Projekte

This paper presents INSYDER, a contentbased visual-information-seeking system for the Web. The Web can be seen as one huge digital library offering a variety of very useful information for business analysts. INSYDER addresses these possibilities and offers powerful retrieval and visualisation functionalities. The main focus during the development was on the usability of the system. Therefore, a variety of well-established visualisation components were employed to support the user during the information-seeking process (e.g. visual query, result table, bar graph, segment view with tile bars, and scatterplot). Also, the retrieval aspects were developed with the goal of increasing the usability of the system (e.g. natural language search, content-based classification, relevance feedback). Extensive evaluations of the retrieval performance and the usability of the visualisation were conducted. The results of these evaluations offered many helpful insights into developing a new visual-informationseeking system called VisMeB.