https://github.com/zobayerakib/student-result-data-analysis__data-analysis-project
https://github.com/zobayerakib/student-result-data-analysis__data-analysis-project
linear-regression machine-learning mathplotlib numpy pandas predictive-analytics random-forest-regression seaborn student-result-analysis
Last synced: 2 months ago
JSON representation
- Host: GitHub
- URL: https://github.com/zobayerakib/student-result-data-analysis__data-analysis-project
- Owner: ZobayerAkib
- License: mit
- Created: 2024-06-12T08:12:33.000Z (almost 2 years ago)
- Default Branch: main
- Last Pushed: 2024-06-12T08:17:06.000Z (almost 2 years ago)
- Last Synced: 2025-12-31T15:06:31.812Z (5 months ago)
- Topics: linear-regression, machine-learning, mathplotlib, numpy, pandas, predictive-analytics, random-forest-regression, seaborn, student-result-analysis
- Language: Jupyter Notebook
- Homepage:
- Size: 572 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: License
Awesome Lists containing this project
README
## Student Result Analysis using Regression Models
### Overview
In this project, we aimed to analyze student results using machine learning regression models. We divided the dataset into an 80:20 ratio for training and testing purposes. Our goal was to predict the scores of various subjects ('math_score', 'history_score', 'physics_score', 'chemistry_score', 'biology_score', 'english_score', 'geography_score') based on the number of weekly self-study hours.
### Model Training and Evaluation
We utilized two regression models for training:
- **Linear Regression (LR)**
- **Random Forest Regression (RDF)**
After training the models, we evaluated their performance using the following metrics:
- **Mean Absolute Error (MAE)**
- **Mean Squared Error (MSE)**
- **Root Mean Squared Error (RMSE)**
- **R-squared (R²)**
### Results
| Model | MAE | MSE | RMSE | R² |
|-------------------------|--------|---------|---------|---------|
| Linear Regression (LR) | 10.5696| 157.1996| 12.5121 | 0.0699 |
| Random Forest (RDF) | 10.3420| 155.1183| 12.4352 | 0.0810 |
### Conclusion
Both models yielded similar results, with the Random Forest Regression slightly outperforming the Linear Regression model. However, the overall predictive performance indicates that the relationship between weekly self-study hours and subject scores may be more complex and may require further investigation or feature engineering.
### Data Source
The dataset used for this analysis is sourced from [Kaggle].