Student Performance Analysis & Prediction

This project performs a descriptive and predictive analysis of the "Students Performance in Exams" dataset. The goal is to first understand the factors that influence student scores (descriptive) and then to build a machine learning model to predict student performance (predictive).

The full analysis is documented in the Jupyter Notebook, with a final summary available in the project report.

Project Overview

1. Descriptive Analysis (EDA): An exploration of the dataset to find key patterns and insights. This includes looking at the impact of gender, parental education, and test preparation on student scores.
2. Predictive Modeling: A classification task to predict a student's "condition" based on their background. A student is classified as being in a "Critical" condition if they score below 40 in at least one subject.

Key Findings

• Correlation: There's a strong positive correlation between reading and writing scores.
• Influential Factors: Both parental level of education and the completion of a test preparation course show a significant positive impact on exam scores.
• Data Imbalance: The dataset is highly imbalanced, with 949 students in "Great" condition and only 51 in "Critical" condition.

Predictive Modeling Results

Three different classification models were trained and evaluated to predict a student's condition. The model performances were:

Algorithm	Test Accuracy(%)	AUC Value
KNN	89.18	0.892
Logistic Regression	88.52	0.885
SVM	86.88	0.868

The K-Nearest Neighbors (KNN) model was the top-performing model for this classification task.

Dataset

• Source: Kaggle: Students Performance in Exams
• Local Copy: The dataset is included in the data/ directory of this repository.

Technologies Used

• Python
• Pandas & NumPy
• Seaborn & Matplotlib
• Scikit-learn
• Jupyter Notebook

How to Use This Repository

1. Clone the repository:

   git clone <your-repository-url>

2. Install dependencies:

   pip install -r requirements.txt

3. Run the analysis: Open and run the Jupyter Notebook to see the full process.

   jupyter notebook notebooks/student_performance_analysis.ipynb

4. View the Report: For a detailed summary of the project, see the PDF in the report/ directory.