Clear and precise queries are a necessity when searching very large document collections, especially when query-based retrieval is the only means of exploration. We propose system-mediated information access as a solution for users ’ well-documented inability to formulate good queries. Our approach is based on two main assumptions: first, on the ability of document clustering to reveal the topical, semantic structure of a problem domain represented by a specialized “source collection,” and, second, on the capacity of statistical language models to convey content. Taking the role of the human mediator or intermediary searcher, a mediation system interacts with the user and supports her exploration of a relatively small source collection, chosen to be representative for the problem domain. Based on the

The utility of an enterprise search system is determined by three key players: the information retrieval (IR) system (the search engine), the enterprise users, and the service provider who delivers the tailored IR service to its designated enterprise users. Currently, evaluations of enterprise search have been focused largely on the IR system effectiveness and efficiency, only a relatively small amount of effort on the user’s involvement, and hardly any effort on the service provider’s role. This paper will investigate the role of the service provider. We propose a method that evaluates the cost and benefit for a service provider of using a mediated search engine – in particular, where domain experts intervene on the ranking of the search results from a search engine. We test our cost and benefit evaluation method in a case study and conduct user experiments to demonstrate it. Our study shows that: 1) by making use of domain experts ’ relevance assessments in search result ranking, the precision and the discount cumulated gain of ranked lists have been improved significantly (144 % and 40 % respectively); 2) the service provider gains substantial return on investment and higher search success rate by investing in domain experts ’ relevance assessments; and 3) the cost and benefit evaluation also indicates the type of queries to be selected from a query log for evaluating an enterprise search engine.