Our SIGIR 2014 paper “A syntax-aware re-ranker for microblog retrieval” by Aliaksei Severyn, Alessandro Moschitti, Manos Tsagkias, Richard Berendsen and Maarten de Rijke is online now. We tackle the problem of improving microblog retrieval algorithms by proposing a robust structural representation of (query, tweet) pairs. We employ these structures in a principled kernel learning framework that automatically…
Category: Publications
SIGIR 2014 paper on evaluating intuitiveness of vertical-aware click models online
Our SIGIR paper “Evaluating Intuitiveness of Vertical-Aware Click Models” by Aleksandr Chuklin, Ke Zhou, Anne Schuth, Floor Sietsma and Maarten de Rijke is available online now. Modeling user behavior on a search engine result page is important for understanding the users and supporting simulation experiments. As result pages become more complex, click models evolve as well in…
SIGIR 2014 paper on personalized document re-ranking online
Our SIGIR 2014 paper “Personalized Document Re-ranking Based on Bayesian Probabilistic Matrix Factorization” by Fei Cai, Shangsong Liang and Maarten de Rijke is available online now. A query considered in isolation provides limited information about the searcher’s interest. Previous work has considered various types of user behavior, e.g., clicks and dwell time, to obtain a better understanding…
SIGIR 2014 paper on email recipient recommendation online
Our SIGIR 2014 paper “Recipient Recommendation in Enterprises using Communication Graphs and Email Content” by David Graus, David van Dijk, Manos Tsagkias, Wouter Weerkamp and Maarten de Rijke is available online now. We address the task of recipient recommendation for emailing in enterprises. We propose an intuitive and elegant way of modeling the task of recipient recommendation,…
ECIR 2014 proceedings online
With 10 days to go to ECIR 2014, the ECIR 2014 proceedings are online now. You can find information about it at the Springer web site or access the online version directly at this page. Now playing: Portico Quartet — Rubidium
ECIR 2014 paper on blending vertical and web results online
“Blending Vertical and Web results: A Case Study using Video Intent” by Damien Lefortier, Pavel Serdyukov, Fedor Romanenko and Maarten de Rijke is available online now. Modern search engines aggregate results from specialized verticals into the Web search results. We study a setting where vertical and Web results are blended into a single result list, a setting…
ECIR 2014 paper on query-dependent contextualization of streaming data online
“Query-dependent contextualization of streaming data” by Nikos Voskarides, Daan Odijk, Manos Tsagkias, Wouter Weerkamp and Maarten de Rijke is available online. We propose a method for linking entities in a stream of short textual documents that takes into account context both inside a document and inside the history of documents seen so far. Our method uses a…
ECIR 2014 paper on cluster-based fusion for microblog search online
“The impact of semantic document expansion on cluster-based fusion for microblog search” by Shangsong Liang, Zhaochun Ren and Maarten de Rijke is available online now. Searching microblog posts, with their limited length and creative language usage, is challenging. We frame the microblog search problem as a data fusion problem. We examine the effectiveness of a recent cluster-based…
ECIR 2014 paper on predicting new concepts in social streams online
“Generating Pseudo-ground Truth for Predicting New Concepts in Social Streams” by David Graus, Manos Tsagkias, Lars Buitinck and Maarten de Rijke is available online now. The manual curation of knowledge bases is a bottleneck in fast paced domains where new concepts constantly emerge. Identification of nascent concepts is important for improving early entity linking, content interpretation, and…
ECIR 2014 paper on click-based recommender evaluation online
“Effects of Position Bias on Click-Based Recommender Evaluation” by Katja Hofmann, Anne Schuth, Alejandro Bellogin and Maarten de Rijke is available online now. Measuring the quality of recommendations produced by a recommender system (RS) is challenging. Labels used for evaluation are typically obtained from users of a RS, by asking for explicit feedback, or inferring labels from…