The Study of Credit Evaluation of Business Websites Using Support Vector Machines

With the rapid developing of the Internet, more and more business Websites are appearing all around the world. The problem about trustworthiness in the business Website also becomes more critical. The deal methods are very different between the business Websites and the traditional enterprises in evaluating the credit rank due to the feature of business Websites themselves. In this paper, a new credit evaluating method, weighted support vector machine (WSVM), is proposed to score business Website credit ranks, which associates with website building time. Firstly, we train WSVMs using 50 enterprises´ credit indexes, then score three credits of enterprise A, B, C. Secondly, we utilize statistics hypothesis validation to prove indiscrimination of polynomial kernel, RBF kernel and sigmoid kernel. The hypothesis validation and the experimental result shows that the forecasting method of this paper is stable, highly accurate, and strongly robust.