The 35th ACM SIGIR Conference was held in Portland, Oregon, USA. Every three years the Gerard Salton Award is handed out for long lasting achievements in the field of information retrieval. This year, Prof. Dr. Norbert Fuhr was awarded with the prize.
My personal highlights from the accepted full paper presentations include:
- Van Dang and Bruce Croft – Diversity by Proportionality: An Election-based Approach to Search Result Diversification The authors propose a voting algorithm that takes into account the underlying aspect proportions of diverse queries.
- Ryen White and Eric Horvitz – Studies of the Onset and Persistence of Medical Concerns in Search Logs. The authors investigate medical web search sessions as expressed in search engine log files. According to their analysis, 80% of their users showcase medical search sessions over a period of several months. Most notably, sympton-driven searches were found to precede concrete conditions by significant time.
- Patrick Pantel et al. – Social Annotations on the Search Results Page: Utility and Prediction Modeling. This work uses social network annotations such as likes, dislikes and expressions of expertise to augment search engine result pages. The evaluation is based on a large-scale simulated social network.
- Eugene Agichtein et al. – Search, Interrupted: Understanding and Predicting Search Task Continuation . Some search tasks continue across session boundaries and resurface across extended stretches of time. The authors identify properties of continuing search task in order to predict whether a given task will end with the current session or resurface at a later point in time.
- Brent Hecht et al. – Explanatory Semantic Relatedness and Explicit Spatialization for Exploratory Search. The authors presented Atlasify, a system for exploring the relatedness of topical and spatial domain of search queries.
- Yu-Heng Lei et al. – Where Is Who: Large-Scale Photo Retrieval by Facial Attributes and Canvas Layout . A sketch-based image retrieval system powered by number and position of people in the images. Face recognition further helps to refine queries by specifying who to search for.
- Mehdi Hosseini et al. – An Uncertainty-aware Query Selection Model for Evaluation of IR Systems. The authors propose a query selection framework to identify the most effective subset of queries for the formation of evaluation corpora.