Trust is a necessary concept to realize the SemanticWeb. But how can we build a "Web of Trust"? We first argue that a small "Web of Trust" for each community is very essential to realize a huge "Web of Trust." Then, we focus on an academic community as a microcosm of a "Web of Trust" and show a web mining approach to generate a social network automatically. Each edge is added using the number of retrieved pages by a search engine which includes both persons' names. Moreover, each edge is given a label, such as "coauthors" or "members of the same project," by applying classification rules to the page content. The relation of persons such as "coauthor" or "same laboratory" can be described by an RDF format. Finally, using the social network, we calculate authoritativeness of a node as a social trust and an individual trust. National Institute of Advance Industrial Science and Technology (AIST), Japan email: email@example.com.