Knowledge Search and Trust-Oriented Search

Abstract

Increased usage of Web search engines in our daily lives means that the trustworthiness of searched results has become crucial. We developed several systems to help users determine the trustworthiness of Web search results. In this paper, first, we describe the extraction of Web knowledge from Web search engine. Extracted Web knowledge is useful not only for evaluation (quality, trustworthiness etc.) of search results, but also for improving conventional Web search, such as supporting user's recall of vocabulary (query keywords) and re-ranking of search results. The proposed extraction mechanism is based on the ideas of the usage of the "structural term cooccurrence relationships" and "aggregating social annotation information", which are obtained by accessing to conventional Web search engines. We also describe our service development platform for extracting knowledge from Web, which wraps accesses to Web search engines as virtual relational database tables.