Presenting a model for predicting the buying behavior of consumers of leather products with the NLP approach
Subject Areas : marketing
hamideh seifi shojaei
1
,
Morteza Mahmoudzadeh
2
,
hossein Bodaghi khajeh Nobar
3
,
nasser Feghhi Farahmand
4
1 - Ph.D student of Business Management, Department of Management , Tabriz Branch, Islamic Azad University, Tabriz, Iran.
2 - Assistant Professor of Management Department, ,Tabriz Branch, Islamic Azad University, Tabriz, Iran.
3 - Assistant Professor of Management Department, Tabriz Branch, Islamic Azad University, Tabriz, Iran.
4 - Associate Professor of Management Department, Tabriz Branch, Islamic Azad University, Tabriz, Iran
Keywords: Purchasing Behavior, Machine Learning, Topic Modeling, Latent Dirichlet Allocation (LDA), Natural Language Processing (NLP).,
Abstract :
This research has been done with the aim of providing a model for predicting the buying behavior of consumers of leather products with the NLP approach. The current research is part of applied research in terms of its exploratory research goal and in terms of results. This research is presented in a mixed form and in the form of two qualitative and quantitative studies. Therefore, the mixed research method has been used to conduct these two studies. For the qualitative part and determining the factors, it was obtained by analyzing and combining the results of previous studies and determining the effective indicators by experts who were selected purposefully and snowballed, using the Delphi method. For the quantitative part, topic modeling and natural language processing (NLP) have been used. The components of purchasing behavior of consumers of leather products with thematic modeling method and LDA algorithm and the resulting results, based on super words, graphs and word repetition graphs, three topics were obtained, the first topic is called purchase intention, the second topic is product quality and topic Third, they were named as product characteristics. According to the presented results, the causal relationships between the selected variable and also the variables that were obtained through the qualitative study, which were presented in the form of a model using the Bayesian network method.
