Flank, S. (1998), A layered approach to nlp-based information retrieval, in `Proceedings of the 36th ACL and 17th COLING', Morgan Kaufmann, pp. 397--403.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....But we could abstract the Ypa to an IR task involving short queries for the access of short documents which is quite different to classical IR tasks like those considered at TREC [17] From this point of view there are at least two approaches that should be mentioned here. Publishers Depot [5] is a commercial product that allows the retrieval of pictures based on the corresponding caption. 2 That means the doc 2 http: www.picturequest.com uments are very short compared to standard IR tasks. Some points are worth mentioning here as we are concerned with similar problems. First of ....

....algorithm has to infer that car washing is some sort of automobile cleaning. Such a knowledge base has been applied to precision content retrieval knowing that standard IR approaches are not very effective for short queries [2] Our approach can finally be placed somewhere between the one used in [5] and [20] 4 The Backend 4.1 Input Data The Backend construction process takes the raw data (the so called YP printing tape) and creates a database that retains as much information and relations as possible for the online enquiry system. This was described in detail in [4] The input data is ....

[Article contains additional citation context not shown here]

Flank, S. A layered approach to NLP-based Information Retrieval. In Proceedings of the 36 th ACL and the 17 th COLING Conferences (Montreal, 1998), pp. 397-- 403.

....by hand and took advantage of the researchers a priori knowledge of the Reuters 22173 domain. No change of representation was involved, and the use of WordNet was limited to the synonymy relation only. Some researches in the IR community have also applied WordNet to retrieval tasks. Sharon Flank [FLA98] recently reported good results using many of the WordNet semantic relations, including hypernymy, to expand short descriptions of pictures for retrieval. Ellen Voorhees looked at the possibility of using WordNet for document indexing [VOO93] and query expansion [VOO94] In these cases, the ....

Sharon Flank. A Layered Approach to NLP-Based Information Retrieval. In Proc. COLING-ACL'98. August, 1998. 397-403.

....This entire process was done by hand and took advantage of the researchers a priori knowledge of the Reuters domain. No change of representation was involved, and the use of WordNet was limited to the synonymy relation only. Some researches have also applied WordNet to retrieval tasks. Sharon Flank [1998] reported good results using many of the WordNet semantic relations, including hypernymy, to expand short descriptions of pictures for retrieval. Ellen Voorhees looked at the possibility of using WordNet for document indexing [Voorhees, 1993] and query expansion [Voorhees, 1994] The latter work ....

Flank, Sharon. 1998. A layered approach to NLP-based information retrieval. COLING-ACL'98. 397-403.

....e.g. Minox dealer in Colchester or Minox Colchester. Ontologies and customised versions of existing language resources like WordNet [24] are being successfully em 3 http: www.northernlight.com ployed to search product catalogues and other document collections held in relational databases [14, 12]. Part of that research is the actual construction of ontologies [9] The cost to create the resources can be enormous and it is difficult to apply these solutions to other domains where the document structure is not known in advance. Quite a different way of dealing with the semistructured ....

S. Flank. A layered approach to NLP-based Information Retrieval. In Proceedings of the 36 th ACL and the 17 th COLING Conferences, pages 397--403, Montreal, 1998.

....process was done by hand and took advantage of the researchers a priori knowledge of the Reuters domain. No change of representation was involved, and the use of WordNet was limited to the synonymy relation only. Some researches have also applied WordNet to information retrieval tasks. Sharon Flank [1998] reported good results using many of the WordNet semantic relations, including hypernymy, to expand short descriptions of pictures for retrieval. Ellen Voorhees looked at the possibility of using WordNet for document indexing [Voorhees, 1993] and query expansion [Voorhees, 1994] The latter work ....

Flank, Sharon. 1998. A Layered Approach to NLP-Based Information Retrieval. COLINGACL '98. 397-403.

....documents while ignoring irrelevant information. To find relevant information, NLP systems apply such techniques as filtering, part of speech tagging, parsing, lexical semantic tagging, building relationships among phrasal and sentential elements, and producing a grammatically coherent framework [19,20,24]. In the NLP approach, natural language is the dominant guide; in our approach, an ontology expressed as a conceptual model is the dominant guide. Unlike many NLP systems, our work does not attempt to extract deeplevel understanding and our work does not depend upon complete sentences. Our ....

Flank, S.: A layered approach to NLP-based information retrieval. Proc. COLING'98 (1998) 397--403

No context found.

Flank, S. (1998), A layered approach to nlp-based information retrieval, in `Proceedings of the 36th ACL and 17th COLING', Morgan Kaufmann, pp. 397--403.

No context found.

S. Flank. A layered approach to NLP-based information retrieval. In Proceedings of the 36th ACL and the 17th COLING Conferences, pages 397--403, Montreal, 1998.

No context found.

S. Flank. A layered approach to NLP-based information retrieval. In Proceedings of COLING'98, pages 397--403, 1998.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC