Hayes, P. and Weinstein, S. (1990). Construe/tis: a system for content-based indexing of a database of news stories. In Annual Conference on Innovative Applications of AI.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....consists of combining two first level classifiers: one learned from the training collection by one training algorithm at the election of the system developer, and one that uses lexical knowledge to classify new instances. The second classifier could be an expert system like that described in [8], a WordNet synonyms based classifier as in our work [2] or a classifier 4 based on data specific to some domain (as spam filtering heuristics like the mail address of the sender for classifying spam email) 6, 18] A second level linear classifier can then be learnt from a separate training ....

Philip J. Hayes and Steven P. Weinstein. Construe/Tis: a system for content-based indexing of a database of news stories. In Alain Rappaport and Reid Smith, editors, Proceedings of IAAI-90, 2nd Conference on Innovative Applications of Artificial Intelligence, pages 49--66. AAAI Press, Menlo Park, US, 1990.

....regular flat category system. The parameters are conf min = 0.25, var max = 0.98. Observe that our shows the same relationship between the hierarchies as reported by D Alessio et al.: E4 hierarchy yields the best result. Wibowo and Williams [30] experienced this collection with another hierarchy [10], perhaps this is the reason why they best result is even lower, 73.74 , than 13 that has been achieved by flat categorizers. Our method achieved remarkable results on flat category system as well. This is comparable with the best known results of flat categorizers (87.8 Weiss et al. [29] ....

P. J. Hayes and S. P. Weinstein. CONSTRUE/TIS: a system for content-based indexing of a database of news stories. In A. Rappaport and R. Smith, editors, Proc. of the 2nd Conference on Innovative Applications of Artificial Intelligence (IAAI-90), pages 49--66, Menlo Park, 1990. AAAI Press.

....wechose topic spotting of newswire stories as the task and the Reuters 21578 corpus for the data. This corpus has become a new benchmark lately in TC evaluations, and is the refined version of several older versions, namely Reuters 22173 and Reuters 21450, on which many TC methods were evaluated[10, 16, 1, 28, 6, 33, 22, 31], but the results on the older versions may not be directly comparable to the results on the new version. For this paper we use the ApteMod version of Reuters 21578, which was obtained by eliminating unlabelled documents and selecting the categories whichhave at least one documentin the training ....

P.J. Hayes and S. P.Weinstein. Construe/tis: a system for content-based indexing of a database of new stories. In Second Annual Conference on Innovative Applications of Artificial Intelligence, 1990.

....they are fully automatic, eliminating the need for manual parameter tuning. i Introduction With the rapid growth of online information, text categorization has become one of the key techniques for handling and organizing text data. Text categorization is used to classify news stories [Hayes and Weinstein, 1990] [Masand et at. 1992] to find interesting information on the WWW [Lang, 1995] Batabanovic and Shoham, 1995] and to guide a users search through hypertext [Joachims et at. 1997] Since building text classifiers by hand is difficult and time consuming, it is desirable to learn classifiers from ....

Hayes, P. and Weinstein, S. (1990). Construe/tis: a system for content-based indexing of a database of news stories. In Annual Conference on Innovative Applications of AI.

....Although several versions are available, the Reuters sets are a notable exception that many researchers use for benchmarking. These sets are based on the Reuters newswire and the first version was originally produced by the Carnegie Group Inc. CGI) and used to evaluate their CONSTRUE system [3]. The other versions made available since then are mostly refinements of the so called Reuters 22173 and Reuters 21450 sets. The documents generally refer to financial news related to different industries and have a title and a content section (we have considered both indistinctly) In this work ....

P.J. Hayes and S. P. Weinstein. Construe/tis: a system for content-based indexing of a database of new stories. In Second Annual Conference on Innovative Applications of Artificial Intelligence, 1990.

....objects. There have been a number of conceptual information retrieval systems described in the literature. These include SCISOR, RESEARCHER and OpEd. 10 An Alternative To Traditional IR systems There has been a great deal of work recently on knowledge based information retrieval systems ( 13] [14]; 12] 15] Knowledge based IR systems rely on an explicit knowledge base, such as rule base [14] semantic network [13] patterns [15] or case frames [12] 11 The SMART Retrieval System (Salton) The SMART system is a sophisticated text retrieval tool based on storing all information terms in ....

....These include SCISOR, RESEARCHER and OpEd. 10 An Alternative To Traditional IR systems There has been a great deal of work recently on knowledge based information retrieval systems ( 13] 14] 12] 15] Knowledge based IR systems rely on an explicit knowledge base, such as rule base [14], semantic network [13] patterns [15] or case frames [12] 11 The SMART Retrieval System (Salton) The SMART system is a sophisticated text retrieval tool based on storing all information terms in a vector of terms. In principle, the terms might be chosen from a controlled vocabulary list or a ....

Hayes, Philip J. and Weistein, Steven P. 1991. Construe-TIS: A system for contentbased indexing of a database of news stories. In Proceedings of the second annual Conference on innovative Applications of Artificial Intelligence. AAAI Press. 49-64.

....an additional construct of the query langauge: Only documents that in addition to the traditional query have a specific category label assigned are returned to the user. A well known system for text categorization is the TCS system which is used with great success in categorizing financial news [9]. A typical pattern in TCS is (and gold (and (not medal) not jewelry) This pattern matches in every documents that contains gold , but none of the words medal or jewelry . It can be used, e.g. to find articles dealing with gold in which gold is not a good. More elaborated pattern ....

P.J. Hayes, S.P. Weinstein. Construe-TIS: A System for Content -Based Indexing of a Database of News Stories, Innovative Applications of Artificial Intelligence 2, AAAI Press / MIT Press, 1991.

....by an extensible, modular system architecture which provides a simple interface for integrating additional operators. In contrast to other approaches, our system PET (Pattern Extraction for Texts) allows to learn classification patterns as they are used for hand crafted categorization rules (cf. (Hayes Weinstein 1991), Agne Hein 1998) Essentially, the relational learning paradigm as realized in Inductive Logic Programming (ILP) cf. Nienhuys Cheng de Wolf 1997) provides an equivalent expressiveness for formulation of classification knowledge. However, our pattern language offers the advantage of ....

Hayes, P., and Weinstein, S. 1991. Construe-TIS: A System for Content-Based Indexing of a Database of News Stories. In Rappaport, A., and Smith, R., eds., Innovative Applications of Artificial Intelligence 2. AAAI Press / MIT Press. 49--64.

....categorization system, using statistics as way of augmenting hand coded knowledge. The context of this research is a commercially developed system [Rau and Jacobs, 1991] that automat ically assigns categories to news stories for custom clip ping and other markets. Like Construe TIS [Hayes and Weinstein, 1990], the work derives from, and coordinates with, NLP efforts, but the system primarily uses a lexico semantic pattern mateher for categorization [Jacobs et al. 1991] Categorization tasks vary greatly in difficulty, but the recall and precision results produced in our tests are similar to those ....

....we have found that exically driven pre processing serves as a complement to parsing and semantic interpretation, both in identifying portions of relevant text and in marking the input text to make it easier to process. Our lexico semantic pattern rules are quite similar to those in CONSTRUE TIS [Hayes and Weinstein, 1990], asso ciating each pattern with an action rule that can ma nipulate text or activate or de activate a category. This 181 type of knowledge structure has proven effective for topic identification as well as other forms of pre processing. Because the pattern matchef is designed as an efficient ....

Philip J. Hayes and Steven P. Weinstein. CONSTRUE/TIS: A system for content-based indexing of a database of news stories. In Proceedings of the Second Annual Conference on Innovative Applicalions of Artificial Intelligence, May 1990.

....testing. We summarize significant differences between test and training sets in Table 2. These differences can bring noise into categorization, because training relies on similarity between training and test documents. Nev ertheless, this 21,50 723 partition has been used be fore [Lewis, 1992; Hayes and Weinstein, 1990] and involves the general case of documents with no cateforins assigned. We have worked wth raw data provided in the Reuters distribution Control charaoters, numbers and several separators like have been removed, and categories dfferent from the TOPICS set have been ignored. For ....

P.J. Hayes and S.P. Weinstein. CONSTRUE/TIS: a system for content-based indexing of a database of newsstories. In Proceedings of the Second Annual Conference on InnovattveApph- catons of Arttfical Intelhgence, 1990.

No context found.

Hayes, P. and Weinstein, S. (1990). Construe/tis: a system for content-based indexing of a database of news stories. In Annual Conference on Innovative Applications of AI.

No context found.

Philip J. Hayes and Steven P. Weinstein. Construe/Tis: a system for contentbased indexing of a database of news stories. In Alain Rappaport and Reid Smith, editors, Proceedings of IAAI-90, 2nd Conference on Innovative Applications of Arti cial Intelligence, pages 49-66. AAAI Press, Menlo Park, US, 1990.

No context found.

Hayes, Philip J. and Weinstein, Steven P. CONSTRUE/TIS: A System for Content-Based Indexing of a Database of News Stories. In A. Rappaport and R. Smith, Innovative Applications of Artificial Intelligence 2, AAAI Press/The MIT Press, 1990.

No context found.

P. J. Hayes and S. P. Weinstein. Construe/Tis: a system for content-based indexing of a database of news stories. In A. Rappaport and R. Smith, editors, Proceedings of IAAI-90, 2nd Conference on Innovative Applications of Artificial Intelligence, pages 49--66. AAAI Press, Menlo Park, US, 1990.

No context found.

P. J. Hayes and S. P. Weinstein, "CONSTRUE/TIS: a system for contentbased indexing of a database of news stories," in Proc. of the 2nd Conference on Innovative Applications of Artificial Intelligence (IAAI90) , A. Rappaport and R. Smith, Eds. Menlo Park: AAAI Press, 1990, pp. 49--66.

No context found.

-225, April, 1995. Hayes, P.J., Weinstein, S.P. Construe-TIS: A System for Content-Based Indexing of a Database of News Stories, 2nd Annual Conference on Innovative Applications of Artificial Intelligence, pp.

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