Maybank is the largest financial institution in Malaysia that started in 1970 It provides different types of financial and banking services to the public which include deposit and investment, loan financing, wealth management, banking and so on Due to an intense competition among the financial institutions or banks, Maybank has provided a broad range of products and services to attract the customers During the marketing of those services and products, Maybank has spent a huge amount of money The purpose is to predict the customers will subscribed deposit or not. Conclusion :
- SVM and logistic regression have the highest accuracy among all but AUC graph in SVM has highest value than all of models
- For learning curve in Cross Validation Score, KNN has overfitting but another model are well fitting
- From scalability of model, Naive Bayes has the fastest fit times but SVM has the slowest fit times
- From the performance model graph, SVM has the constant performance score while the fit times increase
- SVM can be used as the baseline model to predict deposits subscribed by telephonic marketing
- SVM and KNN has the same importance features previous,marital,education and also for Logistic Regression and Naïve Bayes has the same importance features (Age and campaign)
Let's Check out my pdf presentation and html python code jupyter notebook !!! Don't hesitate to contact me if you want to do some correction or discussion ! kurniaanwarraif@gmail.com or https://www.linkedin.com/in/kurnia-anwar-ra-if/ #DataScience #ClassificationModelling #EDA #dibimbing.id