This thesis combines aspects from two approaches to informationaccess, information filtering and information retrieval, in an effortto improve the signal to noise ratio in interfaces to conversationaldata. These two ideas are blended into one system by augmenting asearch engine indexing Usenet messages with concepts and ideas fromrecommender systems theory. My aim is to achieve a situation wherethe overall result relevance is improved by exploiting the qualitiesof both approaches. Important issues in this context are obtainingratings, evaluating relevance rankings and the application of usefuluser profiles.
An architecture called NewsView has been designed as part of the workon this thesis. NewsView describes a framework for interfaces toUsenet with information retrieval and information filtering conceptsbuilt into it, as well as extensive navigational possibilities withinthe data. My aim with this framework is to provide a testbed for userinterface, information filtering and information retrieval issues,and, most importantly, combinations of the three.