Presentation | 2004/10/12 Consumer Behavior Modeling using Bayesian Networks For Analyzing Latent Condition behind Consumers' Behavior Tomoko MURAKAMI, Akihiro SUYAMA, Ryohei ORIHARA, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | In this paper we describe a consumer behavior model using bayesian networks. The model aims to support the marketers. Answer data collected by a questionnaire surveys of the market is often incomplete. It is difficult to collect latent information such as a state of consumers' mind by a questionnaire survey. Considering these facts we aplly bayesian networks to construct a consumer behavior model. The merits of the method are to utilize knowledge of domain's specialist, to introduce unobservable variable and the robustness against the lack of data. We decide the structure of the model utilizing answer data collected by a questionnaire survey to investigate lifestyle of the consumers. Conditional probability for each nodes in the network is calculated based on frequency of the data for observable node and by expectation maximization(EM) algorithm for unobservable nodes. We perform some experimental studies by probability inference for this model to specify the loyal consumers and to estimate if a brand arouses target consumers' interest. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | bayesian networks / consumer behavior model / unobserved variable / probability inference |
Paper # | NC2004-70 |
Date of Issue |
Conference Information | |
Committee | NC |
---|---|
Conference Date | 2004/10/12(1days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | |
Chair | |
Vice Chair | |
Secretary | |
Assistant |
Paper Information | |
Registration To | Neurocomputing (NC) |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Consumer Behavior Modeling using Bayesian Networks For Analyzing Latent Condition behind Consumers' Behavior |
Sub Title (in English) | |
Keyword(1) | bayesian networks |
Keyword(2) | consumer behavior model |
Keyword(3) | unobserved variable |
Keyword(4) | probability inference |
1st Author's Name | Tomoko MURAKAMI |
1st Author's Affiliation | Corporate Research & Development Center, Toshiba Corporation() |
2nd Author's Name | Akihiro SUYAMA |
2nd Author's Affiliation | Corporate Research & Development Center, Toshiba Corporation |
3rd Author's Name | Ryohei ORIHARA |
3rd Author's Affiliation | Corporate Research & Development Center, Toshiba Corporation |
Date | 2004/10/12 |
Paper # | NC2004-70 |
Volume (vol) | vol.104 |
Number (no) | 349 |
Page | pp.pp.- |
#Pages | 6 |
Date of Issue |