Methodologies For Semantic Information Sharing In E-Commerce

This paper studies the practical methodologies for sharing transactional information in E-commerce, emphasizing the solutions to share public commercial information including product information, sales information, offer and demand opportunities, and consumer reviews. The author presents three methodologies for semantic information sharing in E-commerce, which are based on RSS/Atom, microformats and RDF respectively. These methodologies are compatible with Web standards, supported with real cases, underpinned by the two new theories found by the author, and therefore feasible in implementation.In the part of fundamental theories, the author presents the principle of information sharing and the incentive mechanism for information sharing by discussing and extending relevant concepts and theories in information science. Inspired by shelf ready packaging (SRP), the author suggests online ready packaging (ORP) as a new component into the current cooperation between product manufacturers and their online retailers.Current information sharing methods in B2C and C2C E-commerce are primarily human-to-human information communication based on content tree. Information sharing in B2B E-commerce is mainly data interchange between computer systems. The methodologies proposed in this paper are oriented both to humans and to computers, featured in the integration of information communication and data interchange, and pertaining to ways to share E-commerce information on the semantic Web. The research emphases are on the applications of extended RSS/Atom feeds to E-commerce, and on the methods to represent commercial semantic information in Web pages by microformats and RDF.Information from external suppliers can be used for mashup assembly or syndication assembly on client-side. With regards to the utilization of shared commercial information, the author analyzes commercial search syndication, presents a solution for mashing up an online bookstore with an academic library system, and designs a new framework for building a loose federated database based on a group of external semantic Web pages.The author finds the effect of data stratification and information assembly and the jumping law of information value-in-use through theoretically abstracting the information representation commonalities among the three information sharing methodologies. Raw data, structure data and style data are the three fundamental layers of digital information. These three fundamental layers can be separated, and it is also meaningful to combine any two or three layers together. Multiple layers of applicational information can be built upon the same fundamental layers. Information value-in-use increases smoothly during the process of information evolution from a lower ordered state to an adjacent higher ordered state. A mutation occurs when arrival at a higher ordered state, enabling a totally new application and resulting in a jump of information value-in-use.The full text is composed of nine chapters, 225,000 Chinese characters in total, including 45 figures. 31 refs in Chinese and 160 refs in English are cited.