Impact of Hyperparameters on Deep Learning Model for Customer Churn Prediction in Telecommunication Sector

Abstract

In this paper, in order to predict a customer churn in the telecommunication sector, we have analysed several published articles that had used machine learning (ML) techniques. Significant predictive performance had been seen by utilising deep learning techniques. However, we have seen a tremendous lack of empirically derived heuristic information where we had to influence the hyperparameters consequently. Here, we had demonstrated three experimental findings, where a Relu activation function was embedded and utilised successfully in the hidden layers of the deep network. We can also see that the output layer had the service ability of a sigmoid function, in which we had seen a significant performance of the neural network model and obviously it was improved. Furthermore, we had also seen that the model's performance was noticed to be even better, but it was only considered better though when the batch size in the model was taken less than the test dataset’s size, respectively. In terms of accuracy, the RemsProp optimizer beat out the other algorithms such as stochastic gradient descent (SGD). RemsProp was seen even better from the Adadelta algorithm, the Adam algorithm, the AdaGrad algorithm, and AdaMax algorithm as well.

Archimedes Optimization Algorithm-Based Feature Selection with Hybrid Deep-Learning-Based Churn Prediction in Telecom Industries

“…Comparison analysis outcomes of the AOAFS-HDLCP technique with other approaches[20,39,40].Along with that, the AIJOA-CPDE approach illustrated reasonable outcomes with an accu y of 91.28%, prec n of 95.52%, reca l of 91.29%, F score of 94.08%, and an AUC score of 91.29%. However, the AOAFS-HDLCP technique gained the maximum performance with an accu y of 94.65%, prec n of 96.92%, reca l of 94.65%, F score of 95.74%, and an AUC score of 94.65%.…”

“…In Figure 8, the ROC analysis curve achieved by the AOAFS-HDLCP algorithm for 70:30 TRS/TSS dataset is shown. This figure indicates that the AOAFS-HDLCP system Table 4 shows the results of the comparison analysis conducted between the proposed AOAFS-HDLCP method and the existing methods [20,39,40]. The experimental values infer that the DR and LR models exhibited poor results, whereas the SVM, SGD, and RM-SProp approaches achieved slightly increased performance.…”

Application of Classification Methods in Forecasting broadband internet subscribers leaving the network on Broadband Internet Subscribers Leaving the Network

“…The model achieved prediction performance with 74% accuracy, 78% precision, and 68% recall. Domingos and Ojeme et al [16] and Dalli [17] present empirical analysis on the impact of different hyperparameters when using deep neural networks (DNN) to predict customer churn in the banking sector. In this paper, the data set used is loaded from Kaggle.…”

Banks customer churn analysis and prediction using random forest

“…Researchers have developed many models for churn prediction, for example, logistic regression, Naïve Bayes, K-means, and random forest etc. [1][2][3][4][5]. The random forest might be one of the best models among them because it can get good results with fewer data [6][7][8][9].…”