Report on the TREC 2003 Experiment: Genomic and Web Searches
نویسندگان
چکیده
This year we took part in the genomic information retrieval and information extraction tasks, as well as the named page and topic distillation searches. In carrying out the last two tasks, we made use of link anchor information and document content in order to construct Web page representatives. This type of document representation uses multi-vectors to highlight the importance of both hyperlink and document content.
منابع مشابه
Report on the TREC-8 Experiment: Searching on the Web and in Distributed Collections
The Internet paradigm permits information searches to be made across wide-area networks where information is contained in web pages and/or whole document collections such as digital libraries. These new distributed information environments reveal new and challenging problems for the IR community. Consequently, in this TREC experiment we investigated two questions related to information searches...
متن کاملReport on the TREC 11 Experiment: Arabic, Named Page and Topic Distillation Searches
This year we took part in the Arabic cross-language information retrieval track (for us limited to monolingual Arabic retrieval) and also in both named page and topic distillation searches. In the last two tasks, we made use of link anchor information and document content in order to construct Web page representatives. This document representation uses multi-vectors in order to highlight the im...
متن کاملRelevance Propagation for Topic Distillation UIUC TREC 2003 Web Track Experiments
In this paper, we report our experiments on the Web Track TREC-2003. We submitted five runs for the topic distillation task. Our goal was to evaluate the standard language modeling algorithms for topic distillation, as well as to explore the impact of combining link and content information. We proposed a new general relevance propagation model for combining link and content information, and exp...
متن کاملVT at TREC-2003: The Web Track Report
This year, we participated in the Web Track in addition to the Robust Track. We submitted results on both topic distillation and home page/named page finding tasks. As our time and human resources were limited for taking two tasks simultaneously, in this task we only concentrate on testing our ranking function discovery technique, ARRANGER (Automatic Rendering of RANking functions by GEnetic pR...
متن کاملReport on the TREC 2003 Experiments Using Web Topic-Centric Link Analysis
In TREC 2003, our experiments have been concentrated only on the topic distillation task. We first simply apply the term-based technique to the .GOV web collection, and then re-rank the retrieval results using a link analysis algorithm in order to boost the retrieval precision. Our link analysis has been inspired from the original PageRank, but focused on the web topic during the iterative scor...
متن کامل