Projeto referente a Formação Cientista de Dados da Data Science Academy
Customer churn can refer to the loss of customers in a particular company. Customer loyalty and Customer churn always add up to 100%. If a company has a loyalty rate of 60%, then the customer loss rate is 40%. Under the 80/20 profitability rule, 20% of customers are generating 80% of revenue. Therefore, it is important to anticipate the users who are likely to leave the business and the factors that affect the customer's decisions.
In this project, I used an algorithm to predict the Customer Churn of a Telecom Companie.
LR: 0.761848
LDA: 0.766047
NB: 0.827235
KNN: 0.803239
CART: 0.898620
RANDOM FOREST: 0.926215
ADB: 0.855429
GRADIENT: 0.928014
XGB: 0.926215
SVM: 0.855429
The best performance was with the GRADIENT algorithm
model score: 0.928