People often have difficulty in expressing their information needs. Many times this results from a lack of clarity regarding the task at hand, or the way an information or search system works. In addition, people may not know what they do not know. The former is addressed by search systems by providing recommendations, whereas there are no good solutions for the latter problem. Even when a search system makes recommendations, they are limited to suggesting objects such as queries and documents only. They do not consider providing suggestions for strategies, people, or processes. This project will address such problems by investigating the nature of the work a person is doing, predicting the potential problems they may encounter, and providing help to overcome those problems. Such a help could be an object such as a document or a query, a strategy, or a person. This whole process is referred to as Information Fostering. Beyond creating a general-purpose recommender system, Information Fostering is an idea for providing proactive suggestions and help to information seekers. This could allow them to avoid potential problems and capture promising opportunities in search before it is too late. In order to meet these goals, the project will carry out three lab studies. Through these studies, a new system will be created. This system will be integrated in a user's Web browser to provide real-time assessment of the information seeking process, as well as recommendations for queries, documents, strategies, and people. The outcomes of this project will make it possible and easier for a user with even low information literacy to be able to leverage the power of information. Such users may use information for multiple life contexts, including healthcare (e.g. caring for a sick family member), financial well-being (e.g., deciding on an investment portfolio), work (e.g., reviewing a business proposition), and education (e.g., compiling a report).
Current systems face challenges in understanding the problems that information seekers face due to their inability to express their information needs, recognizing a potential problem during a search episode, and identifying support needed that goes beyond what a typical search system could provide. Most recommender systems try to mitigate these problems by suggesting information objects (queries, documents), disregarding a deeper understanding of the task at hand or the possibility of recommendations that involve process/strategy, people, and other forms. The project will advance our understanding of these information-seeking problems at the task level, and of when and how help could be offered to information seekers. The offered help would go beyond recommending alternative queries and documents and would include recommending search strategies. This will be done in three phases with different user studies: (1) extracting the nature of task, problems, and help to build Task Model and Problem-Help Model; (2) testing the validity of Task Model and Problem-Help Model in being able to detect tasks, problems, and help; and (3) creating an Information Fostering system and evaluating its effectiveness in various search tasks. There will be three major intellectual outcomes: (1) Task Model that detects the nature of a search task using implicit signals such as browsing behaviors; (2) Problem-Help Model that uses behavioral data and other contextual factors, including the nature of the task, to explicate possible problems and potential solutions without explicitly asking from the searchers; and (3) a general-purpose recommender system framework, called Information Fostering, that proactively creates recommendations in real time for enhancing one's information seeking process and helping one avoid potential problems or grab an opportunity before it is too late. The results from this project will be disseminated through the project website, which will include technical reports, publications, and links to datasets and open-source software developed in this project.
