AcademicPerformanceML

Student Exam Outcome Prediction

This project predicts whether a student will pass or fail their final examination based on their performance in the first two examinations and several demographic, social, and behavioral factors. The goal is to build a machine learning model that can classify students into "Pass" or "Fail" categories with high accuracy.

Dataset Overview

The dataset contains information about students, including their exam scores and personal attributes. Below are the features used for prediction:

Key Features

Academic Performance:

First exam score Second exam score

Demographic Information:

School Gender (Sex) Age Address (Urban/Rural)

Family Information:

Family size Mother's job Father's job

Social & Behavioral Factors:

Free time Health status Travel time to school Study time Internet access

Target Variable:

Final exam result (Pass/Fail)

Objective

This project aims to leverage machine learning algorithms to accurately classify students' final exam outcomes based on the above features.

Methodology

The project explored multiple classification models to identify the best-performing one. The following models were tested: Logistic Regression Random Forest Classifier Gradient Boosting

Best Performing Model: Logistic Regression

Logistic Regression emerged as the best model for this task, achieving an accuracy of 93% on the test data. This high performance is attributed to the linear relationship between the input features and the target variable.

Results

• Accuracy: 93%
• Precision, Recall, and F1-Score: Evaluated for both "Pass" and "Fail" classes, showing balanced performance across both outcomes.
• ROC-AUC Score: High, indicating a strong ability of the model to distinguish between the two classes.

How to Use

1. Clone the Repository:

  git clone 
https://github.com/yourusername/StudentExamOutcomePrediction.git 
cd StudentExamOutcomePrediction

2. Install Dependencies:

pip install -r requirements.txt

3. Run the Model: Execute the main script to train the logistic regression model and test its performance.

4. Input New Data: Provide new student data in the required format to predict whether they will pass or fail.

Conclusion

This project demonstrates the effectiveness of logistic regression in predicting student outcomes based on academic performance and other influencing factors. The model can be further improved by incorporating additional features or fine-tuning hyperparameters. For questions or contributions, feel free to open an issue or pull request.