Optimized Attention Enhanced Temporal Graph Convolutional Network Espoused Research of Intelligent Customer Service System based on Natural Language Processing Technology

Zhifeng Wei,Hongyan Wang,Qiang Xu,Yi Qu,Wei Xing

DOI: https://doi.org/10.1080/08839514.2024.2327867

IF: 2.777

2024-03-20

Applied Artificial Intelligence

Abstract:Consumers have begun to move their attention away from product functioning and toward value probably extracted from items. Companies have begun to use customer service systems (CSS) in response to this trend, which are business models that give clients with not solitary tangible items as well as intangible facilities. Even with substantial investigation on Smart CSS frameworks, rare of this frameworks considered customers active data producers actively creating data for the Smart CSS. Furthermore, the majority of them offered a generic remedy rather than a personalized one. To classify customer service systems, performance metrics, such as precision, accuracy, F1-score, Recall (Sensitivity), Specificity, Error rate, Computation time, and RoC are considered. The performance of AETGCN-NGOA-CSS approach attains 19.11%, 24.12% and 28.13% high specificity, 24.93%, 23.04%, and 9.51% lower computation time, 15.2%, 25.45% and 13.91% higher ROC and 8.45%, 20.98%, and 27.55% higher accuracy compared with existing methods, such as developing personalized recommendation system in smart product service system depend on unsupervised learning model (CSS-BERT), Cognitive Decision-Making approaches in Data-driven Retail Intelligence: Consumer Sentiments, Choices, Shopping Behaviors (CSS-CDMA), e-Commerce Online Intelligent Customer Service System under Fuzzy Control (CSS-FFNN), respectively.

computer science, artificial intelligence,engineering, electrical & electronic

What problem does this paper attempt to address?

### Problems the Paper Attempts to Solve

This paper aims to address several key issues present in existing intelligent customer service systems (CSS):

1. **Insufficient Proactivity of Customer Data**: Most existing intelligent customer service systems fail to fully consider the role of customers as proactive data producers, meaning the vast amount of data generated by customers during service usage is not effectively utilized.
2. **Lack of Personalized Service**: The solutions provided by most existing intelligent customer service systems are relatively generic and fail to offer personalized services based on the specific needs of each customer.
3. **Performance Metrics Improvement**: There is room for improvement in performance metrics such as precision, accuracy, F1 score, recall (sensitivity), specificity, error rate, computation time, and ROC in existing intelligent customer service systems.

To address these issues, the paper proposes an optimized Attention-Enhanced Temporal Graph Convolutional Network (AETGCN-NGOA-CSS), which combines natural language processing techniques to improve the performance and personalization level of intelligent customer service systems.

### Main Contributions

1. **Introduction of the AETGCN-NGOA-CSS Model**: This model enhances the processing capability and classification accuracy of customer data by optimizing the attention mechanism and temporal graph convolutional network.
2. **User Needs Identification**: By using customer journey maps and sentiment analysis, the specific needs of customers are identified, providing a foundation for personalized services.
3. **Exploration of Key Data**: Key data is collected and analyzed from various data sources (such as social media comments, rapid consumer feedback, etc.) to better understand customer needs.
4. **Optimization Algorithm Application**: The Northern Goshawk Optimization Algorithm (NGOA) is used to optimize the weight parameters of the AETGCN model, further enhancing the model's performance.

Through these methods, the paper aims to build a more efficient, accurate, and personalized intelligent customer service system to meet the demand for high-quality customer service in modern enterprises.