Predictive Analysis of Customer Retention Using the Random Forest Algorithm

Retaining customers is becoming a measurement focus in an industry with increasing competition. The concept of customer retention has become research study the sales industry, because it difficult to retain and easily switch other brands. Customer repurchase decisions business world are very competitive. satisfaction directly proportional rate, if not satisfied then automatic rate will be low. If company able meet expectations, have serious impact on company, namely moving services. Service factors, price, profit value, trust affect retention. One factors that influence consumers service quality. A predictive plan needed data mining using random forest algorithm. algorithm method generates number trees from sample data, where creation one tree during training does depend previous tree, decision based most voting. voting results several formed boundaries used as class determination classification process votes winners determine class. This aims analyze loyalty, satisfaction. So can make easier monitor at company. seen percentage about 81.12% 18.87% churn. result feature evaluation shows customer_activity highest retention, followed by subtotal qty.