Research On Information Credibility Of E - Commerce Subject

The booming E-commerce market in China is an important driving part of the world’s economy. While driven by the temptation of huge profits in this growing market, part of the trading participants (especially the sellers) ignore their reliability and harm the benefits of other participants. Since the relevant laws and regulations are incomplete and the market regulatory is not clear, the offending participants will not get punished in time. Meanwhile, the related trading platform and the whole e-commerce transactions market are suffering big losses. For example, when some of the Taobao’s sellers have deceived their consumers then it resulted in the whole Taobao mart’s credit crisis. In another case, the’Guangdong Huijin Gold Exchange’ leaded to the rectification storm of the whole bulk stock exchange systems in2012.The government puts the greatest emphasis on the trading participants’ reliability in E-commerce (especially for those related to the online bulk stock transaction). In order to provide a way to regulate the bulk stock online trading, the National Science and Technology Department leads the research and development for bulk stock credible trading technology. The Fudan University, the Tsinghua University and the Itrus China Company all cooperate in a project titled "Research and development for trusted bulk stock online transaction technology and its application". For which one of the sub-projects is the research on methods for evaluating reliability of trading participants’information for bulk stock.This thesis presents a methodology for modeling the trading enterprises’ information reliability, which is in large part based on the survey of e-business trading characteristics and enterprise data characteristics from ECData.org. Aspects such as integrity, formativeness, updating timeliness, information irregularities, authoritative and corporate credibility are all taken into consideration in order to evaluate the reliability of an enterprise’s information. As the corporate credibility evaluation is the key to evaluate its reliability, the design of credibility indicators and evaluation algorithm is also an important aspect of this research. The evaluation of enterprises’ credibility lies in its credit and reputation. Trading enterprises will evaluate other trading participants’ credit. Experts with related knowledge or working experience are asked to evaluate enterprises’reputation for set of different indicators. These include transaction subject and object of a transaction, the transaction process feature information, reputation indicators include social responsibility, emotional attractiveness and other4other aspects. The weight for each of these indicators is calculated using the AHP and entropy method. The main data source for the credit assessment is trading contracts, while reputation assessments bases on the experts’ knowledge accumulation. Single calculated reliability value affects the final enterprise credibility by reliability transferring functions, designed to consider the ’recent behavior trust’ principle. Finally, the designed reliability model and reliability evaluation model are evaluated by conducting experiments that take usage of the ECdata.org data.