At this phase, we have the data in the data warehouse and it's ready for mining, this repository includes all the work that has been established to make the machine learning models and the data mining processes.
Use the package manager pip or conda to install The Requirment Packages.
- Numpy
- Pandas
- Matplotlib
- Sklearn
- Keras
- Pytorch
Here is a result from mining the data and deducing that the algorithmics subjects are the critical units to obtain in order to be successful in the rest of the curriculum of Computer science:
Also The results of the machine learning models for the first problem about predicting the success or the failure of the student based on his socio demographic profile information
| Algorithm | Accuracy |
|---|---|
| Deep NN | 81% |
| Naïve Bayes | 68% |
| Logistic Regression | 80% |
| SVM | 81% |
| Decision Tree | 78% |
| Random Forest | 79% |