الفهرس | Only 14 pages are availabe for public view |
Abstract Abstract Credit has become more stringent for banks in international markets, diverting focus to internal clients and their fund-raising deposits. This push has led to an increased demand for awareness of a customer’s deposit and personal loan conduct, especially their reaction to telemarketing and cyber marketing campaigns. banks have a huge number of customer-related records. Machine learning techniques have a tremendous opportunity to analyze the data in this field and turn it into useful knowledge for maintaining and gaining clients. Missing data problems is Rumor in the real world both observational and experimental types of research. The most used single regression approaches are the full-case study, and mean imputation methods result in overly skewed estimates of all missed value situations where more than 10% of the data of the subject is missing. Moreover, the best effective single-imputation technique turned out to be single regression, but standard errors are underestimated because missing values uncertainty is not integrated. The aim of this research is to predict the success of cyber marketing, and we propose an approach for missing data imputing based on machine learning algorithms advanced. The dataset consists of instances and 14 characteristics. With SPSS, the WEKA instrument, and python, were used to assess and compare four classification models (i.e., Decision Tree, Random Forest, XG Boost, and support vector machines): ROC and accuracy. Overall, with 97.46% accuracy, the DT has the best efficiency. Key words: Machine Learning; Classification Problem; Prediction Model; Missing Data; Auto Direct Marketing. |