Research Interest

I am a Ph.D. student at Tampere University, at the Department of Information Studies .
My supervisor is Professor Kalervo Järvelin.

Since 2005, I am a member of the IRiX (Research Group on Information Retrieval in Context). at University of Tampere, Department of Information Studies. The IRiX Research Group focuses on interactive information retrieval from a task performance perspective. The aim is to study information searching and the role of information retrieval systems in support of human actions and tasks.

In 2000, I joined the FIRE (Finnish Information Retrieval Expert Group) research group at University of Tampere, Department of Information Studies. FIRE members pursue basic and applied research on all aspects of information storage and retrieval, including bibliographic, text and multimedia databases, database design and reorganization, and document management.

Research interest:

• Collaborative Information Seeking and Retrieval
• Task-based IS&R and Contextual approaches to Information Access.
• Patent IR
• User and Usage studies, Use Cases and Evaluation.
• Digital Libraries
• Crisis Management

Research Issues

Task- and Context-based approach to information seeking and retrieval (IS&R).

Information retrieval should not be treated in isolation, rather as embedded in the information seeking process. Furthermore, within a professional setting such as the Patent domain, the information seeking and retrieval activities are embedded in work-tasks. In order to acquire a wider understanding of the processes that undergo IR, we need to adopt a broader and contextual perspective, which includes the tasks people are performing.

In short, our general goal of the project is to investigate the relationship between work-task situations and their characteristics and users information retrieval processes in order to understand the task performance process and elicit requirements for interactive information systems design. How are work-tasks related to the information seeking and retrieval process? How do types of tasks and task characteristics influence the information seeking and retrieval process?

The study has the following point of departure:

• Today, many IR systems have problems supporting both tasks and users.
• The context of the tasks is vague and not well understood
• There are relatively few studies that empirically connect and relate the work task to the information seeking and retrieval.
• The concept of task is seldom taken into account for the design of IR systems and user interface.

IR in the Patent domain.
The study is performed within a real life work setting, the Swedish Patent- and Registration Office (PRV), where patent engineers are performing real work tasks including information seeking and retrieval tasks, which involves real information needs. Both qualitative and quantitative data are collected using methods such as interviews, electronic "diaries" and focus observations. Finally, log statistics from searches are also collected.

The project is organised in collaboration with the Swedish Institute of Computer Science (SICS) and the Swedish Patent- and Registration Office (PRV) in Sweden, Stockholm and University of Tampere, Department of Information Studies.

The results of the study will be presented as scientific articles/papers and finally reported in a doctoral dissertation.

Users and tasks
Users have varying preferences, information needs, tasks and goals and information seeking strategies (searching, browsing, navigation etc), at both an individual, group and domain levels. IR techniques are more and more frequently being used in complex task oriented on-line systems whose main objectives are not only to retrieve information. We need to relate the information need to a task or a set or sequence of tasks, and design tools that supports different information seeking strategies to fit in an appropriate task context. Systems today most often are not designed to be adaptive: neither to tasks nor to users. We also need to understand the changing nature of these tasks during an information seeking episode(s) and to observe what users attempt to do in their task context, i.e. discrepancies between what users really want to do and what they really do.

Information Acces Evaluation
IR evaluation must be done in real situations and with real users with real needs and tasks. Like users, different IR system have their specific characteristics and functionalities that influence how the information could be accessed. Precision and recall alone cannot provide the understanding and other metrics are needed. We need to combine qualitative and quantitative evaluation (both system- and user-based) methods in a more dynamic evaluation framework, incorporating e.g. HCI methods (such as usability studies).

Implications for design of information access
The evaluation should reveal important implications and aspects of how the system (and interface) could be improved as well as knowledge on how users interact with IR systems in terms of the collaborative support between the user and the IR system, the tranformation of ill-formulated information needs, and to relate different strategies to different IR techniques.