The goal of this project was to use machine learning algorithms to correctly classify Enron employees who were known to have committed fraud from a data set that contained a general selection of employees who had committed fraud and who had not committed fraud. The report is presented here. The software used for this task was Python and the library used was the sk-learn package.
The machine algorithms used were:
- Gaussian Naive Bayes
- Support Vector Machine
- K-Nearest Neighbors
- Adaboost
- Random Forest
Recursive elimination of features was utilized, as was Select-K-Best and Principal Component Analysis.
| File Name | Description |
|---|---|
| References.txt | References Used |
| enron_free_responses.markdown | Markdown Report with Questions |
| my_classifier.pkl | Pickle File of Classifier |
| my_dataset.pkl | Pickle File of Data Set |
| my_feature_list.pkl | Pickle File of Features Used |
| poi_id.py | Python Code With ML Algorithms |
| tester.py | Python Code for Cross-Validation |