Previous studies investigating task based search often take the form of lab studies or large scale log analysis. In lab studies, users typically perform a designed task under a controlled environment, which may not reflect their natural behaviour. While log analysis allows the observation of users' natural search behaviour, often strong assumptions need to be made in order to associate the unobserved underlying user tasks with log signals. We describe a field study during which we log participants' daily search and browsing activities for 5 days, and users are asked to self-annotate their search logs with the tasks they conducted as well as to describe the task characteristics according to a conceptual task classification scheme. This provides us with a more realistic and comprehensive view on how user tasks are associated with logged interactions than seen in previous log- or lab-based studies; and allows us to explore the complex interactions between task characteristics and their presence in naturalistic tasks which has not been studied previously. We find a higher number of queries, longer timespan, as well as more task switches than reported in previous log based studies; and 41% of our tasks are zero-query tasks implying that large amounts of user task activities remain unobserved when only focused on query logs. Further, tasks sharing similar descriptions can vary greatly in their characteristics, suggesting that when supporting users with their tasks, it is important to know not only the task they are engaged with but also the context of the user in the task. Copyright is held by the owner/author(s). Publication rights licensed to ACM.

, ,
Behavior-aware Search Evaluation for Information Retrieval
ACM SIGIR Conference on Information Interaction and Retrieval
Centrum Wiskunde & Informatica, Amsterdam (CWI), The Netherlands

He, J, & Yilmaz, E. (2017). User behaviour and task characteristics: A field study of daily information behaviour. In CHIIR 2017 - Proceedings of the 2017 Conference Human Information Interaction and Retrieval (pp. 67–76). doi:10.1145/3020165.3020188