Modelling Time Series Customer Preference Based on E-commerce Website
- 10.2991/assehr.k.211209.525How to use a DOI?
- time series customer preference; e-commerce website; opinion mining; COA based DENFIS
In the research of customer preference for products, we need to collect enough customer data first, and then we can analyze it to realize the effective establishment of customer preference model. Although a reasonable model can be obtained, the customer’s own preference presents a dynamic state, which changes disorderly with the change of time. This requires a systematic analysis of the time series data of customer preference based on different time periods, so as to complete the modeling of customer preference. Through research, this paper proposes a new time series customer preference modeling method, which can effectively process and analyze online customer comment data of e-commerce websites, including opinion mining and chaos optimization algorithm (COA) based dynamic evolutionary neuro fuzzy inference system (DENFIS). Finally, a case verification method is adopted, and a hair dryer is selected to verify the effectiveness and rationality of the method. The comparison results show that the proposed COA based DENFIS approach performs better than K-means-based ANFIS, fuzzy c-means-based ANFIS, subtractive cluster-based ANFIS and DENFIS in terms of mean absolute percentage error.
- © 2021 The Authors. Published by Atlantis Press International B.V.
- Open Access
- This is an open access article under the CC BY-NC license.
Cite this article
TY - CONF AU - Huimin Jiang AU - Chunsheng Li AU - Farzad Sabetzadeh PY - 2021 DA - 2021/12/15 TI - Modelling Time Series Customer Preference Based on E-commerce Website BT - Proceedings of the 2021 3rd International Conference on Economic Management and Cultural Industry (ICEMCI 2021) PB - Atlantis Press SP - 3222 EP - 3227 SN - 2352-5428 UR - https://doi.org/10.2991/assehr.k.211209.525 DO - 10.2991/assehr.k.211209.525 ID - Jiang2021 ER -