Baldwin, B., Doran, C., Reynar, J. C., Niv, M., Srinivas, B., and Wasson, M. (1997). EAGLE: An extensible architecture for general linguistic engineering. In Proceedings of RIAO-97, pages 271--283, Montreal.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....lists. Gale, Church and Yarowsky [5] showed that words strongly tend to exhibit only one sense in a document or discourse. This is also one of the assumptions of the document centered approach advocated in this proposal. The description of the EAGLE workbench for linguistic engineering [2] mentions a case normalization module which uses a heuristic that a capitalized word in a mandatory position should be downcased if it is found lowercased in the same document. This also employs a database of bigrams and unigrams of lowercased and capitalized words found in unambiguous positions ....

B. Baldwin, C. Doran, J. Reynar, M. Niv, B. Srinivas and M. Wasson. Eagle: An extensible architecture for general linguistic engineering. In Proceedings of RIAO '97, Montreal, June 1997.

....changeovers 2 and new product introductions. The documents were drawn from a variety of news sources such as Reuters, Business wire, PR news, LA Times and San Fransisco Chronicle. To do this task, we employed the EAGLE (An Extensible Architecture for General Linguistic Engineering) system[Baldwin et al. 1996] which has been under development since 1995 at the University of Pennsylvania. The EAGLE system provides a flexible architecture to integrate a variety of Natural Language tools in novel ways for text processing. It contains document preprocessing tools for sentence segmentation, case ....

....syntactic analysis tools such as a morphological analyzer, a combination of three partof speech taggers, supertagger and statistical parser; and discourse processing tools for detecting coreference information. A detailed description of each of these components of the system is provided in [Baldwin et al. 1996]. A pattern description language called 2 The fields of this template were not identical to those in the corresponding MUC 6 template [Committee, 1995] Mother of PERL (MOP) Doran et al. 1996] has also been developed in conjunction with the EAGLE system. MOP allows a user to specify patterns ....

[Article contains additional citation context not shown here]

Breckenridge Baldwin, Christine Doran, Jeffrey Reynar, Michael Niv, B. Srinivas, and Mark Wasson. EAGLE: An Extensible Architecture for General Linguistic Engineering. Manuscript, Department of Computer and Information Sciences, University of Pennsylvania, 1996.

....Figure 3.10 shows a portion of a transcript of an episode of NPR s All Things Considered annotated with links indicating which phrases refer to the same entities. Unfortunately, pronoun resolution remains an unsolved problem and most computational systems today address only a subset of it [Baldwin, 1997]. However, references made using repetitions of people s names, the names of companies or organizations, and place names can be easily detected. In the same way that identical word n grams are unlikely to arise independently in different topic segments, repetitions of proper names are also ....

....[Yarowsky, 1992] language modeling (e. g [Beeferman et al. 1997a] coreference resolution [Kehler, 1997] and identifying the most likely genre for a document [Losee, 1996, Kessler et al. 1997] Pronoun resolution techniques often search the preceding context for candidate antecedents (e.g. [Baldwin, 1997]) Using data found in arbitrary windows of words has yielded state of the art systems. However, performance on various tasks should improve if more motivated segments were used. Words related to one topic would not erroneously be counted as co occurring with words from a neighboring topic ....

[Article contains additional citation context not shown here]

Baldwin, B., Doran, C., Reynar, J. C., Niv, M., Srinivas, B., and Wasson, M. (1997). EAGLE: An extensible architecture for general linguistic engineering. In Proceedings of RIAO-97, pages 271--283, Montreal.

....query which are represented in the summary are covered. A phrase in the document is considered to cover a phrase in the query if it is coreferent with it. This approach maximizes the space of entities retained in the summary with minimal redundancy. The software is built upon the CAMP NLP system [3]. Problem Statement Given the relative immaturity of summarization technologies and their evaluation, it is worthwhile to describe our approach in detail and the problems it is intended to solve. An important aspect of our technique is that we produce sentence extraction summaries which are ....

....include: named entity recognition, tokenization, sentence detection, part of speech tagging, morphological analysis, parsing, argument detection, and coreference resolution. Many of the techniques used for these tasks perform at or near the state of the art and are described in more depth in [16, 12, 11, 9, 6, 2, 3]. The system produces coreference annotated documents which serve as the input to the summarization algorithm. Relating the query to the document The relationships discussed previously are approximated via a series of associations between tokens in the query, headline, and the body of the ....

Breck Baldwin, Christine Doran, Jeffrey C. Reynar, Michael Niv, B. Srinivas, and Mark Wasson. EAGLE: An extensible architecture for general linguistic engineering. In Proceedings of RIAO-97, Montreal, 1997.

No context found.

Baldwin, B., C. Doran, J. Reynar, M. Niv, and M. Wasson. EAGLE: An Extensible Architecture for General Linguistic Engineering. Proceedings RIAO, Computer-Assisted Information Searching on Internet, Montreal, Canada, 1997.

....query which are represented in the summary are covered. A phrase in the document is considered to cover a phrase in the query if it is coreferent with it. This approach maximizes the space of entities retained in the summary with minimal redundancy. The software is built upon the CAMP NLP system [2]. Problem Statement Given the relative immaturity of summarization technologies and their evaluation, it is worthwhile to describe our approach in detail and the problems it is intended to solve. An important aspect of our technique is that we produce sentence extraction summaries which are ....

....include: named entity recognition, tokenization, sentence detection, part of speech tagging, morphological analysis, parsing, argument detection, and coreference resolution. Many of the techniques used for these tasks perform at or near the state of the art and are described in more depth in [12, 9, 8, 7, 5, 1, 2]. The system produces coreference annotated documents which serve as the input to the summarization algorithm. Relating the query to the document The relationships discussed previously are approximated via a series of associations between tokens in the query, headline, and the body of the ....

Breck Baldwin, Christine Doran, Jeffrey C. Reynar, Michael Niv, B. Srinivas, and Mark Wasson. EAGLE: An extensible architecture for general linguistic engineering. In Proceedings of RIAO-97, Montreal, 1997.

No context found.

Breck Baldwin, Christine Doran, Jeffrey C. Reynar, Michael Niv, B. Srinivas and Mark Wasson. 1997. EAGLE: An Extensible Architecture for General Linguistic Engineering.

No context found.

B. Baldwin, C. Doran, J.C. Reynar, B. Srinivas, M. Niv, and M. Wasson. EAGLE: An Extensible Architecture for General Linguistic Engineering. In In Proceedings of RIAO'97, Montreal, June 1997.

No context found.

B. Baldwin, C. Doran, J.C. Reynar, B. Srinivas, M. Niv, and M. Wasson. EAGLE: An Extensible Architecture for General Linguistic Engineering. In In Proceedings of RIAO'97, Montreal, June 1997.

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