Some highlights from the oral paper presentations:
- Ryen White – Beliefs and Biases in Web Search. (Best paper) The author investigates the importance of several well-known cognitive biases in Web Search. Most notably, searchers were found to often stick to their original beliefs even if confronted with unanimously refuting evidence.
- Mikhail Ageev et al. – Improving Search Result Summaries By Using Searcher Behavior Data. The authors employ eye gaze and cursor movement tracking techniques to assess the quality of SERP snippets, identify snippet quality criteria and subsequently maximise those features during the generation of new snippets.
- Leif Azzopardi et al. – How Query Cost Affects Search Behavior. Following a number of economic cost vs. gain models, the authors simulate user behavior under different query formulation and result inspection costs. Their findings are compared to the outcomes of a lab-based user study.
- Pernilla Qvarfordt et al. – Looking Ahead: Query Preview in Exploratory Search. For standing, ongoing and exploratory information needs in which users revisit identical or near-identical queries, the authors explore alternative interfaces that highlight which documents had been previously seen and which ones have been modified since the last visit.
- Ahmed Hassan et al. – Toward Self-Correcting Search Engines: Using Underperforming Queries to Improve Search. The authors propose explicit models of searcher satisfaction and search engine switching which can be used as quality criteria during the result ranking step.
- Milad Shokouhi et al. – Fighting Search Engine Amnesia: Reranking Repeated Results. 40% of all web search sessions contain high-ranked results that the user has encountered before. Information about previous interaction with these documents can be used for re-ranking of subsequent document occurrences.
- Miles Efron – Query Representation for Cross-Temporal Information Retrieval. Based on machine translation and vocabulary alignment techniques, the author presents a method for retrieving historical documents using contemporary Query terms.
- Peter Golbus et al. – A Mutual Information-based Framework for the Analysis of Information Retrieval Systems. Using different notions of mutual information, the authors investigate retrieval system evaluation to contrast different result lists against actual performance differences.