{ "paper_id": "M92-1019", "header": { "generated_with": "S2ORC 1.0.0", "date_generated": "2023-01-19T03:13:22.216409Z" }, "title": "SRI INTERNATIONAL FASTUS SYSTE M MUC-4 TEST RESULTS AND ANALYSI S", "authors": [ { "first": "Douglas", "middle": [ "E" ], "last": "Appelt", "suffix": "", "affiliation": { "laboratory": "", "institution": "SRI International", "location": { "settlement": "Menlo Park", "country": "California" } }, "email": "" }, { "first": "John", "middle": [], "last": "Bear", "suffix": "", "affiliation": { "laboratory": "", "institution": "SRI International", "location": { "settlement": "Menlo Park", "country": "California" } }, "email": "" }, { "first": "Jerry", "middle": [ "R" ], "last": "Hobbs", "suffix": "", "affiliation": { "laboratory": "", "institution": "SRI International", "location": { "settlement": "Menlo Park", "country": "California" } }, "email": "" }, { "first": "David", "middle": [], "last": "Israel", "suffix": "", "affiliation": { "laboratory": "", "institution": "SRI International", "location": { "settlement": "Menlo Park", "country": "California" } }, "email": "" }, { "first": "Mabry", "middle": [], "last": "Tyso", "suffix": "", "affiliation": { "laboratory": "", "institution": "SRI International", "location": { "settlement": "Menlo Park", "country": "California" } }, "email": "" } ], "year": "", "venue": null, "identifiers": {}, "abstract": "The system that SRI used for the MUC-4 evaluation represents a significant departure from syste m architectures that have been employed in the past. In MUC-2 and MUC-3, SRI used the TACITUS tex t processing system [I], which was based on the DIALOGIC parser and grammar, and an abudctive reasone r for horn-clause logic. In MUC-4, SRI designed a new system called FASTUS (a permutation of the initia l letters in Finite State .Automata-based Text Understanding System) which we feel represents a significant advance in the state of the art of text processing. The system shares certain modules with the earlie r TACITUS system, namely modules for text preprocessing and standardization, spelling correction, Hispani c name recognition, and the core lexicon. However, the DIALOGIC system and abductive reasoner, which wer e the heart and soul of the previous system, were replaced by a system whose architecture is based on cascade d finite-state automata. Using this system we were capable of achieving a significant level of performance on the MUC-4 task with less than one month devoted to domain-specific development. In addition, the system is extremely fast, and is capable of processing texts at the rate of approximately 3,200 words per minute , measured in CPU time on a Sun SPARC-2 processor. (Measured according to elapsed real time, the system about 50% slower, but the observed time depends on the particular hardware configuration involved .) OVERVIEW OF THE FASTUS ARCHITECTUR E The architecture of the FASTUS system is described in detail in the associated system summary. It can be summarized as a three-phase process. The first phase consists of scanning the text to identify prope r names, correcting spelling, and similar preprocessing tasks to ensure that the text is in a standardized forma t for the remainder of the processing .", "pdf_parse": { "paper_id": "M92-1019", "_pdf_hash": "", "abstract": [ { "text": "The system that SRI used for the MUC-4 evaluation represents a significant departure from syste m architectures that have been employed in the past. In MUC-2 and MUC-3, SRI used the TACITUS tex t processing system [I], which was based on the DIALOGIC parser and grammar, and an abudctive reasone r for horn-clause logic. In MUC-4, SRI designed a new system called FASTUS (a permutation of the initia l letters in Finite State .Automata-based Text Understanding System) which we feel represents a significant advance in the state of the art of text processing. The system shares certain modules with the earlie r TACITUS system, namely modules for text preprocessing and standardization, spelling correction, Hispani c name recognition, and the core lexicon. However, the DIALOGIC system and abductive reasoner, which wer e the heart and soul of the previous system, were replaced by a system whose architecture is based on cascade d finite-state automata. Using this system we were capable of achieving a significant level of performance on the MUC-4 task with less than one month devoted to domain-specific development. In addition, the system is extremely fast, and is capable of processing texts at the rate of approximately 3,200 words per minute , measured in CPU time on a Sun SPARC-2 processor. (Measured according to elapsed real time, the system about 50% slower, but the observed time depends on the particular hardware configuration involved .) OVERVIEW OF THE FASTUS ARCHITECTUR E The architecture of the FASTUS system is described in detail in the associated system summary. It can be summarized as a three-phase process. The first phase consists of scanning the text to identify prope r names, correcting spelling, and similar preprocessing tasks to ensure that the text is in a standardized forma t for the remainder of the processing .", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Abstract", "sec_num": null } ], "body_text": [ { "text": "The second phase consists of a finite-state machine that accepts the sequence of words from the text, an d produces as output a sequence of linguistic consituents -noun groups consisting of determiners, prenominals and head noun, verb groups consisting of auxilliaries plus the main verb together with any intervenin g adverbs, and particles, which is a catch-all category including prepositions, conjunctions, and genitive markers . The output of the second pass is filtered to include only the longest consitutents spanning any give n portion of the sentence .", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "", "sec_num": null }, { "text": "The linguistic consituents from the second phase are given as input to another finite-state machine . The transitions of this third-phase machine are based on the head of each constituent, and each transition build s some piece of an \"incident.\" structure, which can be thought of as a \"proto-template .\" When a final state of the machine is reached, the incident, structure that has been produced through that point is saved, an d merged with all other incident structures produced by the same sentence . (There may be several, because the machines are non-deterministic) . These incident structures are then merged with incident structure s from the rest of the text according to a set of merging heuristics . The incident structures are converted t o the format of MUC-4 templates in a post-processing phase .", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "", "sec_num": null }, { "text": "In the course of designing the system, we paramaterized a number of characteristics of the system's operation because we believed that the parameterized behavior would reflect tradeoffs in recall versus precision . Subsequent testing revealed that many of these parameters result in both higher recall and higher precisio n when in one state or the other, and therefore we left them permanently in their most advantageous state . Those parameters that seemed to affect recall the the expense of precision were set to produce a test ru n in which we attempted to maximize the system's recall . The effect of these parameters could be described in general as distrusting the system's filters' ability to eliminate templates corresponding to stale dates , uninteresting countries, and military incidents . We observed a small but measurable increase in recall at the expense of precision by distrusting our filters .", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "CONTROLLING THE FASTUS SYSTE M", "sec_num": null }, { "text": "The following parameters were implemented and tested on 300 texts before arriving at the decisions fo r the settings on the final run .", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "CONTROLLING THE FASTUS SYSTE M", "sec_num": null }, { "text": "\u2022 Conservative Merging. When this option is selected, the system would not merge incidents that ha d non-overlapping targets with proper names . When not selected, any merges consistent with the inciden t types were permitted . Testing revealed that merging should always be conservative .", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "CONTROLLING THE FASTUS SYSTE M", "sec_num": null }, { "text": "\u2022 Civilian Target Requirement. This filter would reject any template that did not have at least on e non-military target, including templates that identified a perpetrator, but no physical or human targe t at all . This option appears to produce a recall-precision tradeoff of about one or two points .", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "CONTROLLING THE FASTUS SYSTE M", "sec_num": null }, { "text": "\u2022 Subjectless Verb Groups . This parameter would allow the system to generate an incident structur e from a verb together with its object, even if its subject could not be determined . Although early tests showed a recall-precision tradeoff, subsequent and more thorough testing indicated that this shoul d always be done .", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "CONTROLLING THE FASTUS SYSTE M", "sec_num": null }, { "text": "\u2022 Filter Many-Target Templates. This filter would disallow any template that had more than 100 targets , on the supposition that such templates often result from vague or general, and hence irrelevant, descriptions . This turns out to be a correct heuristic, but only if the number of targets is evenly divisibl e by 100 . (An airline bombing with 307 victims is certainly interesting, while \"70,000 peasants hav e been killed\" is probably vague) .", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "CONTROLLING THE FASTUS SYSTE M", "sec_num": null }, { "text": "\u2022 Military Filtering. This heuristic causes the system to eliminate all military targets from templates , on the belief that we may have incorrectly merged a military incident with a civilian incident and incorrectly reported the union of the two . Tests show that this filtering improves precision slightly .", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "CONTROLLING THE FASTUS SYSTE M", "sec_num": null }, { "text": "\u2022 Liberal Perpetrator Org. Setting this parameter would cause the system to pick any likely perpetrato r organization out of the text, ignoring whatever the text actually says . 'Testing showed that thi s parameter had no effect, which was such a surprising result that we distrust it, and regard our testin g as inconclusive .", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "CONTROLLING THE FASTUS SYSTE M", "sec_num": null }, { "text": "\u2022 Spelling Correction This parameter controls how much spelling correction the system does . Our experiments indicated that spelling correction hurts, primarily because novel proper names get corrected t o other words, and hence lost . We tried a weaker version of spelling correction which would correct onl y misspelled words that did not occur on a large list of proper names that we had assembled . This showe d an improvement, but spelling correction still had a small negative effect . This was also a surprisin g result, and we were not willing to abandon spelling correction, and ran all tests with weak spellin g correction enabled, although to some extent a complete lack of spelling correction is compensated fo r by the presence of common misspellings of important domain words like \"guerrilla\" and \" assassinate\" in the lexicon .", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "CONTROLLING THE FASTUS SYSTE M", "sec_num": null }, { "text": "\u2022 Stale Date Filtering . This parameter causes filtering of any template that has a date that is earlie r than two months before the date of the article . Eliminating this filtering produces an increase in recal l at the expense of precision, the magnitude of which depends on how well our date detection currentl y works . We would expect about a one-point tradeoff.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "CONTROLLING THE FASTUS SYSTE M", "sec_num": null }, { "text": "\u2022 Weak Location Filtering . If the system's location dection finds that the location of an incident i s impossible according to the system ' s location database, it eliminates the template . If this flag is set , the template will be produced using only the country as the location . Testing shows that this is always desireable .", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "CONTROLLING THE FASTUS SYSTE M", "sec_num": null }, { "text": "On TST3, we achieved a recall of 44% with precision of 55% in the all-templates row, for an F-scor e (3 = 1) of 48 .9 . On TST4, the test on incidents from a different time span, we observed, surprisingly, a n identical recall score of 44%, however our precision fell to 52%, for an F-score of 47 .7 . It was reassuring to see that there was very little degradation in performance moving to a time period over which the syste m had not been trained . We also submitted a run in which we attempted to maximize the system's recal l by not filtering military targets, and allowing incidents with stale dates . On TST3, this led to a two-poin t increase in recall at the expense of one point in precision . On TST4, our recall did not increase, howeve r our precision fell by a point, giving us a lower F-score on this run . These results were consistent with ou r observations during testing, although our failure to produce even a small increase in recall on TST4 wa s somewhat disappointing .", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "THE RESULTS ON TST3 AND TST4", "sec_num": null }, { "text": "The runtime for the entire TST3 message set on a SPARC-2 processor was 11 .8 minutes (about 1 6 minutes of elapsed real time with our configuration of memory and disk) . These times are quite consisten t with our runs over the development sets . During the course of development, the overall run time for 10 0 messages increased approximately 50%, but we attribute this increase to the decision to treat more sentence s as relevant . It appears possible to increase the coverage of the system without an unacceptable increase i n processing time .", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "THE RESULTS ON TST3 AND TST4", "sec_num": null }, { "text": "During December of 1991 we decided to implement a preprocessor for the TACITUS system, at whic h point the FASTUS architecture was born . The system was originally conceived as a preprocessor for TACI-TUS that could be run in a stand-alone mode . Considerably later in our development we decided that th e performance of FASTUS on the MUC-4 task was so high that we could make FASTUS our complete system . Most of the design work for the FASTUS system took place during January . The ideas were tested ou t on finding incident locations in February, and with some initial favorable results in hand, we proceded wit h the implementation of the system in March . The implementation of the second phase of processing wa s completed in March, and the general outline of phase three was completed by the end of April . On May 6, we did the first test of the FASTUS system on TST2, which had been withheld as a fair test, and we obtained a score of 8% recall and 42% precision . At that point we began a fairly intensive effort to hill-clim b on all 1300 development texts, doing periodic runs on the fair test to monitor our progress, culminating i n a score of 44% recall, 57% precision in the wee hours of June 1, when we decided to run the official test . A s the chart in Figure 1 points out, the rate of progress was rapid enough that even a few hours of work coul d be shown to have a noticeable impact on the score . Our scarcest resource was time, and our supply of it wa s eventually exhausted well before the point of diminishing returns .", "cite_spans": [], "ref_spans": [ { "start": 1271, "end": 1279, "text": "Figure 1", "ref_id": "FIGREF0" } ], "eq_spans": [], "section": "DEVELOPMENT HISTORY", "sec_num": null }, { "text": "FASTUS was more successful than we ever dreamed when the idea was originally conceived . In retrospect , we attribute its success to the fact that its processing is extremely well suited to the demands of the task . The system's phase-3 works successfully because the input from phase-2 is already reliably processed . Phase two does only the linguistic processing that can be done reliably and fast, ignoring all the problems of makin g attachment decisions, and the ambiguity introduced by coordination and appositives . This input is adequate for phase-3 because the domain pragmatics are sufficiently constrained that given this initial chunking, th e relevant information can be reliably detected and extracted .", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "CONCLUSION S", "sec_num": null }, { "text": "One source of frustration with the development of this system is that we never had the opportunity t o produce a decent developer's interface . We believe that phase-2 is almost completely domain independent , with all the domain specific knowledge embedded in the phase-3 automata . We feel that with some carefu l thought devoted to such an interface, we could produce a general text processing system that could b e brought up to our current level of performance on a MUC-like or TIPSTER-like task in even less than th e three and a half weeks of effort that we required . Another discovery of this experience is that a MUG-like task is much easier than anyone ever thought . Although the full linguistic complexity of the MUC texts is very high, with long sentences and interestin g discourse structure problems, the relative simplicity of the information-extraction task allows much of thi s linguistic complexity to by bypassed -indeed much more than we had originally believed was possible . Th e key to the whole problem, as we see it from our FASTUS experience, is to do exactly the right amount o f syntax, so that pragmatics can take over its share of the load . For the MUC task, we think FASTUS display s exactly the right mixture .", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "CONCLUSION S", "sec_num": null }, { "text": "Finally, we point out that while FASTUS is an elegant engineering achievement,, the whole host o f linguistic problems that were bypassed are still out there, and will have to be addressed eventually fo r more complex tasks . and to achieve ever higher performance on simpler tasks . The nature of competitiv e evaluations is that they force everyone to deal with the easiest problems first . However, the hard problems cannot be ignored forever, and scientific progress requires that they be addressed .", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "CONCLUSION S", "sec_num": null } ], "back_matter": [ { "text": "This research was funded by the Defense Advanced Research Projects Agency under Office of Nava l Research contracts N00014-90-C-0220, and by an internal research and development grant from SRI International .", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "ACKNOWLEDGEMENTS", "sec_num": null } ], "bib_entries": { "BIBREF0": { "ref_id": "b0", "title": "Description of the TACITUS System as Used for MUC-3", "authors": [ { "first": "J", "middle": [], "last": "Hobbs", "suffix": "" } ], "year": 1991, "venue": "Proceedings of the MUC-3 Workshop", "volume": "", "issue": "", "pages": "200--206", "other_ids": {}, "num": null, "urls": [], "raw_text": "Hobbs, J ., et al ., \"Description of the TACITUS System as Used for MUC-3,\" Proceedings of the MUC-3 Workshop, 1991, pp . 200-206 .", "links": null } }, "ref_entries": { "FIGREF0": { "type_str": "figure", "text": "Plot of F-Score versus Date for FASTUS Developmen t", "num": null, "uris": null } } } }