Home » HIIT-wide Focus Area: Interactive Intent Modelling for Exploratory Search
HIIT-wide Focus Area: Interactive Intent Modelling for Exploratory Search
The talk discusses the efforts undertaken in one of the two HIIT-Wide Focus Areas: Augmented Research. It is an interdisciplinary research initiative in which researchers at HIIT combine their expertise in Human-Computer Interaction, Machine Learning, Information Retrieval, and Ubiquitous Computing with a target to augment scientific information access using computation. The first part of the talk will discuss the vision of the HIIT-Wide Focus Area Research Initiative and give background on how to benefit and collaborate wit other researchers through the focus area. We have collected one of the largest databases of scientific literature and related data in the world that can be utilized in research. We aim pushing research methods and results achieved in the focus area towards systems with real-world impact.
The second part of the talk, I will present an example of practical work conducted as a part of the focus area: SciNet, an information access system for scientific literature that interactively directs users in their exploratory search tasks by predicting their search intents from interaction data. The system combines methods of optimized interactive data visualization, reinforcement learning based user modeling, and active relevance feedback for information retrieval. I will present the computational methods behind the system and evidence of effectiveness of our approach from user studies undertaken to evaluate the performance of the methods in real-world information seeking tasks.
All HIIT researchers are encouraged to join and learn what we are up to and how the work of individual researchers could benefit from the focus area initiative.
About the speaker:
Dr. Tuukka Ruotsalo is a post-doctoral researcher at the Helsinki Institute for Information Technology (HIIT). His research interests are in the intersection of Information retrieval, Human-Computer Interaction, and User Modeling. He obtained his doctoral degree at Aalto University in 2010. Before joining HIIT, he was a post-doc at UC Berkeley (2010-2011) .