J. Chu-Carroll and B. Carpenter, "Vector-based natural language call routing," Computational Linguistics, vol. 25, no. 3, pp. 361--388, 1999. Home/Search Document Not in Database Summary Related Articles Check |
This paper is cited in the following contexts:
Improving Latent Semantic Indexing Based Classifier With.. - Li, Chou (2002) (Correct)
....(synonym) and the literal terms in a query may not match those of a relevant document. Latent Semantics Indexing (LSI) is a powerful approach and successfully applied to information retrieval [1] natural language understanding and, particularly, to the problem of natural language call routing [2,3,4]. LSI is a vector space based approach. It assumes that there is some underlying or latent structure in the usage of terms in the document and query, which is obscured by the variability of the language and the choice of terms. It represents the terms (features) and documents (categories) as ....
....many times more rows (terms) than the number of columns (documents) The accuracy of a LSI classifier is therefore subject to the selection of terms. In addition to accuracy, term selection for large data set can significantly improves the runtime performance. In the previous LSI based approach [2,3], terms are selected based on their occurrence statistics (or frequencies) in the training data. Terms with occurrence less than a pre set threshold are thrown away. Terms selected or discarded in this process may or may not be salient. It relies on other language processing resources (e.g. stop ....
[Article contains additional citation context not shown here]
Chu-Carrol, J. and Carpenter B., "Vector-Based Natural Language Call Routing", Computational Linguistics, 25(3):361--389, 1999.
An Information Theoretic Approach for Using Word Cluster.. - Li, Liu, Chou (2003) (Correct)
.... ways to express a given concept (synonym) and the literal terms in a query may not match those of a relevant destination (document) This leads to the study and application of various natural language understanding and information retrieval techniques in NLCR, such as latent semantic indexing [2,3,4]. In natural language processing, word term classes (or clusters) are formed by clustering word terms that have some common properties or similar semantic meanings. They are regarded as more robust than word terms, because word class clustering process can be viewed as a mapping of the surface ....
....Joint word term and word class based LSI algorithm In this subsection, we describe the approach and implementation of an IG enhanced joint word term and word class based LSI classifier. The focus is on the joint word terms and word class IG extension part in the proposed approach and we refer to [2,3,4] for other details of LSI based classifier. The training corpus for LSI based classifier is a collection of documents with corresponding categories labeled. It is usually first processed by a linguistic analysis module to convert words in the document into a sequence of raw terms. This module is ....
Chu-Carrol, J. and Carpenter B., "Vector-Based Natural Language Call Routing", Computational Linguistics, 25(3):361--389, 1999.
Simplifying Design Specification For Automatic Training Of.. - Natural Language Call (Correct)
No context found.
J. Chu-Carroll and B. Carpenter, "Vector-based natural language call routing," Computational Linguistics, vol. 25, no. 3, pp. 361--388, 1999.
A Study of Two Different Machine Learning Approaches to Dialogue .. - Sahlgren (2002) (Correct)
No context found.
Chu-Carroll, J. and Carpenter, B. 1999. \Vector-based Natural Language Call Routing." Computational Linguistics, 25(3): 361-388.
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