This notebook conducts an in-depth exploration of a dataset containing students' exam results, sourced from Kaggle . The primary goal of this analysis is to uncover the various factors that influence students' academic performance. These factors include variables such as family education levels, the type of lunch students receive, test preparation course participation, and etc. Through a combination of data preprocessing, descriptive statistics, and data visualization techniques, we aim to gain insights into the relationships between these variables and students' exam scores.
Additionally, machine learning models will be employed to predict student success in exams based on the previously mentioned variables. This analysis provides an opportunity to not only shed light on the educational landscape but also to offer actionable recommendations for educators and policymakers.
| Model | Accuracy |
|---|---|
| Linear Regression | 0.23 |
Low Accuracy of model can be attributed to several factors:
- Insufficient Data
- Poor Quality of Data
- Poor data preparation
- feature engineering / feature selection
- Hyperparameter Tuning
- and many more unseen factors!