Integrating KPSO and C5.0 to analyze the omnichannel solutions for optimizing telecommunication retail

作者:

Highlights:

• This study analyzes different groups of buyers of the omnichannel.

• The telecommunications retailers provide precise customer demand.

• The telecommunications retailers are good at using the omnichannel solution.

摘要

Telecommunication system providers offer many special number commodities, and all marketing staff sell the commodities and special number combinations according to their marketing experience; consequently, telecommunication retailers find it difficult to consider both user demands and profits in their marketing strategies. This study proposes a classification model integrating K-means Particle Swarm Optimization (KPSO) and C5.0. The particles' PSO only followed pbest and gbest when moving, resulting in the disadvantage that PSO may easily fall into a local optimal solution. First, clustering analysis is carried out using the KPSO clustering method. Second, classification rules for clustering results are formulated by the C5.0 classification method, and a classification model is established in order to achieve effective descriptions of the clustering rules. The methods proposed herein can help retailers find and utilize complementary tariff products for mobile numbers as the basis for future sales and procurement. This study also analyzes the best media for customers' mobile phone purchase methods and utilizes different groups of buyers of the omnichannel. The proposed model is also able to categorize new future tariffs and can further conceptualize the clustering results in the analysis of telecom tariffs and product mix. Finally, the results effectively assist the telecommunications retail industry when considering procurement, product projects, sales, and marketing solutions.

论文关键词:K-means particle swarm optimization,C5.0,Clustering analysis,Classification method,Omnichannel

论文评审过程:Received 26 January 2017, Revised 14 September 2017, Accepted 21 December 2017, Available online 22 December 2017, Version of Record 18 April 2018.

论文官网地址:https://doi.org/10.1016/j.dss.2017.12.009