I-Corps: Translation Potential of an Online Platform for Collaborative and Personalized Information Discovery

The broader impact of this I-Corps project is the development of a general software tool to improve online information management and discovery. The goal is to help helps communities of users organize saved information around the questions that the community members ask and where they ask it. As a result, the information needs of community members can be anticipated and resolved, which minimizes user effort and mitigates many of the problems of finding online information. This technology has the potential to be applied in any setting where groups of individuals share common contexts and goals. For example, in an educational setting, students may use the tool to efficiently ask and answer questions on course-based material, such as lectures, assignments, and readings. In an enterprise setting, knowledge workers can leverage the tool to reduce the reliance on tacit knowledge, thus making the organization more efficient. This technology may help mitigate inefficient online information discovery by using saved information to guide the learning process. This I-Corps project utilizes experiential learning coupled with a first-hand investigation of the industry ecosystem to assess the translation potential of a collaborative software platform with the goal of helping mitigate inefficient online information discovery by using saved information to guide the learning process. The solution is based on the use of large language models and search algorithms. It uses generative large language models to help predict the questions that a user will have in a given context, a novel framework for using the historical interactions of community members to guide future community members in similar contexts, and an actively developed software platform that supports these technologies via various interfaces, including both a website and a browser extension. In addition, the platform may provide an ecosystem for researchers to learn more about many previously understudied problems in information retrieval, such as proactive search, through the de-identified data collected by the platform. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.