Ensuring continuity in customer payments and analysis with data mining techniques

While companies continue to provide services to their customers, they want to receive the required service fees from their customers in return for this service. However, in some cases, customers may delay their payment for various reasons. This is an undesirable situation that will adversely affect the service provider company. The purpose of this work is to ensure that customers make their payments on time and continuously. This includes examining the data set containing customer invoice payment information and estimating the payment habits of customers using data mining techniques. Within the scope of the study, the invoice payment information data of 10000 different customers belonging to a company operating in the information sector were anonymized within the scope of the law on protection of personal data and modeled with naive bayes, decision trees, support vector machines, random forests, logistic regression algorithms. In the study, the success rates were compared by running the models created with the above data mining classification and prediction methods. Afterwards, the success of the combined model created with the community learning method was evaluated by comparing the algorithm with the highest success rate and the community learning method. As a result of the study, it was seen that the selection of parameters and the use of methods to optimize the model were effective in the success of the algorithms used.