No abstract

In India, the banks have a formidable edge in maintaining their customer retention ratio for past few decades. Downfall makes the private banks to reduce their operations and the nationalised banks merge with other banks. The researchers have used the traditional and ensemble algorithms with relevant feature engineering techniques to better classify the customers. The proposed algorithm uses a Meta classifier instead of an ensemble algorithm with an adaptive genetic algorithm for feature selection. Churn prediction is the number of customers who wants to terminate their services in the banking sector. The model considers twelve attributes like credit score, geography, gender, age, etc, to predict customer churn. The project consists of five modules as follows. First is the pre-processing module that identifies the missing data and fills the value with mean and mode. Second is the data transformation module where, the categorical data is converted into numerical data using label encoding to fasten the computations. The converted numerical data is normalized using the standard scalar technique. The feature selection module identifies the essential attributes using DragonFly and Firefly (Hybrid Fly) algorithms. The classification module designs an intelligent Meta learner, which combines the Ensemble Algorithm Extreme Gradient Boosting (XGBOOST) with base classifiers as "Extra Tree Classifier" and "Logistic Regression" to predict the churn customers.

Missing data padding is an important problem that is faced in real time. This makes the task of data processing challenging. This paper aims to design a solution for this problem which is ways different from traditional approaches. The proposed method is based on co-cluster sparse matrix learning (CCSML) method. This algorithm learns without reference class, and even with data continuous missing rate as high as the existing techniques. This method is based on a tensor optimization model and labeled maximum block. The computational models of sparse recovery learning are based on low-rank matrix and co-clusters of genome-wide association study (GWAS) data matrices, and the performance is better than existing techniques.

No abstract

No abstract