A Case Study of Neural Network Technique Applied to Buyer''s Response Prediction for a Plastic Technique R&D Service Provider in Taiwan

• Main Authors: Yu-Chi Bau, 包栯綺
• Other Authors: 周世玉
• Format: Others
• Language: en_US
• Published: 2008
• Online Access: http://ndltd.ncl.edu.tw/handle/03949830892577382686

碩士 === 國立中興大學 === 行銷學系所 === 96 === In recent years companies in various industries tended to outsource part of their R&D services to a professional third party to save their costs and speed up new product development. In light of customer retention, a customer’s response to the R&D service has great impact on the possibility of reusing the service from the same service provider. Realizing how customers respond has become a critical issue for a R&D service provider, such as PIDC (Plastic Industry Development Center). Besides, if PIDC know their buyer’s loyalty tendency and propensity to switch to other providers, it can implement adequate relationships management policy with limited marketing resources. The independent variables, including effective communication, trust, satisfaction, commitment, and customer’s cultural market orientation, are used to predict buyer’s response. The predictive performance of models, neural network, decision tree, linear regression, logistic regression and discriminant analysis, are compared. In the prediction of buyer’s response, neural network is a better predictive model than other models. Finally, we suggest that PIDC use the neural network model as their predictive model to manage its relationship marketing policies effectively and efficiently.