Gazelle optimization and conditional variational auto encoder for telecom user service recommendation based on churn analysis
Year: 2024
Type: article
Abstract: Customer churn analysis in telecommunication industry is a very essential factor to be achieved and it makes direct impact to retaining customers and generating income. Various current approaches are utilized to enhancing the customer experience in telecommunication-network service with churn analysis. But those approaches has lots of challenges such as acquisition cost will be high within the networks and also the satisfaction level of recommending the network for the customers was not in a rapid-flow. To-overcome these concerns, Conditional Variational Auto Encoder (CVAE) is developed to enhance the customer experience in telecommunication-network by predicting the churn-customers. The text’s was gathered and pre-processed using tokenization, stemming, stop word removal, spell correction, handling negation, character normalization and lemmatization. Subsequently, Contrastive information extraction with Generation Transformer(CGT) is used to extract features. Appropriate Features are chosen by using t-DSNE (t-Distributed Stochastic Neighbor Embedding). Optimal number of components are selected using GOA. Finally, CVAE is used for predicting churn customers and recommend the high priority network to the user. From the experiment analysis, the proposed approach attains an accuracy of 97.2%, a precision value of up to 94.5%, and a specificity range of 98.1% for service recommendation. Whereas accuracy of 96.45, precision of 93.5% and specificity of 96% is achieved for churn prediction. Thus, the suggested-method is the better choice for enhancing the customer experience in telecommunications-network.
Source: Expert Systems with Applications
Institution: Sri Sivasubramaniya Nadar College of Engineering
Language: English
Subfield: Marketing
Domain: Social Sciences
Open Access status: hybrid
Estimated APC paid: $3,320
OpenAlex ID: W4405611420