54

Student Grade Prediction

A simple linear regression in multiple variable model , to get the prediction of the grade of the student

GitHub

Student Grade Prediction using Multiple Linear Regression

This project implements a multiple linear regression model to predict a student's grade based on several input features, including Socioeconomic Score, Study Hours, Sleep Hours, and Attendance (%). The model is trained using a dataset that contains these input features along with the corresponding grades.

Problem Statement

The goal of this project is to create a predictive model that estimates the grades of students based on the following features:

  • Socioeconomic Score: A numerical value representing the socioeconomic background of the student.
  • Study Hours: The total number of hours a student has studied.
  • Sleep Hours: The number of hours the student sleeps on average.
  • Attendance (%): The percentage of classes attended by the student.

The model uses multiple linear regression to learn the relationship between these input features and the target variable (Grades) and can be used to predict the grade of a student for unseen data.

Features

  • Socioeconomic Score: A numeric value that reflects the socioeconomic background of the student, which may influence their academic performance.
  • Study Hours: The number of hours the student spends studying, which is expected to have a direct correlation with grades.
  • Sleep Hours: The amount of sleep the student gets, which can affect cognitive function and academic performance.
  • Attendance (%): The percentage of classes the student attends, which is a factor in determining their overall understanding of the course material.

Approach

1. Data Preprocessing

  • Normalization: The features (Study Hours, Sleep Hours, Attendance, Socioeconomic Score) are normalized to have zero mean and unit variance to ensure that the gradient descent converges efficiently.

2. Model: Multiple Linear Regression

The multiple linear regression model is used to find the optimal coefficients (theta) for the relationship:

Grades = θ₀ + θ₁ × Socioeconomic Score + θ₂ × Study Hours + θ₃ × Sleep Hours + θ₄ × Attendance

  • Training: The model is trained using gradient descent, minimizing the cost function (Mean Squared Error).

3. Prediction

After training the model, the learned coefficients are used to predict a student's grade based on their feature values.

Evaluation

  • The performance of the model is evaluated by plotting the cost function against the number of iterations and monitoring the convergence of the gradient descent.
  • The final model is used to predict the grade of a student based on their input features.

Requirements

  • numpy: For numerical computations.
  • matplotlib: For visualizations.
  • pandas: (Optional) If you plan to use data from CSV files and perform more advanced data manipulation.

Steps to Run the Project

  1. Clone the repository: 
    git clone https://github.com/adityaamehra/student-grade-prediction.git
  2. Install dependencies: 
    pip install numpy matplotlib
  3. Run the Python script to train the model and make predictions: 
    python grade_prediction.py

Example Output

The model will output the final computed values for theta (model parameters) and plot the cost function's convergence over time. After training, you can input a student's features and get a predicted grade.

Future Improvements

  • Feature Engineering: Add more features, such as student participation in extracurricular activities or parental involvement, to improve predictions.
  • Model Tuning: Experiment with different learning rates for gradient descent and other optimization techniques.
  • Cross-validation: Implement k-fold cross-validation to better estimate model performance and avoid overfitting.