Filters








2,199 Hits in 4.4 sec

Text filtering in MUC-3 and MUC-4

David D. Lewis, Richard M. Tong
1992 Proceedings of the 4th conference on Message understanding - MUC4 '92  
Acknowledgments We thank DARPA for supporting the authors' travel to the MUC-4 workshop, and Beth Sundheim for her aid with this paper and her splendid handling of the MUC-4 process.  ...  Text Filtering Effectiveness in MUC-3 and MUC-4 In this section we present data on the effectiveness of system level text filtering in MUC-3 and MUC-4 systems.  ...  Use of Text Filtering in MUC-3 and MUC-4 Systems In this section we briefly describe the text filtering processes performed by each of the sites participating in MUC-3 and/or MUC-4.  ... 
doi:10.3115/1072064.1072069 dblp:conf/muc/LewisT92 fatcat:6nl63dsr55dlln5bqa4nhepzpa

Page 446 of Computational Linguistics Vol. 19, Issue 3 [page]

1993 Computational Linguistics  
In addition, results of three adjunct tests were reported at MUC-4. The first was an analysis of the text-filtering capabilities of the MUC-3 and MUC-4 systems (Lewis and Tong 1992).  ...  There were two blind test sets in MUC-4 labeled TST3 and TST4.  ... 

MUC-4 evaluation metrics

Nancy Chinchor
1992 Proceedings of the 4th conference on Message understanding - MUC4 '92  
So even if a system's official scores were the same for MUC-3 and MUC-4, the system had improved in MUC-4 .  ...  Given the differences in scoring between MUC-3 and MUC-4, the MUC-4 systems' scores represent a larger improvement over MUC-3 performance than the numbers themselves suggest .  ... 
doi:10.3115/1072064.1072067 dblp:conf/muc/Chinchor92 fatcat:sftc3xlrjfga3hsy54mpwbwedy

BBN PLUM

Ralph Weischedel, Damaris Ayuso, Sean Boisen, Heidi Fox, Herbert Gish, Robert Ingria
1992 Proceedings of the 4th conference on Message understanding - MUC4 '92  
GOALS Our mid-term to long-term goals in data extraction from text for the next one to three years are to achieve muc h greater portability to new languages and new domains, greater robustness, and greater  ...  Probabilistic models can contribute to robustness by allowing for words, constructions, and forms not anticipated ahead of time and by looking for the most likely interpretation in context. 7 19a.  ...  ACKNOWLEDGMENT S The work reported here was supported in part by the Defense Advanced Research Projects Agency and was monitored by the Rome Air Development Center under Contract No .  ... 
doi:10.3115/1072064.1072073 dblp:conf/muc/WeischedelABFGI92 fatcat:vl4xooqvlfdtfppmbembxlbr44

Appendix B: MUC-7 Test Scores Introduction

Nancy Chinchor
1998 Message Understanding Conference  
This appendix contains the summary score reports for each system in the MUC-7 evaluation.  ...  The reports are ordered by task (Template Element, Template Relation, Scenario Template, Named Entity, and Coreference) and secondarily by site and language (in alphabetical order).  ...  In Proceedings, Fourth Message Understanding Conference (MUC-4). Morgan Kaufmann. San Mateo, CA. .  ... 
dblp:conf/muc/Chinchor98a fatcat:n3sw7jg2bbdyfglnxlkc5hch4e

GE NLToolset

Lisa Rau, George Krupka, Paul Jacobs, Ira Sider, Lois Childs
1992 Proceedings of the 4th conference on Message understanding - MUC4 '92  
INTRODUCTIO N We report on the GE results from the MUC-4 conference and provide an analysis of system performance . In general, MUC-4 was a very successful effort for GE .  ...  MUC-4 task .  ...  The move from MUC-3 to MUC-4 produced some clear improvements in the test as well as the systems .  ... 
doi:10.3115/1072064.1072074 dblp:conf/muc/RauKJSC92 fatcat:gv22xsxg4zbfpg653yy2ev362a

Language Systems, Inc

Christine A. Montgomery, Bonnie Glover Stalls, Robert S. Belvin, Robert E. Stumberger
1991 Proceedings of the 3rd conference on Message understanding - MUC3 '91  
In evaluating these scores, which place LSI about two/thirds of the way down in a results-ranked list of MUC-3 participants, it should be noted that our MUC-3 system reflects a major redevelopment of the  ...  Had we chosen instead to go with the fairly robust previous version o f the DBG system (described in [2] and [3], recently evaluated for Rome Laboratory by KSC, Inc .), our MUC-3 scores would certainly  ...  more variety in the sentenc e types in the MUC-3 texts, and the time and resources available were substantially less than the 18 months an d $225K dedicated to the Rome effort .  ... 
doi:10.3115/1071958.1071971 dblp:conf/muc/MontgomerySBS91 fatcat:t6nhmmg6ifeq7bwvk6rrjaimi4

An empirical study of automated dictionary construction for information extraction in three domains

Ellen Riloff
1996 Artificial Intelligence  
In this paper, we describe experiments with AutoSlog in two additional domains: joint ventures and microelectronics.  ...  We compare the performance of AutoSlog across the three domains, discuss the lessons learned about the generality of this approach, and present results from two experiments which demonstrate that novice  ...  Acknowledgements We would like to thank Wendy Lehnert for her help in setting up both of the experiments described in Section 4, David Fisher and Jon Peterson for designing and programming the AutoSlog  ... 
doi:10.1016/0004-3702(95)00123-9 fatcat:hh6fusripfhp7lmtdwnl4yzpb4

Tipster/MUC-5 information extraction system evaluation

Beth M. Sundheim
1993 Proceedings of a workshop on held at Fredericksburg, Virginia September 19-23, 1993 -  
27 MUC-5 CLASS OF PARTICIPANT PARTICIPATION SYSTEM EJV EME JJV JM E BBN Tipster PLUM X X X X GE/CMU Tipster SHOGUN X X X X TEXIRACT X X Language Sys ., Inc . non-Tipster DBG X MITRE non-Tipster ALEMBIC  ...  She would also like to acknowledge the active participation of th e ARPA/SISTO sponsors, Thomas Crystal and George Doddington, and the long-suffering assistance of the MUC -  ...  ACKNOWLEDGEMENT S The author would like to acknowledge the extensive support throughout Tipster Phase 1 from Nanc y Chinchor and Gary Dungca at SAIC .  ... 
doi:10.3115/1119149.1119168 dblp:conf/tipster/Sundheim93a fatcat:qm4mk3a4kzdrxgb6lpabju23u4

Overview of the fourth message understanding evaluation and conference

Beth M. Sundheim
1992 Proceedings of the 4th conference on Message understanding - MUC4 '92  
Isolation of Text Filtering Performance Overall, approximately 50% of the texts in the MUC-3 and MUC-4 corpora are irrelevant to the information extraction evaluation task.  ...  One of the fundamental ways in which MUC-3 and MUC-4 are distinct from earlier efforts is in their choice of texts: MUC-3 and MUC-4 made use of news articles on the subject of Latin American terrorism,  ... 
doi:10.3115/1072064.1072066 dblp:conf/muc/Sundheim92 fatcat:iivua2se7reg7f6zrspmgvang4

Tipster/MUC-5

Beth M. Sundheim
1993 Proceedings of the 5th conference on Message understanding - MUC5 '93  
27 MUC-5 CLASS OF PARTICIPANT PARTICIPATION SYSTEM EJV EME JJV JM E BBN Tipster PLUM X X X X GE/CMU Tipster SHOGUN X X X X TEXIRACT X X Language Sys ., Inc . non-Tipster DBG X MITRE non-Tipster ALEMBIC  ...  She would also like to acknowledge the active participation of th e ARPA/SISTO sponsors, Thomas Crystal and George Doddington, and the long-suffering assistance of the MUC -  ...  ACKNOWLEDGEMENT S The author would like to acknowledge the extensive support throughout Tipster Phase 1 from Nanc y Chinchor and Gary Dungca at SAIC .  ... 
doi:10.3115/1072017.1072023 dblp:conf/muc/Sundheim93 fatcat:27a465zzhng6pocvujkeznn24y

Advanced decision systems' CODEX

Laura Blumer Balcom, Richard M. Tong
1991 Proceedings of the 3rd conference on Message understanding - MUC3 '91  
This strategy might change in th e future, depending on the robustness of the parser in analyzing MUC-3 texts .  ...  4 MUC-3 Effort by Category Table 5 : 5 Number of Knowledge Base Structures for MUC-3  ... 
doi:10.3115/1071958.1071965 dblp:conf/muc/BalcomT91 fatcat:dtgaw7qtkbgpzjhuwcttsdkopy

MUC-5 evaluation metrics

Nancy Chinchor, Beth Sundheim
1993 Proceedings of the 5th conference on Message understanding - MUC5 '93  
The official MUC-5 metrics express error rates while the official MUC-4 metric s express performance in terms of recall and precision (used for MUC-5 only as "unofficial" metrics) .  ...  INTRODUCTION The metrics used for the Fifth Message Understanding Conference (MUC-5) evaluation are a major update to those used for MUC-4 in 1992.  ...  Evaluation Metric s Table 3 : 3 System-dependent Error Metrics . Table 4 : 4 Richness-normalized error. Table 5 : 5 Recall-Precision Evaluation Metrics . Table 6 : 6 Text Filtering. .  ... 
doi:10.3115/1072017.1072026 dblp:conf/muc/ChinchorS93 fatcat:ihxhnpwnhbbs5n2aam5ku7nvvy

Description of the LINK system used for MUC-4

Steven L. Lytinen, Sayan Bhattacharyya, Robert R. Burridge, Peter M. Hastings, Christian Huyck, Karen A. Lipinsky, Eric S. McDaniel, Karenann K. Terrell
1992 Proceedings of the 4th conference on Message understanding - MUC4 '92  
Prior to MUC-4, LINK ha d been used to extract information from free-form texts in two narrow application domains .  ...  As a comparison, the lexicons for the previous domains contained only 300-500 words, compared wit h 6700 words in our MUC-4 test configuration .  ...  It also performs a few simple editing tasks, such as separating the text into sentences, and removing text that i s in brackets .  ... 
doi:10.3115/1072064.1072106 dblp:conf/muc/LytinenBBHHLMT92a fatcat:2zrt3n54wngihkv2jfa5k6heo4

A Performance Evaluation of Text-Analysis Technologies

Wendy G. Lehnert, Beth Sundheim
1991 The AI Magazine  
does not represent an official analysis of MUC-3, its methodology, or results. The views expressed herein are solely those of the authors as participants in the evaluation and conference.  ...  The official conference report is available in proceedings form (MUC-3 1991). Notes 1.  ...  Figure 3 . 3 Figure 3. An Official TST2 Score Report (from the University of Massachusetts). Figure 4 . 4 Figure 4. The Official MUC-3 Scores for 15 Participating Research Sites.  ... 
doi:10.1609/aimag.v12i3.905 dblp:journals/aim/LehnertS91 fatcat:mfualy6henfnln47jfnvyx7wuu
« Previous Showing results 1 — 15 out of 2,199 results