What is Information Fostering?

  • Providing proactive support for seeking, sharing, and making sense of information.
  • Suggesting questions to ask and answering the questions you should be asking.
  • Giving you guidance for the future that you may not know exists.
  

Information Fostering is currently being funded through the National Science Foundation (NSF) grant III-1717488, awarded to Dr. Chirag Shah.

Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.
 
Dr. Shah received Amazon Research Award for project "Addressing Cold Start Problem in Personalization and Recommendation Using Proactive Information Retrieval" (February 2019-January 2020; $100,000). This will complement and help expand the work being done under the NSF-funded project.
 
Last update: March 2020

Research Outcomes

  • A new framework for proactive recommendations: Reported in a Perspective Paper at ACM CHIIR 2018.
  • Task Model: Reported in papers at ACM SIGIR 2018, ACM ICTIR 2018, and ACM SIGIR 2019.
  • Problem-Help Model: Published in the Journal of Information Processing and Management (IP&M).
  • Coagmento (a free, open-source tool) for conducting interactive IR and user studies: Presented at ACM CHIIR 2019.

Current Research Activities

  • We are doing more analysis using the data generated by a user study we recently completed. This work has already generated interesting results for creating a Problem-Help Model, which is under review.
  • Exploring reinforcement learning as a method to connect Task Model, Intention Model, and Problem-Help Model to recommendations -- first, reactive, and eventually, proactive.

Past Research Activities

  1. Information seeking barriers, failures, and solutions. In an effort to understand where people fail in information seeking situations and how to best help them, even when they don't explicitly ask for help, we collected data from a qualitative survey involving 208 real-life examples of information seeking failures. In addition, 10 semi-structured interviews with 10 different participants were analyzed using various theoretical frameworks of tasks, strategies, and barriers.
  2. Generative query recommendations. Sometimes, to help an information seeker, we need to go beyond just incremental changes to their existing expression of information need. This is where we have developed new techniques for generating queries that allow a searcher to make a leap rather than just a step. Given a user’s past queries, we model their past queries as a set of topics and recommended the least explored topic that was still relevant to their search, according to a threshold criterion. We then generate multi-word query terms from this topic, using a skipgram model that could create coherent phrases.
  3. Strategy recommendations. We have conducted research in recommending strategies (a sequence of steps involving queries, documents, and relevant information) for exploratory search tasks. We were able to accurately predict the performance of a user a few steps ahead within their search, and moreover could offer effective recommendations. The recommendations offered became more effective as the recommendations were given later in the search process. They moreover greatly enhanced user performance, according to quantitative simulations and qualitative judgments of the simulations and user performance.
  4. Collaborator recommendation. Some times, to help someone in their information seeking process, we should recommend a person to collaborate with. Using 120 participants in 60 pairs working on an exploratory search task, we built a model for finding suitable collaborators for an individual during different stages of his/her search. This model was then tested on real-life data obtained from Bing search logs and Internet Explorer browser logs. The data consisted of 8,969 search sessions by 8,051 users who were working on an exploratory search task during the overlapping timeframe.

Dissertations

  • Liu, J. (2020). A State-Based Approach to Supporting Users in Complex Search Tasks. Information Science, Rutgers University.
  • Mitsui, M. (2018). Adopting a Graphical Perspective in Interactive Information Retrieval Research. Computer Science, Rutgers University.

Publications

    2023

  1. Sarkar, S. & Shah, C. (2023). A Synthetic Search Session Generator for Task-Aware Information Seeking and Retrieval. Demo in Proceedings of ACM Web Search and Data Mining (WSDM) Conference. Singapore. February 27-March 2, 2023.

    2. 2022

  2. Liu, J. & Shah, C. (2022). Leveraging User Interaction Signals and Task State Information in Adaptively Optimizing Usefulness-Oriented Search Sessions. Proceedings of ACM/IEEE-CS Joint Conference on Digital Libraries (JCDL). Cologne, Germany.

    3. 2021

  3. Wang, Y. & Shah, C. (2021). Authentic vs. Synthetic: An Investigation of the Influences of Study Settings and Task Configurations on Search Behaviors. Journal of the Association for Information Science and Technology (JASIST), 73 (3), p. 362-375.
  4. Shah, C., & White, R. W. (2021). Task Intelligence for Search and Recommendation. Synthesis Lectures on Synthesis Lectures on Information Concepts, Retrieval, and Services. Morgan & Claypool Publishers.
  5. Shah, C., & White, R. W. (2021, March 14-19). Bridging task expressions and search queries. Proceedings of ACM Conference on Human Information Interaction and Retrieval (CHIIR), 319-323.

    6. 2020

  6. Mehrotra, R., Shah, C., & Carterette, B. (2020). Investigating Listeners’ Responses to Divergent Recommendations. Proceedings of Fourteenth ACM Conference on Recommender Systems (RecSys '20). Association for Computing Machinery. September 22-26, 2020. New York, NY, USA, 692–696.
  7. Liu, J., Sarkar, S., & Shah, C. (2020). Identifying and Predicting the States of Complex Search Tasks. Proceedings of ACM Conference on Human Information Interaction and Retrieval (CHIIR). March 14-18. 2020. Vancouver, Canada.

    8. 2019

  8. Liu, J., & Shah, C. (2019). Proactive identification of query failure. Proceedings of the Association for Information Science and Technology, 56(1), 176-185. [Best Paper Award]
  9. Sarkar, S., Mitsui, M., Liu, J., & Shah, C. (2019). Implicit Information Need as Explicit Problems, Help, and Behavioral Signals. Journal of Information Processing & Management (IP&M).
  10. Mitsui, M., & Shah, C. (2019). Bridging Gaps: Predicting User and Task Characteristics from Partial User Information. In Proceedings of ACM SIGIR 2018 Conference. 10 pp. July 21-25, 2019. Paris, France.
  11. Liu, J., & Shah, C. (2019). Interactive IR User Study Design, Evaluation, and Reporting. Synthesis Lectures on Information Concepts, Retrieval, and Services. Series Edited by Gary Marchionini. Morgan & Claypool Publishers. (93 pages) ISBN: 9781681735795
  12. Soltani, D., Mitsui, M., & Shah, C. (2019). Coagmento v3.0: Rapid Prototyping of Web Search Experiments. Demo in Proceedings of the 2019 Conference on Human Information Interaction and Retrieval (CHIIR '19). ACM, New York, NY, USA, 367-371. DOI: https://doi.org/10.1145/3295750.3298917

    13. 2018

  13. Mitsui, M., & Shah, C. (2018). The Broad View of Task Type Using Path Modeling. In Proceedings of ACM International Conference on Theory of Information Retrieval (ICTIR). September 14-17, 2018. Tianjin, China.
  14. Mitsui, M., Liu, J., & Shah, C. (2018). How Much is Too Much? Whole Session vs. First Query Behaviors in Task Prediction. In Proceedings of ACM SIGIR 2018 Conference. 4 pp. July 8-12, 2018. Ann Arbor, MI.
  15. Shah, C. (2018). Information Fostering - Being Proactive with Information Seeking. In Proceedings of ACM Conference on Human Information Interaction and Retrieval (CHIIR). March 11-15, 2018. New Brunswick, NJ.

Talks

    2019

  1. Investigating Tasks and Taskers with Studies of Searching Online and in the Wild. DUB Seminar. University of Washington. November 13, 2019.
  2. Going Beyond Retrieval: Task Fulfillment and Fostering with Information. RMIT University. Melbourne, Australia. October 22, 2019.
  3. Going from 'What' and 'How' to 'Who' and 'Why': Task Analysis in Online and Physical Contexts. University College London, UK. March 15, 2019.

    4. 2018

  4. Information Fostering - Being Proactive in Retrieval and Recommendation. Amazon, Seattle. Nov. 14, 2018.
  5. Information Fostering: The Art and Science of Providing Answers without Questions. Spotify, NYC. April 30, 2018.
  6. Information Fostering: Being Proactive in Information Seeking and Retrieval. ACM Conference on Human Information Interaction and Retrieval (CHIIR). March 13, 2018.

    7. 2017

  7. Information Fostering: Being Proactive in Information Seeking. National Institute of Informatics (NII). Tokyo, Japan. July 24, 2017.
  8. Information Fostering: Being Proactive in Information Seeking. Microsoft Research. Redmond, Washington. March 29, 2017.

    9. 2016

  9. Information Fostering: Proactively Complementing Information Seeking. GESIS in Cologne, Germany. June 30, 2016.
  10. Information Fostering: Being Proactive in Information Seeking. University of Texas at Austin. March 4, 2016.
  11. Information Fostering: Proactively Complementing Information Seeking. Guest lecture in PhD 610 (Seminar in Information Science). Rutgers University. October 13, 2016.
  12. Information Fostering: Proactively Supporting Information Seeking. University of Tampere, Finland. October 21, 2016.

    13. 2015

  13. Minority Report for Information Seeking: Fixing Things Before They Break in Search. University Technology at Sydney, Australia. October 21, 2015.