Skip to content

Latest commit

 

History

History
33 lines (25 loc) · 2.03 KB

README.md

File metadata and controls

33 lines (25 loc) · 2.03 KB
Raw

Admission-Prediction

Create a Machine Learning model that predict the chances of a student getting admission

Admission-process Image Credit: Donetsk National Medical University

Business Understanding

The main goal of this project is to collect and analyze student data that contain featutres that determines the likelihood of getting admission for Masters in the university. We will help students check the possibility of getting admitted with their different exam scores.

Analytical Approach

This is a supervised regression problem where we will determine the chances of getting admission. 0 means no chance, 1 shows a very high chance and 0.5 is the cutoff. We will use the different regression algorithms to solve this problem.

Data Collection

The data was gotten from Kaggle https://www.kaggle.com/mohansacharya/graduate-admissions.

Data Understanding

The data contains the following features:

  • TOFEL and GRE scores
  • University Ratings
  • CGPA of applicant
  • Scores out of 5 for the Statement of Purpose (SOP) and Letter of Recommendation (LOR)
  • Whether the graduate course is Researched based or not.

Modeling

  • I scaled all the values of the features to keep them be in the same range for easy model building
  • Linear Regression, Lasso Regression, Linear SVR, Thiel San Regressor , Least Angle Regression, Ridge Regression, Elastic Net, Bayesian Regressor and Random Forest Regression models were all built.
  • Mean Absolute Error (MAE) and Root mean squared error (RMSE) were the metrics used to evaluate the performance of the model.
  • Linear Regression has the lowest MAE and RMSE, so it is the chosen model.

Web Application

I built a web application using Streamlit and deployed it to Streamlit Cloud.

The web app : Admsission Prediction App