enliktjioe / bda2020

Repository made for Business Data Analytics Course in Spring 2019/2020 (University of Tartu)

BDA2020

• Repository made for Business Data Analytics Course in Spring 2019/2020 (University of Tartu)
• I made it open-source, use it responsibly

Homework (HW) keywords

• HW1: Introduction (Python and Pandas, Data Understanding, Data Preparation, Gathering Interesting Facts)
• HW2: Customer Segmentation and Regression Problem (RFM, Kmeans, DBSCAN, Imputation, Correlation, MAE, RMSE, R-square)
• HW3: Classification Problems (Employee Attrition, Pandas Get Dummies, DecisionTree, SVM, SVC, Logistic Regression, RandomForestClassifier, XGBClassifier)
• HW4: Cross-sell / Up-sell using Recommendations (Market Basket Analysis, Collaborative Filtering, apriori, association_rules, one-hot encoding, LabelEncoder, fuzz, SVD, KNNBasic, NormalPredictor, Neural Network, Embedding, Reshape, Dense)
• HW5: A/B Testing & Uplift Modelling (Hypothesis, QQ-plot, shapiro test, KNN, Logistic Regression, XGB Model)
• HW6: Knowledge graphs & Fairness and Explainability (triples, networkx, tensorflow, ampligraph, ComplEx, XGBoost, eli5 feature importances, shap)

Practice Session (Lab) keywords

• Lab01: Python Basic, Data Science Tutorial (Numpy, Pandas, Matplotlib), Work with Missing Values, Pandas Profiling Report)
• Lab02: Data Science and Descriptive Data Analysis (Data cleaning, loc, iloc, Data Viz)
• Lab03: Customer Segmentation (Clustering) (RFM, Kmeans, Gaussian Mixture Models(GMM))
• Lab04: Customer Lifecycle Management - Regression (LinearRegression, MSE, R2, MAE)
• Lab05: Customer Lifecycle Management: Churn Prediction - Classification (MinMaxScaler, Logistic Regression, XGB)
• Lab06: Model Evaluation - Classification (Unbalanced dataset, random oversampling, ROC/AUC, Precision, Recall, K-Fold Cross Validation)
• Lab07: Recommendation System (Item-based collaborative filtering, Content-Based Filtering, CountVectorizer)
• Lab08: Cross-sell/Up-sell: Recommendations system using Neural networks + Embeddings (ANN, Deep Learning, Keras,
• Lab09: A/B Testing
• Lab10: Uplift Modeling
• Lab11: Fraud Detection (Hyperparameter tuning, GridSearchCV)
• Lab12: Knowledge Graph (ampligraph, knowledge embeddings, PCA,
• Lab13: Fairness and Explainability (feature importances, eli5, Skater, SHAP,

Troubleshooting

URLError: <urlopen error [SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed (_ssl.c:847)>

• Found on macOS 10.12.6 using Jupyter Notebook, insert this code cell

import ssl ssl._create_default_https_context = ssl._create_unverified_context