President University Repository

PREDICT CHURN TELECOMMUNICATION CUSTOMER USING MACHINE LEARNING IN PYTHON

Show simple item record

dc.contributor.author Triferyawan, Okky Dwi
dc.date.accessioned 2023-03-27T08:06:19Z
dc.date.available 2023-03-27T08:06:19Z
dc.date.issued 2022
dc.identifier.uri http://repository.president.ac.id/xmlui/handle/123456789/10787
dc.description.abstract Customers play a very important role in every aspect of business, whether it's a business in technology or non-technology. Therefore, when companies want to in-crease their efforts in retaining customers, they must be able to accurately predict in advance, whether customers will churn in the future or not is a very powerful tool for the team in the company. By using machine learning technology and artificial intelli-gence the possibility to predict churn increases a lot. Our proposed methodology is consist of five phases. In the first and the second one is for exploratory data analysis (EDA) and data pre-processing is performed. In the third phase is the prediction pro-cess, most popular predictive models based on data mining techniques have been ap-plied, such as, logistic regression, support vector machine (SVC), random forest, deci-sion tree and naive bayes. In addition, logistic regression classification has been im-proved it with hyperparameter tuning (random search) so that our model performance improves and becomes more accurate. Finally, the results obtained on the test predic-tions were evaluated using logistic regression models, support vector machines, ran-dom forests, decision trees, naive bayes and logistic regression with hyperparameter tuning. It was found that logistics with hyperparameter tuning provided the highest accuracy of each model with a score of 0.80312, therefore it will be used as a model for prediction applications later. en_US
dc.language.iso en_US en_US
dc.publisher President University en_US
dc.relation.ispartofseries Information System;012201800005
dc.subject Python en_US
dc.subject Machine Learning en_US
dc.subject Exploratory Data Analysis en_US
dc.subject Churn Prediction en_US
dc.subject Prediction Model en_US
dc.subject Hyperparameter Tuning en_US
dc.subject Predict Application in Streamlit en_US
dc.title PREDICT CHURN TELECOMMUNICATION CUSTOMER USING MACHINE LEARNING IN PYTHON en_US
dc.type Thesis en_US


Files in this item

This item appears in the following Collection(s)

Show simple item record

Search Repository


Advanced Search

Browse

My Account