A Real-Time Eye Tracking Based Query Expansion Approach via Latent Topic Modeling

Abstract

Formulating and reformulating reliable textual queries have been recognized as a challenging task in Information Retrieval (IR), even for experienced users. Most existing query expansion methods, especially those based on implicit relevance feedback, utilize the user's historical interaction data, such as clicks, scrolling and viewing time on documents, to derive a refined query model. It is further expected that the user's search experience would be largely improved if we could dig out user's latent query intention, in real-time, by capturing the user's current interaction at the term level directly. In this paper, we propose a real-time eye tracking based query expansion method, which is able to: (1) automatically capture the terms that the user is viewing by utilizing eye tracking techniques; (2) derive the user's latent intent based on the eye tracking terms and by using the Latent Dirichlet Allocation (LDA) approach. A systematic user study has been carried out and the experimental results demonstrate the effectiveness of our proposed methods.

Implicit Estimation of Paragraph Relevance From Eye Movements

“…They use relevance estimates to identify recently read paragraphs that are relevant to the user and, eventually, to reformulate the search query. Chen et al (2015) presented a query expansion method based on eye tracking and topic modeling. They identified fixated terms and modeled the user's latent intent using the Latent Dirichlet Allocation (LDA) for topic modeling.…”

“…However, eye movements highly depend on the user characteristics, the task at hand, and the content visualization (Buchanan et al, 2017). Related approaches use eye tracking to infer the perceived relevance of text documents with respect to previously shown trigger questions (Salojarvi et al, 2003(Salojarvi et al, , 2004(Salojarvi et al, , 2005aBuscher et al, 2008a;Loboda et al, 2011;Gwizdka, 2014a;Bhattacharya et al, 2020a,b), and to extend (Buscher et al, 2008b;Chen et al, 2015) or generate search queries (Hardoon et al, 2007;Ajanki et al, 2009). A common disadvantage of approaches for gaze-based relevance estimation is that they are tested using documents with constrained layouts and topics such as single sentences (Salojarvi et al, 2003(Salojarvi et al, , 2004(Salojarvi et al, , 2005a or short news articles that fit on the screen at once (Buscher et al, 2008a;Loboda et al, 2011;Gwizdka, 2014a;Bhattacharya et al, 2020a,b).…”

Modeling multiple interactions with a Markov random field in query expansion for session search

“…For example, Buscher et al used eye‐tracking data to keep track of document parts that the user reads, and then the information at the subdocument level is used as implicit feedback for query expansion and document re‐ranking . More recently, Chen et al have proposed a query expansion model based on the real‐time reading content captured by an eye tracker.…”