Machine Learning applications with python, sklearn, tensorflow, keras.
Contents
-
Fraud_Detection.ipynb: Credit card fraud detection with deep learning and Random Forest. Apply sampling approaches to tackle imbalance data problem.
-
Kmeans.ipynb: Apply clustering on online retail transactional data with RFM clustering approach using K-means and heirachical clustering methods.
-
Minimizing churn rate.ipynb: Customer churn prediction with Logistic Regression and applying feature selection with Recurssive Feature Elimination.
-
Mnist.ipynb: Classifying MNIST data with tensorflow-keras and CNN
-
Preditcting_liklihood_of_esigning.ipynb: Predicting the likelihood of e-signing a loan based on financial history with Logistic Regression, Support Vector Machine, Random Forest and fine tuning with grid search.
-
Spam and Ham classification.ipynb: Text classification for spam or ham detection with Naive Bayes.
-
TalkingData.ipynb: Performing TalkingData AdTracking Fraud Detection Challenge from Kaggle with boosting methods, AdaBoost, Gradient Boosting and XGBoost.
-
catboost_and_shap.ipynb: Modeling with catboost for default prediction ans mdel explanation with SHAP.
-
movie_recommender_system.ipynb : Applying item-based collaberative filtering for movie recommendation.
-
predict_chichago_crimerates.ipynb: Time series prediction with facebook Prophet.