Understanding and Supporting Collaborative Sensemaking in Collaborative Information Seeking.

Abstract

With the ubiquity of current information retrieval systems, users move beyond individual searching to performing complex and exploratory information seeking tasks together with collaborators for social, leisure or professional purposes. As a consequence, collaborative information seeking (CIS) systems become popular to support users for CIS tasks. These CIS systems aimed at enhancing the awareness of each others activities between collaborators but provide little support for collaborative sensemaking of the CIS task and related information together. To design tools for collaborative sensemaking, we lack an understanding of how users perform collaborative sensemaking and what support they need for collaborative sensemaking in CIS. Therefore, the aim of this thesis is to understand user strategies for collaborative sensemaking and the challenges they face in collaborative sensemaking, and to design tools to support collaborative sensemaking in CIS. In this thesis, we first present an exploratory study that investigates how users perform collaborative sensemaking, and the challenges they encountered in CIS. A follow-up study then compared the collaborative sensemaking behaviour and challenges users encountered between different CIS tasks. Through a comparative analysis, we acquired an understanding of the difference of collaborative sensemaking behaviour according to task as well as the general patterns in collaborative sensemaking behaviour and the challenges that users face. Based on the findings from our user studies, we proposed and designed a tool MakeSenseTogether, with novel topic-related features, to support collaborative sensemaking behaviour. An evaluation study of MakeSenseTogether shows that the topic-related features improved user experience of collaborative sensemaking in CIS. This thesis contributes to our understanding of collaborative sensemaking in CIS in two ways. Firstly, we gain a comprehensive understanding of the general process of collaborative sensemaking and the challenges users face. Secondly, we proposed novel topic-related features which can improve users experience in collaborative sensemaking.