The Turn: Integration of Information Seeking and Retrieval in Context (The Information Retrieval Series)
The SMART Retrieval System—Experiments in Automatic Document Processing
The SMART Retrieval System—Experiments in Automatic Document Processing
Focused access to sparsely and densely relevant documents
Proceedings of the 33rd international ACM SIGIR conference on Research and development in information retrieval
DTD based costs for tree-edit distance in structured information retrieval
ECIR'13 Proceedings of the 35th European conference on Advances in Information Retrieval
WANTED: focused queries for focused retrieval
ECIR'13 Proceedings of the 35th European conference on Advances in Information Retrieval
Estimating structural relevance of XML elements through language model
Proceedings of the 10th Conference on Open Research Areas in Information Retrieval
Hi-index | 0.00 |
This paper describes our participation at INEX 2010. We participated in two different tracks: ad-hoc and data-centric. We first propose a classification of INEX topics and analyze several characteristics of the relevance assessments from INEX 2009 for each of the topic classes. The goal of our study is to investigate whether there are differences in relevance judgements between topic classes in order to use this information at retrieval time. We also present the experiments we performed on the INEX 2010 data. In the ad-hoc track we study the performance effects of changing the article order (fetching phase) while in the data-centric track we experiment with the use of different indices and retrievable element types. Our main finding is that indexing uniquely movie documents leads to much better performance than indexing the complete collection.