Tarragon Consulting at TREC 2017
نویسنده
چکیده
Tarragon Consulting Corporation (henceforth Tarragon) contributed two runs to the new Common Core track. Both were manual runs using the NIST judged topics. Both used Solr as the base search engine with the queries semi-automatically constructed from the Topic descriptions and augmented with information from Wordnet and Wikipedia. Results are generally below the published median scores but for several topics our results are very competitive. And we did contribute a significant number of “unique relevant” documents to the judged pool.
منابع مشابه
Knowledge-Based Access to the Bio-Medical Literature, Ontologically-Grounded Experiments for the TREC 2003 Genomics Track
The Tarragon Consulting team participated in the primary task of the TREC 2003 Genomics Track. We used a combination of knowledge-engineering and corpus analysis to construct semantic models of the interactions between genes/proteins and other biological entities in the organism, and then used automatic methods to convert these models into evidential queries that could be executed by the K2 sea...
متن کاملInformation Needs and Automatic Queries
Tarragon Consulting Corporation participated in the adhoc retrieval task of the TREC 2004 Genomics Track. We used a standard deployment of the K2 search engine from Verity, Inc. in which we exploited the free-text query parser to interpret the information need statements provided in the task. The primary goal of our participation was to establish a performance baseline using “of-the-shelf” tool...
متن کاملDocument Retrieval Using The MPS Information Server (A Report on the TREC-7 Experiment)
This paper summarizes the results of the work conducted by FS Consulting, Inc. as part of the Seventh Text Retrieval Experiment Conference (TREC-7). We participated in two components of TREC: (1) Category C, running the Ad Hoc experiments, producing two sets of official results (fsclt7m and fsclt7a) and (2) the Very Large Collection (VLC) track, running the entire experiment, producing the requ...
متن کاملDocument Retrieval Using The MPS Information Server (A Report on the TREC-4 Experiment)
This paper summarizes the results of the experiments conducted by FS Consulting, Inc. as part of the Six Text Retrieval Experiment Conference (TREC-6). We participated in Category C and ran the ad-hoc experiments, producing three sets of official results (fsclt6, fsclt6r and fsclt6t), only one of which was judged (fsclt6). We also produced two sets of unofficial results as part of a database me...
متن کاملDocument Retrieval Using The MPS Information Server ( A Report on the TREC - 6 Experiment )
This paper summarizes the results of the experiments conducted by FS Consulting, Inc. as part of the Six Text Retrieval Experiment Conference (TREC-6). We participated in Category C and ran the ad-hoc experiments, producing three sets of official results (fsclt6, fsclt6r and fsclt6t), only one of which was judged (fsclt6). We also produced two sets of unofficial results as part of a database me...
متن کامل