Customer Behavior Analytics and Visualization
- Publication Type:
- Thesis
- Issue Date:
- 2021
Open Access
Copyright Clearance Process
- Recently Added
- In Progress
- Open Access
This item is open access.
Customer behavior refers to the study of customers and the procedure they use to pick, use, and dispose of products or services. The understanding of customer behavior analysis (CBA) is essential for improving business strategies. The existing studies have explored useful information to analyze customers' behaviors. However, they often fail to allow the analysts, including business management, development, decision-making, etc. The existing research on CBA is limited with four main challenges. First, the analysis of the absence of useful private information and the presence of asymmetric information of customers. Second, exploring customer behavior with multi-dimensional and temporal data is necessary for any competitive and global business to improve its strategies. Third, the estimation of the correlation between claim analysis and risk management is key to avoiding fraud; Fourth, the lack of quantitative research necessitates performance analysis at the class, instance levels, and model visualization. Several approaches to addressing these issues were introduced that are inconsistent with models of rational choice. Nowadays, data mining has become a standard support method for gaining interesting insight into customer behavior. Even though rapid and accurate identification of customer demands is critical to business management, it is not feasible to design all approaches to meet all criteria to be developed. Therefore, this thesis aims to exploit novel data mining techniques blending with visual analytics (VA) to explore customer behavior and provide valuable insight for decision-making support. Insurance data such as questionnaires, demographic, and claim data are used as a testbed to demonstrate our techniques. This thesis is categorized into four main themes: (1) pattern mining (PM) for discovering adverse behavior (AB); (2) visual analytics (VA) for exploring customer behavior; (3) natural language interaction driven data visualization (NLI-driven-DV) to analyze customer claim behavior and manage risk; (4) deep visual analytics (DVA) to provide a wide range of performance evaluations of different methods for understanding customer behavior (UCB). This is one of the first studies to utilize data mining techniques blending with visual analytics (VA) for exploring customer behavior from the insurance business aspect. The empirical results of this thesis show the advantages and effectiveness of the developed methods valuable for researchers and insurance managers (IMs). Moreover, various aspects of insurance data have been researched and integrated into sophisticated visual interactive systems (VIS) to gain a deeper understanding of customer behavior and to better business plans and make decisions.
Please use this identifier to cite or link to this item: