
WEEK – 1 (From Date: 09-05-2023 to Date:14-04-2023)

Objective of the Activity Done:

• Learnt how to install Anaconda prompt on my laptop
• Learnt Basics of python
• Learnt some of the concepts like list, tuple, set, dictionaries.
Detailed Report:
09/05/2023:
Introduction to the internship and how to create an environment for executing the programs.
Basic rules of python were explained like declaring variables, taking input, printing the output.
11/05/2023:
Creating lists, tuples, sets and dictionaries and performing operations like inserting values, deleting the
values etc. using them.
Accessing and printing random integers using numpy library.
Rules in python were discussed.
 ACTIVITY LOG FOR WEEK-2

Day Date Brief description of the Learning Person In-Charge

daily activity Outcome Signature
16/05/2023 Data Analysis , Machine
6:00 PM - Manipulation with numpy Learning
8:00 PM and pandas Python data Libraries
science package to Numpy-Hands
manipulate, calculate and on Pandas-
Day-3
analyze data Hands on

18/05/2023 Exploratory Data Data

6:00 PM - Visualization in Python Visualization
8:00 PM with matplotlib Learn how Matplotlib-
to explore, visualize, and Hands on
extract insights from data Seaborn-
Day-4
Hands on
 WEEKLY REPORT
WEEK – 2 (From Date: 15/05/2023 to Date: 21/05/2023)

Objective of the Activity Done:

• Learnt about numpy library.
• Learnt about pandas library.
• Learnt about matplot library.
• Learnt about seaborn library.
Detailed Report:
16/05/2023:
Learnt about numpy library and pandas library in detail. Learnt many functions in numpy
library like arrange, exp, sqrt, max, min, log, argmin, square, std, var, mean, round. Worked a lot with all
the functions discussed and performed many operations using them.
Sqrt and square functions:
Used for finding sqrt and square of all the elements present in the array and prints the output.
Max and min functions:
Used to find maximum and minimum element in the array.
Argmax and argmin functions:
Used to find minimum and maximum index from the array.
Std ,var and mean functions:
Used to find the mean, variance and standard deviation of all the elements present in the array.
PANDAS LIBRARY:
In pandas library I learnt how to create a data frame , how to access the data frame, finding size and shape
of the data frame etc.
I learnt many functions like shape, head, sample etc.
18/05/2023:
Matplotlib and Seaborn:
Through these libraries I have learnt how to create different types of plots like countplot, violinplot,
boxplot, heat map, swarm plot, distplot, joinplot.
When we want to know about the single categorical data we use count plot. If we want to work with single
numerical data then we use dist plot. When we are working with two categorical data items then we use
count plot and when we wanted to work with two numerical data items we use strip, voilin, swarm plots.
 ACTIVITY LOG FOR WEEK-3

Day Date Brief description of the Learning Person In-Charge

daily activity Outcome Signature
23/05/2023 Statistical Thinking in Measures of
6:00 PM - Python (Part 1) Build the Central
8:00 PM foundation you need to Tendency
think statistically and to Measures of
speak the language of your Dispersion
Day-5
data IQR Statistics-
Hands-On

25/05/2023 Supervised Learning and Supervised

6:00 PM - UnSupervised Learning Learning
Day-6 8:00 PM Classification, Regression, Unsupervised
Fine-tuning your model Learning
Linear
Regression
Metrics in
Linear
Regression
Hands-on in
Linear
Regression

WEEKLY REPORT
WEEK – 3 (From Date: 22-05-2023to Date: 28/05/2023)

Objective of the Activity Done:

• Learnt about IQR statistics
• Learnt about Linear Regression
• Learnt about Supervised learning and Unsupervised learning
Detailed Report:
23/05/2023:
The interquartile range (IQR) is a statistical measure used to describe the spread or dispersion of a dataset.
It is based on quartiles, which divide the dataset into four equal parts. To calculate the IQR, first we need
to arrange the data in ascending order.
Here is a step-by-step description of how to calculate the IQR for a dataset with 100 lines:
1. Sort the data: Arrange the 100 lines of data in ascending order from the smallest to the largest value.
2. Calculate the lower quartile (Q1): Find the median of the lower half of the data. Since we have 100 lines,
the lower half is made up of the first 50 lines. Q1 is the value at the midpoint of this lower half.
3. Calculate the upper quartile (Q3): Find the median of the upper half of the data. Again, since we have
100 lines, the upper half is made up of the last 50 lines. Q3 is the value at the midpoint of this upper half.
4. Calculate the IQR: Subtract Q1 from Q3 to obtain the IQR. Mathematically, IQR = Q3 - Q1.
The IQR provides a measure of the spread of the middle 50% of the dataset. It is resistant to outliers, which
means it is not significantly affected by extreme values in the data. The larger the IQR, the greater the
dispersion or variability within the dataset.
Additionally, the IQR is often used in conjunction with box plots to visualize the distribution of data. In a
box plot, the IQR is represented by a box that spans from Q1 to Q3, with a line (whisker) extending from
each end to indicate the range of the data within 1.5 times the IQR.
25/05/2023:
Supervised Learning:
Supervised learning is a machine learning approach where the algorithm learns from labeled training data.
In supervised learning, the training data consists of input variables (features) and corresponding output
variables (labels or target values). The goal is for the algorithm to learn a mapping function that can predict
the correct output given new, unseen input data. The algorithm learns from the training data by adjusting its
internal parameters or model based on the provided input-output pairs. Common supervised learning
algorithms include linear regression, decision trees, support vector machines (SVM), and neural networks.
Unsupervised Learning:
Unsupervised learning is a machine learning approach where the algorithm learns from unlabeled data. In
unsupervised learning, the training data only consists of input variables (features) without any
corresponding output labels. The goal of unsupervised learning is to discover patterns, structures, or
relationships in the data without prior knowledge of the expected outcomes. The algorithm explores the
data and identifies inherent patterns or clusters based on similarities or differences in the input features.
Unsupervised learning techniques include clustering algorithms such as k-means, hierarchical clustering,
and dimensionality reduction methods like principal component analysis (PCA) or t-distributed stochastic
neighbor embedding (t-SNE).

ACTIVITY LOG FOR WEEK-4

Day Date Brief description of the Learning Person In-Charge

daily activity Outcome Signature
30/05/2023 Logistic regression Logistic
6:00 PM - Regression
8:00 PM Metrics in
Logistic
Regression
Day-7
Hands-on in
Logistic
Regression

WEEKLY REPORT
WEEK – 4 (From Date: 29/05/2023to Date: 04/06/2023)

Objective of the Activity Done:

 • Learnt about Logistic regression in detail
• Learnt metrics of Logistic Regression
Detailed Report:
30/05/2023:
Logistic regression is a statistical model used for binary classification problems, where the outcome variable
is categorical and has two classes. It is a type of supervised learning algorithm that predicts the probability
of an instance belonging to a particular class.

Key points about logistic regression:

1. Target Variable: Logistic regression is used when the target variable is binary or categorical, with two
possible outcomes (e.g., yes/no, true/false, 0/1).
2. Probability Estimation: Instead of directly predicting the class labels, logistic regression estimates the
probability of an instance belonging to a specific class using a logistic function (also known as the sigmoid
function). This function maps any real-valued number to a probability between 0 and 1.
3. Model Interpretation: Logistic regression provides interpretable results. It calculates the coefficients
(weights) associated with each feature, indicating the impact of each feature on the predicted probability.
These coefficients can be interpreted as the change in the log-odds of the target class per unit change in the
corresponding feature.
4. Assumptions: Logistic regression assumes a linear relationship between the independent variables
(features) and the log-odds of the target class. It also assumes that there is no multicollinearity among the
features, no influential outliers, and the residuals follow a logistic distribution.
5. Training the Model: The logistic regression model is trained using maximum likelihood estimation, which
involves finding the coefficients that maximize the likelihood of the observed data. This is typically done
using optimization algorithms such as gradient descent.
6. Decision Boundary: Logistic regression uses a decision boundary to separate the two classes. The
boundary is determined by a threshold probability (usually 0.5). Instances with predicted probabilities above
the threshold are classified as one class, while those below the threshold are classified as the other class.
7. Evaluation: Model performance in logistic regression is often assessed using evaluation metrics such as
accuracy, precision, recall, and F1 score. Additionally, techniques like ROC curves and AUC (Area Under
the ROC Curve) can be used to evaluate the model's discrimination power.

Logistic regression is a widely used and well-established algorithm for binary classification tasks. It is
especially useful when interpretability of the model and understanding the impact of features on the outcome
is important.
ACTIVITY LOG FOR WEEK-5

Day Date Brief description of the Learning Person In-Charge

daily activity Outcome Signature
06/06/2023 SVM, Linear Regression Support
6:00 PM - Vector
8:00 PM Machine
Hands on in
SVM
Day-8

08/06/2023 Preprocessing for Machine Exploratory

6:00 PM - Learning in Python Data Analysis
Day-9 8:00 PM Introduction to Data Missing
Preprocessing, Values
Standardizing Data Outliers
Standardizatio
 n
normalization
Feature
Scaling and
Selection

WEEKLY REPORT
WEEK – 5 (From Date: 05-06-2023 to Date: 11-06-2023)

Objective of the Activity Done:

• Learnt about SVM(Support Vector Machine)
• Learnt about Exploratory Data Analysis(EDA)
Detailed Report:
06/06/2023:
Support Vector Machine:
Support Vector Machine (SVM) is a powerful supervised learning algorithm used for classification and
regression tasks. It works by finding the optimal hyperplane that best separates different classes in the
feature space. SVM aims to maximize the margin, the distance between the hyperplane and the nearest data
points of each class, making it robust to outliers. By using kernel functions, SVM can handle nonlinear
decision boundaries effectively. SVM offers strong generalization capabilities, even with high-dimensional
data. It is versatile and widely used in various applications such as image recognition, text categorization,
and bioinformatics. However, SVM's computational complexity can be a challenge with large datasets.

08/06/2023:
Exploratory Data Analysis:
EDA is basically about cleaning the dataset by removing the missing values duplicate outliers from the
dataset. It is also used for converting the categorical data into numerical data.
Exploratory Data Analysis (EDA) is a crucial initial step in data analysis, where the primary goal is to
understand the data and gain insights from it. EDA involves examining and summarizing the main
characteristics of the dataset, uncovering patterns, identifying outliers, and exploring relationships between
variables.

Key aspects of exploratory data analysis include:

1. Data Cleaning: This involves handling missing values, dealing with outliers, and ensuring data integrity
by removing or imputing erroneous or inconsistent values.

2. Descriptive Statistics: Computing summary statistics such as mean, median, standard deviation, and
quartiles provides an overview of the data's central tendency, dispersion, and shape.

3. Data Visualization: Creating visual representations of the data through histograms, box plots, scatter plots,
and other visualizations helps identify patterns, trends, and relationships among variables.
4. Feature Engineering: EDA can help in identifying potential features or transformations that may enhance
the predictive power of machine learning models.

5. Univariate and Bivariate Analysis: Examining individual variables (univariate analysis) and exploring
relationships between pairs of variables (bivariate analysis) helps understand the distributions, correlations,
and dependencies in the data.

6. Hypothesis Generation: EDA often involves generating hypotheses about the data, which can be tested
further using statistical methods or machine learning algorithms.

EDA plays a crucial role in guiding subsequent analysis steps, such as selecting appropriate models,
identifying relevant variables, and detecting data issues that may affect the quality of the analysis. By
exploring and understanding the data, EDA helps uncover patterns, validate assumptions, and gain insights
that contribute to informed decision-making and problem-solving.

ACTIVITY LOG FOR WEEK-6

Day Date Brief description of the Learning Person In-Charge

daily activity Outcome Signature
13/06/2023 Tree Based Models Decision Tree
6:00 PM - Classification and Bagging
8:00 PM Regression Trees Boosting
Random
Forest
Day-10

15/06/2023 Machine Learning Project Modelling

6:00 PM - "Use data science Linear
Day-11 8:00 PM packages, analysis, Regression-
visualization, create Python
model, extract pure data Logistic
etc Regression-
Python
Decision Tree,
Bagging,
Boosting,
Random
Forest -
Python

WEEKLY REPORT
WEEK – 6 (From Date:12-06-2023 to Date:18-06-2023 )

Objective of the Activity Done:

• Learnt about many ensemble models like decision tree classifier, Random Forest Classifier, Ada
boost Classifier.
Detailed Report:

Random Forest is a popular and powerful supervised learning algorithm used for classification and
regression tasks. It is an ensemble method that combines multiple decision trees to make predictions.
Here are the key characteristics of Random Forest:
1. Ensemble of Decision Trees: Random Forest builds an ensemble of decision trees by training each tree on
a random subset of the data and a random subset of the features. Each tree independently predicts the target
variable, and the final prediction is obtained through voting (for classification) or averaging (for regression)
of the individual tree predictions.
2. Bagging: Random Forest employs a technique called bagging (bootstrap aggregating), which creates
multiple subsets of the training data through sampling with replacement. This helps in creating diverse trees
and reducing overfitting.
3. Feature Randomness: At each split in a decision tree, Random Forest considers only a subset of features
chosen randomly. This randomness encourages tree diversity and reduces the risk of selecting only the most
important features.
4. Robustness to Outliers: Random Forest is robust to outliers and noisy data because it aggregates
predictions from multiple trees, reducing the impact of individual outliers.
5. Variable Importance: Random Forest provides a measure of feature importance by assessing the reduction
in prediction accuracy when a specific feature is randomly permuted. This can help identify the most
influential features in the dataset.
6. Handling High-Dimensional Data: Random Forest performs well even with high-dimensional data, as it
can effectively handle a large number of features without feature selection.
7. Model Interpretability: Although individual trees in a Random Forest can be complex, the overall model
provides insights into feature importance and can be visualized to understand the decision-making process.
Random Forest has wide applicability and has been successfully used in various domains, including finance,
healthcare, and image recognition. It is known for its robustness, accuracy, and ability to handle large and
complex datasets.

The decision tree algorithm is a supervised learning method that uses a tree-like structure to make decisions
or predictions. It is a popular and interpretable machine learning algorithm that can be used for both
classification and regression tasks.

Here are the key features of the decision tree algorithm:

1. Tree Structure: The algorithm constructs a tree-like structure where each internal node represents a
decision based on a feature, and each leaf node represents a class label (in classification) or a predicted value
(in regression).
2. Feature Selection: The decision tree algorithm selects the best features to split the data based on certain
criteria. One common criterion is the "information gain" in classification, which measures the reduction in
entropy or impurity after the split. In regression, the "mean squared error" or similar metrics are used.
3. Recursive Partitioning: The decision tree algorithm employs a recursive partitioning process, where the
dataset is split into subsets based on the selected feature at each node. This process is repeated recursively
for each subset until a stopping condition is met (e.g., reaching a certain depth or having a minimum number
of samples).
4. Interpretability: Decision trees are highly interpretable, as the resulting tree structure can be easily
visualized and understood. The decision path from the root to a leaf node represents a set of rules that can be
followed to make predictions or classify new instances.
5. Handling Missing Values: Decision trees can handle missing values by using surrogate splits or assigning
missing values to the most frequent class or average value.
6. Overfitting: Decision trees are prone to overfitting, especially when the tree becomes too deep and
complex. Techniques like pruning, setting a maximum depth, or using regularization methods can help
mitigate overfitting.
7. Ensemble Methods: Decision trees are often used as building blocks in ensemble methods like Random
Forest and Gradient Boosting, where multiple trees are combined to improve predictive performance.
 CHAPTER 4: OUTCOMES DESCRIPTION

• Through this internship I have learned many new skills and acquired knowledge on unknown projects.
• By this internship I was able to create projects using the datasets and was able to perform operations and fit a
model to them.
• I was able to calculate the accuracy occurred for each model and able to find which model is best suit for the given
data set
• I was able to perform operations using both regression and classification
• I also learn time management by completing all the concepts explained by the instructor and able to understand
them
 Daily Activity GPS Photos

Day-1, Date:09/05/2023 Day-2, Date:11/05/2023

Day-3, Date:16/05/2023 Day-4, Date:18/05/2023

Day-5, Date:23/05/2023 Day-6, Date:25/05/2023

Day-7, Date:30/05/2023 Day-8, Date:06/06/2023

Day-9, Date:08/06/2023 Day-10, Date:13/06/2023

Day-11, Date:15/06/2023