Student Performance Prediction

Group Name

Student Performance Prediction

Project Title

Student Final Grade Prediction Using Machine Learning Models

Group Members

• Shreyeska Silwal
• Sagun Bahadur Pradhan
• Srijan Ayer

Project Description

Project Idea and Approach

The goal of this project is to develop a machine learning model to predict students' final grades based on a variety of factors, including study habits, attendance, extracurricular activities, gender, and parental support. We aim to explore and compare several machine learning algorithms to determine which provides the most accurate prediction.

The process involved:

1. Data Preprocessing: This step involved handling categorical data and preparing the features for machine learning models. We used LabelEncoder to convert categorical variables like gender and parental support into binary values. The dataset was split into training and test sets with a 70-30 ratio for model evaluation.

2. Model Selection: We experimented with three models:

   • Random Forest Classifier: A robust ensemble method known for handling large datasets with higher accuracy.
   • Decision Tree Classifier: A simple, interpretable model that helps to understand the influence of each feature on the final prediction.
   • Linear Regression: A basic regression model to check its performance on predicting numerical grades.

3. Model Evaluation: We compared the models using the accuracy metric to assess how well the models predicted the final grades based on the test dataset.

Tech Stack Used

• Programming Language: Python
• Libraries and Packages:
  • pandas: For data manipulation and preprocessing.
  • sklearn.preprocessing.LabelEncoder: For converting categorical variables into numerical formats.
  • sklearn.model_selection.train_test_split: For splitting the dataset into training and testing sets.
  • sklearn.ensemble.RandomForestClassifier: Random Forest model for prediction.
  • sklearn.tree.DecisionTreeClassifier: Decision Tree model for prediction.
  • sklearn.linear_model.LinearRegression: Linear Regression model for prediction.
  • sklearn.metrics.accuracy_score: For evaluating model accuracy.

Models and Results

After running the models on the test dataset, we obtained the following accuracy scores:

• Random Forest: 0.0
• Decision Tree: 0.0
• Linear Regression: 33.33%

Conclusion

From the current setup, Linear Regression showed better performance than Random Forest and Decision Tree, though the model needs further optimization and tuning to improve accuracy.