A5005530783- Customer churn and segmentation
- Face and Expression Recognition
- Smart Agriculture and AI
- Advanced Clustering Algorithms Research
- Imbalanced Data Classification Techniques
- Complex Network Analysis Techniques
- Consumer Market Behavior and Pricing
- Plant Pathogens and Fungal Diseases
- Customer Service Quality and Loyalty
- Artificial Immune Systems Applications
- Horticultural and Viticultural Research
- Technology Use by Older Adults
- Plant Disease Management Techniques
- Tree Root and Stability Studies
- Data Mining Algorithms and Applications
- Recommender Systems and Techniques
- Data Stream Mining Techniques
- Advanced Image Processing Techniques
China Agricultural University
2021-2024
Abstract In a competitive market, it is of great significance to divide customer groups develop customer-centered personalized products. this paper, we propose segmentation method based on the K-means algorithm and improved particle swarm optimization (PSO) algorithm. As PSO easily falls into local extremum, hybrid (IHPSO) proposed improve accuracy. The full factorial design used determine optimal parameter combination; roulette operator select excellent particles; then, selected particles...
Abstract In an increasingly competitive market, predicting the customer’s consumption behavior has a vital role in customer relationship management. this study, new classifier for prediction is proposed. The proposed methods are as follows: (i) A feature selection method based on least absolute shrinkage and operator (Lasso) Principal Component Analysis (PCA), to achieve efficient eliminate correlations between variables. (ii) An improved genetic-eXtreme Gradient Boosting (XGBoost)...
Purpose The purpose of this study is to propose a new consumer value segmentation method for low-dimensional dense market datasets quickly detect and cluster the most profitable customers enterprises. Design/methodology/approach In study, comprehensive bases (CSB) with richer meanings were obtained by introducing weighted recency-frequency-monetary (RFM) model into common (SB). Further, method, CSB-MBK algorithm was proposed integrating CSB mini-batch k-means (MBK) clustering algorithm....
Purpose Conventional frequent itemsets mining ignores the fact that relative benefits or significance of “transactions” belonging to different customers are in most relevant applied studies, which leads failure obtain some association rules with lower support but from higher-value consumers. Because not all financially attractive firms, it is necessary their values be determined and transactions weighted. The purpose this study propose a novel consumer preference method based on conventional...
An appropriate optimal number of market segments (ONS) estimation is essential for an enterprise to achieve successful segmentation, but at present, there a serious lack attention this issue in segmentation. In our study, independent adaptive ONS method BWCON-NSDK-means++ proposed by integrating new internal validity index (IVI) Between-Within-Connectivity (BWCON) and stable clustering algorithm Natural-SDK-means++ (NSDK-means++) novel way. First, complete the evaluation dimensions existing...