(A) study on next generation business models of B2B EC using integration of artificial intelligence tools and qualitative reasoning

Abstract

Electronic marketplaces are becoming important players in several industries because they promise to greatly improve economic efficiency, reduce margins between price and cost, and speed up complicated business deals. The services they provide will expand many companies`` purchasing and selling abilities, and will make prices more dynamic and responsive to economic conditions. This thesis proposes several knowledge-based systems using data mining techniques and two new business models in B2B EC. The proposed business models have deep knowledge bases of application domains based on close relation with buyers or suppliers then support them useful information that shapes the transaction -- price, availability, quality, quantity, and so on. The specialist originator and the sell-side asset exchange are developed to support for buyers and suppliers, respectively. We propose a hybrid prediction system of neural network (NN) and memory based reasoning (MBR) with self-organizing map (SOM) and knowledge augmentation technique using qualitative reasoning (QR). A SOM augmented NN and MBR expert system have good strengths in prediction and in learning the dynamic behavior over a period of time. A QR system rebuilds static knowledge into time-dependent knowledge using QR processes that are easily describes by common sense from qualitative characteristics.