Salton, G. and Buckley, C. (1989). On the automatic generation of content links in hypertext. Technical Report TR89--993, Department of Computer Science, Cornell University.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....document into hypertext, only structural links are identified. Rada [Rada, 1992] suggests the usage of semantic nets as an intermediate form that a textbook is placed in, before its transformation into a hypertext which includes content links as well as structural links. Salton et al. [Salton and Buckley, 1989, Salton et al. 1993, Salton et al. 1994a] use the information provided by the computation of the similarity between fragments of documents in order to identify content links. Smeaton et al. in [Smeaton and Morrissey, 1995] use also the similarity between document fragments but they selectively ....

Salton, G. and Buckley, C. (1989). Automatic generation of content links for hypertext. Technical Report TR89-993, Department of Computer Science, Cornell University, Ithaca, New York.

....most comprehensive study is the Chameleon project [19] The automatic generation of hypertext links (although not specifically URLs) has also been addressed in the hypertext literature. Important work based on lexical and statistical properties of the input text is reported in Salton and Buckley [20], and two surveys by Furuta [21, 22] The use of AI and natural language processing in particular in the construction of hypertext links was suggested in Mayfield and Nicholas [12] The use of semantic nets in document generation is explored in two papers by Rada [23, 24] The notion of schema in ....

Gerard Salton and Chris Buckley. On the automatic generation of content links in hypertext. Technical Report 89-993, Department of Computer Science, Cornell University, 1989.

.... dA fi fi fi Theta fi fi fi d B fi fi fi (4.1) 4.4. 2 Linking with SMART Researchers at Cornell University have developed methods for automatic link creation using an encyclopaedia as a corpus [SAB94] The method used has been adapted from those described in Cornell technical reports [SB89, SAB92] and James Allan s Ph.D. dissertation [All95] The latter have been proven effective [All95, p. 15] Each sentence and word group is represented by a vector composed of term weights. The premise that the best indexing terms are those that occur frequently in individual documents but ....

G. Salton and C. Buckley. On the automatic generation of content links in hypertext. TR 89-993, Department of Computer Science, Cornell University, April 1989.

....spam filtering, survailence, fraud detection) p(CjX ) Determine authorship. p(CjX ) ffl With in text level Automatic generation of index Finding inconsistencies between documents Extract key information from text ffl Hybrid Automatic hyperlinking, cross referencing [Salton and Buckley, 1989]. Finn A ffi rup Nielsen, DSP, IMM, DTU May 26, 1999 WHAT WE WOULD LIKE TO DO IN CONNECTION WITH NEUROINFORMATICS ffl Application of machine learning techniques on texts to do something useful Independent Component Analysis (ICA) Kolenda and Hansen, 1999] Generalizable Gaussian ....

Salton, G. and Buckley, C. (1989). On the automatic generation of content links in hypertext. Technical Report TR89--993, Department of Computer Science, Cornell University.

....a given string exactly. In contrast, we propose true semantic connections from anchors in the text to the semantic net. Most of the work that has been done on the problem of automatic construction of hypertext links stems from the information retrieval community. For example, Salton and Buckley [20] have shown how to apply information retrieval methods to the problem of automatic conversion of text to hypertext. However, these methods are almost exclusively statistical in nature. Thus, although these methods can discover associations between texts, they cannot uncover the nature of such ....

Gerard Salton and Chris Buckley. On the automatic generation of content links in hypertext. Technical Report 89-993, Department of Computer Science, Cornell University, 1989.

....it is clear that both nodes shown in Table 2 are related with regard to their topic in general. How to generate links efficiently The general complexity for processing n documents is O(n 2 ) Which node granularity to choose: documents, chapters, paragraphs, sentences, or phrases In [27] a hierarchical approach is described that reduces complexity of linking on a sentence level: First, matching documents are looked for; second, sentence linking is done within matching documents. 4 Natural Language Parsing A document s syntactic structures give important information about the ....

G. Salton and C. Buckley. On the automatic generation of content links in hypertext. Technical Report TR 89-993, Department of Computer Science, Cornell University, April 1989.

....spam filtering, survailence, fraud detection) p(CjX ) Determine authorship. p(CjX ) ffl With in text level Automatic generation of index Finding inconsistencies between documents Extract key information from text ffl Hybrid Automatic hyperlinking, cross referencing [Salton and Buckley, 1989]. 2 WHAT WE WOULD LIKE TO DO IN CONNECTION WITH NEUROINFORMATICS ffl Application of machine learning techniques on texts to do something useful Independent Component Analysis (ICA) Kolenda and Hansen, 1999] Generalizable Gaussian Mixture (model) GGM) Hansen et al. 1999] ffl ....

Salton, G. and Buckley, C. (1989). On the automatic generation of content links in hypertext. Technical Report TR89--993, Department of Computer Science, Cornell University.

No context found.

SALTON, G. And BUCKLEY, C. On the automatic generation of content links in hypertext. Department of Computer Science, Cornell University Technical Report 89-993, 1989.

No context found.

G. Salton and C. Buckley. On the automatic generation of content links in hypertext. Technical Report 89-993, Cornell University, April 1989.

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