Simultaneous Co-Clustering and Classification in Customers Insight

3rd International Conference on Mathematics, Science and Education 2016

数学;自然科学;教育

Anggistia, M.^1 ; Saefuddin, A.^1 ; Sartono, B.^1

Department of Statistics, Bogor Agricultural University, Indonesia^1

关键词: Co-clustering;    Customer behaviour;    Heterogeneous dataset;    Homogeneous group;    Marketing strategy;    Prediction accuracy;    Predictive modeling;    Product characteristics;   

Others  :  https://iopscience.iop.org/article/10.1088/1742-6596/824/1/012033/pdf
DOI  :  10.1088/1742-6596/824/1/012033

来源: IOP

【 摘 要 】

Building predictive model based on the heterogeneous dataset may yield many problems, such as less precise in parameter and prediction accuracy. Such problem can be solved by segmenting the data into relatively homogeneous groups and then build a predictive model for each cluster. The advantage of using this strategy usually gives result in simpler models, more interpretable, and more actionable without any loss in accuracy and reliability. This work concerns on marketing data set which recorded a customer behaviour across products. There are some variables describing customer and product as attributes. The basic idea of this approach is to combine co-clustering and classification simultaneously. The objective of this research is to analyse the customer across product characteristics, so the marketing strategy implemented precisely.