Scribd
Upload
18 views19 pages

Forecasting-Future-College-Admissions-Using-Machine-Learning

This project utilizes machine learning, specifically the K-Nearest Neighbors algorithm, to predict college admissions and improve academic planning by analyzing historical student data. The proposed system aims to provide clarity for students regarding their admission chances, optimize resource allocation for colleges, and assist policymakers in creating equitable admission processes. Future enhancements include refining the model with advanced algorithms and incorporating additional student and college-specific parameters.

Uploaded by

konkasireesha11
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
18 views19 pages

Forecasting-Future-College-Admissions-Using-Machine-Learning

This project utilizes machine learning, specifically the K-Nearest Neighbors algorithm, to predict college admissions and improve academic planning by analyzing historical student data. The proposed system aims to provide clarity for students regarding their admission chances, optimize resource allocation for colleges, and assist policymakers in creating equitable admission processes. Future enhancements include refining the model with advanced algorithms and incorporating additional student and college-specific parameters.

Uploaded by

konkasireesha11
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 19

COLLEGE ADMISSION PREDICTION

USING MACHINE LEARNING


ALGORITHM
This presentation outlines a project utilizing machine learning to
predict college admissions, improving the college admission
process and academic planning using KNN.

BATCH NO: C3 NAME OF THE GUIDE: Mr. D Varun Prasad

TEAM MEMBERS: Kandula Sriya (21H71A05I5)

Ramichetty Chandini (21H71A05D8)

Bala Renuka (21H71A05G6)

Sakhamuri Soma Sekhar (21H71A05H7)


Table of Contents
1. Project Overview
2. Problem Statement
3. Existing System
4. Proposed System
5. Literature Review
6. Methodology
7. Model Evaluation and Validation
8. Benefits and Impact
9. Output
10.Reference
11.Conclusion
Problem Statement
Predicting College Admissions Forecasting Enrollment Trends

Current admission processes are often complex and Colleges struggle to accurately predict enrollment
opaque. Students lack clarity about their chances of numbers for different branches, leading to resource
admission to specific colleges and branches. misallocation and inefficient planning.
Project Overview
1. Data Collection
1
Gathering historical data on student demographics, academic performance, and admission outcomes.

2. Data Preprocessing
2
Cleaning, transforming, and preparing the data for analysis and model training.

3. Model Selection
3
Choosing appropriate machine learning algorithms for classification and forecasting.

4. Model Training
4 Training the selected models on the preprocessed data to identify patterns and
relationships.

5. Model Evaluation
5 Assessing the performance and accuracy of the trained models
using appropriate metrics.
Existing
System
The existing system uses a dataset for predicting graduate admissions with the following
parameters:
•GRE Scores: Out of 340
•TOEFL Scores: Out of 120
•University Rating: Out of 5
•Statement of Purpose (SOP) and Letter of Recommendation (LOR) Strength: Out of 5
•Undergraduate GPA: Out of 10
•Research Experience: Either 0 (No) or 1 (Yes)
•Chance of Admit: Ranging from 0 to 1
Purpose: To help students shortlist universities by predicting their chances of admission based on
their profiles.

The existing system is a machine learning pipeline for regression tasks, using multiple models like
K-Nearest Neighbors, Decision Trees, Random Forest, AdaBoost, Extra Trees, XGBoost, and
CatBoost. It includes a robust data handling mechanism, evaluation metrics (R² score and Mean
Squared Error), and visualization tools (Plotly, Seaborn, Matplotlib).
Proposed System

The proposed system uses machine learning techniques to predict college admissions based on
historical data. It employs the K-Nearest Neighbors (KNN) algorithm to estimate the most
probable institute, degree, and program a student can get based on their entrance exam rank,
category, reservation quota, and seat pool.
The key features of the proposed system include:
• Data preprocessing and encoding: Converts categorical features (e.g., category,
quota) into numerical values using Label Encoding.
• Machine learning model (KNN): Trains the model to make predictions based on input
parameters.
• User-friendly interface: Allows students to input their details and receive predictions
instantly.
• Data visualization: Provides insights into admission trends, quota-based seat distribution,
and program-wise selection patterns.
Literature Review
Predictive Analytics College Admission
Forecasting
Several studies have explored
the use of machine learning for Researchers have developed
predicting student success and models to predict student
retention rates. The use of admissions to specific colleges
predictive models to analyze based on various factors,
student demographics, including academic
academic performance, and performance, extracurricular
social factors has been activities, and personal
investigated. attributes.

Enrollment Trend Forecasting


The use of time-series models to forecast enrollment trends for
different branches in higher education institutions has gained
prominence.
Methodology
Data Preprocessing and Feature Selection Model Selection and Training

Data cleaning and transformation to ensure data Selection of appropriate machine learning algorithms
quality and accuracy. Feature selection to identify the for classification and forecasting. Training models on
most relevant factors impacting college admissions. the preprocessed data to identify patterns and
relationships.
Model Evaluation and
Validation
Evaluation Metrics
Evaluating the performance of the trained models
using accuracy, precision, recall, and other relevant
metrics.

Cross-Validation
Using cross-validation techniques to ensure the
robustness and generalizability of the models.
Benefits and Impact

Students Colleges Policymakers


Gain clarity in choosing colleges and Optimize resource allocation, Design equitable and transparent
branches, making informed manage enrollment effectively, and admission processes, promoting
decisions about their future. provide better academic planning. access to higher education.
Output

The interface of our project

We are using the streamlit library and running


the code
With the help of the show data, we can
view our dataset

Which Category gets more seats in


colleges
Total colleges visualizations:

Student per year graph:


Institute course provided: 5-Year Courses offered by the institute

4-Year Courses offered by the institute


Prediction
output
Conclusion
This project delivers a comprehensive and robust solution for accurately predicting college admissions and branch-
wise enrollment trends, enabling a more data-driven and strategic approach to higher education management. The
developed machine learning models provide valuable insights for students, allowing for more informed decision-
making regarding college choices and branch selection. Simultaneously, colleges benefit from optimized resource
allocation and improved enrollment management. Policymakers can leverage this predictive capability to create
more equitable and transparent admissions processes, ultimately expanding access to higher education
opportunities. The successful implementation of this project sets a precedent for applying advanced analytics to
address crucial challenges within the higher education sector.
Future
Enhancement
The path for the future of this college prediction system has a vast potential for
further improvement and practical application. It can be made a more precise and
simpler tool by adding more features and refining the current model. Future
improvements include using improved machine learning algorithms such as Random
Forests, Decision Trees, Support Vector Machines (SVM), or Neural Networks to make
improved predictions. Adjustment of the hyperparameters through methods such as
Grid Search or Random Search and deep learning methods (ANNs, CNNs, RNNs) also
raises the model's predictability. Also incorporating more student-related parameters,
i.e., entrance test scores, extracurricular activities, past performance, and
geographical and demographic information makes the overall prediction stronger.
Incorporating college-specific parameters such as the quality of staff, placement
achievements, and fee structure further narrows the predictions.
References
We are taking reference from the forecasting sales algorithm and the link is provided below

https://www.kaggle.com/code/badl071/forecasting-future-sales-using-machine-learning

A Research paper which predicts college based on the colleges and the link as follows

https://www.researchgate.net/publication/375243523_COLLEGE_PREDICTION_SYSTEM

Reference to different Machine learning is taken from the below research paper

https://www.researchgate.net/publication/344717762_Machine_Learning_Algorithms_-A_Review

You might also like