Projeto completo de Churn Prediction para a redução de cancelamentos de contrato em uma empresa de telecomunicações, utilizando técnicas de Análise de Dados, Ciência de Dados e diferentes modelos de Machine Learning
Churn Prediction by Luis Guimarães - Jupyter Notebook Viewer
-
Bibliotecas
- pandas
- seaborn
- matplotlib
- numpy
- Dython
- sklearn
- imblearn
- lightgbm
- xgboost
- shap
-
Pacotes
- associations
(dython.nominal) - mutual_info_classif
(sklearn.feature_selection) - LabelEncoder
(sklearn.preprocessing) - RandomUnderSampler
(imblearn.under_sampling) - LabelEncoder
(sklearn.preprocessing) - MinMaxScaler
(sklearn.preprocessing) - train_test_split
(sklearn.model_selection) - LGBMClassifier
(lightgbm) - XGBClassifier
(xgboost) - cross_val_score
(sklearn.model_selection) - LogisticRegression
(sklearn.linear_model) - RandomForestClassifier
(sklearn.ensemble) - DecisionTreeClassifier
(sklearn.tree) - SGDClassifier
(sklearn.linear_model) - KNeighborsClassifier
(sklearn.neighbors) - GradientBoostingClassifier
(sklearn.ensemble) - LinearSVC
(sklearn.svm) - GridSearchCV
(sklearn.model_selection) - RandomizedSearchCV
(sklearn.model_selection) - cross_val_predict
(sklearn.model_selection) - confusion_matrix
(sklearn.metrics) - classification_report
(sklearn.metrics) - roc_auc_score
(sklearn.metrics) - Explainer
(shap)
- associations
- Churn Rate: o que é e como reduzir para sua empresa crescer (https://resultadosdigitais.com.br)
- CHURN: você sabe como calcular a taxa de cancelamento de clientes? (https://www.cortex-intelligence.com/)
- IBM Developer (https://developer.ibm.com/)
- Python ValueError: could not convert string to float (https://itsmycode.com/)
- Dython (http://shakedzy.xyz/dython/)
- How To Find Correlation Value Of Categorical Variables. (https://medium.com/@knoldus/)
- Mutual Information: Locate features with the most potential. (https://www.kaggle.com/code/ryanholbrook)
- Everything you need to know about Min-Max normalization: A Python tutorial (https://towardsdatascience.com/)
- Como lidar com dados desbalanceados em problemas de classificação (https://medium.com/data-hackers/)
- Qual a melhor métrica para avaliar os modelos de Machine Learning (https://www.flai.com.br/juscudilio/)
- Difference between GridSearchCV and RandomizedSearchCV (https://www.kaggle.com/)
- Introdução ao Scikit-learn - Parte 3: avaliando a qualidade do modelo via cross-validation (http://computacaointeligente.com.br/)
- How to interpret and explain your machine learning models using SHAP values (https://m.mage.ai/)
- Introduction to SHAP Values and their Application in Machine Learning(https://towardsdatascience.com/)
- Explain Your Model with the SHAP Values (https://towardsdatascience.com/)