Heterogeneous network-based trust analysis: a survey
ACM SIGKDD Explorations Newsletter
| Hi-index | 0.00 |
In a variety of domains, the amount of information grows rapidly; new sources and types of information are proliferating. In recent years, the world-wide web information has been growing at a dramatic pace and might have become the most important information source for most people. In daily life, when people encounter some problems, they tend to retrieve the information from the Web search engines. Many business search engines are efficient at identifying the best web sites for any given keyword query. Unfortunately, the information on the web is not always correct. Moreover, different web sites often provide different information on a subject. One headache problem we have to face is how to distinguish between what is true and what is wrong. In this paper, we propose a computational method to find answers from web search results. Intuitively, if a value is provided by many accurate web sites, it is very likely that the value is correct. We adopt an iterative method to compute the accuracy of web sites and the confidence of values. Finally we find the most likely correct answers and return to the users for further choice.