acl acl2010 acl2010-108 knowledge-graph by maker-knowledge-mining
Source: pdf
Author: Clifton McFate
Abstract: A robust dictionary of semantic frames is an essential element of natural language understanding systems that use ontologies. However, creating lexical resources that accurately capture semantic representations en masse is a persistent problem. Where the sheer amount of content makes hand creation inefficient, computerized approaches often suffer from over generality and difficulty with sense disambiguation. This paper describes a semi-automatic method to create verb semantic frames in the Cyc ontology by converting the information contained in VerbNet into a Cyc usable format. This method captures the differences in meaning between types of verbs, and uses existing connections between WordNet, VerbNet, and Cyc to specify distinctions between individual verbs when available. This method provides 27,909 frames to OpenCyc which currently has none and can be used to extend ResearchCyc as well. We show that these frames lead to a 20% increase in sample sentences parsed over the Research Cyc verb lexicon. 1
Reference: text
sentIndex sentText sentNum sentScore
1 edu Abstract A robust dictionary of semantic frames is an essential element of natural language understanding systems that use ontologies. [sent-5, score-0.523]
2 This paper describes a semi-automatic method to create verb semantic frames in the Cyc ontology by converting the information contained in VerbNet into a Cyc usable format. [sent-8, score-0.651]
3 This method captures the differences in meaning between types of verbs, and uses existing connections between WordNet, VerbNet, and Cyc to specify distinctions between individual verbs when available. [sent-9, score-0.136]
4 This method provides 27,909 frames to OpenCyc which currently has none and can be used to extend ResearchCyc as well. [sent-10, score-0.481]
5 We show that these frames lead to a 20% increase in sample sentences parsed over the Research Cyc verb lexicon. [sent-11, score-0.555]
6 Higher order predicates built into Cyc’s formal language, CycL, allow efficient inferencing about context and meta-language reasoning above and beyond first-order logic rules (Ramachandran et al, 2005). [sent-14, score-0.173]
7 Such applications use NL-to-Cycl parsers which use Cyc semantic frames to convert natural language into Cyc representations. [sent-18, score-0.523]
8 These frames represent sentence content through a set of propositional logic assertions that first reify the sentence in terms of a real world event and then define the semantic relationships between the elements of the sentence, as described later. [sent-19, score-0.613]
9 Because these parsers require semantic frames to represent sentence content, existing parsers are limited due to Cyc’s limited coverage (Curtis et al, 2009). [sent-20, score-0.586]
10 The goal is to increase this coverage by automatically translating the class frames in VerbNet into individual verb templates. [sent-21, score-0.662]
11 However, the semantic frames remain mostly hand-made in ResearchCyc2 and nonexistent in the open-license OpenCyc3. [sent-23, score-0.523]
12 Translating VerbNet frames into Cyc will expand the natural language capabilities of both. [sent-24, score-0.506]
13 There has been previous research on mapping existing Cyc templates to VerbNet, but thus far these approaches have not created new templates to address Cyc’s lapses in coverage. [sent-25, score-0.193]
14 Correspondences between a few VerbNet frames and ResearchCyc templates have also been mapped out through the VxC VerbNet Cyc 2 http://research. [sent-29, score-0.577]
15 A notable exception to the hand-made paradigm is Curtis et al’s (2009) TextLearner which uses rules and existing semantic frames to handle novel sentence structures. [sent-36, score-0.564]
16 Given an existing template that fits some of the syntactic constraints of the sentence, TextLearner will attempt to create a new frame by suggesting a predicate that fits the missing part. [sent-37, score-0.326]
17 Often these are general underspecified predicates, but TextLearner is able to use common sense reasoning and existing facts to find better matches (Curtis et al, 2009). [sent-38, score-0.128]
18 While TextLearner improves its performance with time, it is not an attempt to create new frames on a large scale. [sent-39, score-0.481]
19 Creating generalized frames based on verb classes will increase the depth of the Cyc Lexicon quickly. [sent-40, score-0.577]
20 Furthermore, automatic processes like those in TextLearner could be used to make individual verb semantic frames more specific. [sent-41, score-0.619]
21 3 VerbNet VerbNet is an extension of Levin’s (1993) verb classes that uses the class structure to apply general syntactic frames to member verbs that have those syntactic uses and similar semantic meanings (Kipper et al, 2000). [sent-42, score-0.851]
22 The syntactic roles in the frame are appended with general thematic roles that fill arguments of semantic predicates. [sent-46, score-0.283]
23 Each event is broken down into a tripartite structure as described by Moens & Steedman (1988) and uses a time modifier for each predicate to indicate when specific predicates occur in the event. [sent-47, score-0.24]
24 This approach is transferable to Cyc’s semantic templates in which syntactic slots fill predicate arguments in the context of a specific syntactic frame. [sent-50, score-0.309]
25 4 Method The general method for creating semantic templates in Cyc requires creating Verb Class Frames and then using Cyc predicates and heuristic rules to create individual frames for each member verb. [sent-53, score-0.855]
26 1 OpenCyc The existing semantic templates are accessible through the ResearchCyc KB. [sent-55, score-0.159]
27 OpenCyc was used so as to minimize the effect of existing semantic frames on new frame creation. [sent-59, score-0.696]
28 2 Knowledge Representation The primary difficulty with integrating VerbNet frames into Cyc was overcoming differences in knowledge representation. [sent-62, score-0.481]
29 Cyc semantic templates reify events as an instance of a collection of events. [sent-63, score-0.198]
30 The following is a frame for the VerbNet class Give as presented in the Unified Verb Index4. [sent-67, score-0.195]
31 In Cyc the has Pos ses s ion relationship to and Recipient is represented with the predicates giver and givee. [sent-75, score-0.238]
32 Thus an individual VerbNet semantic predicate often has a many-toone mapping with Cyc predicates. [sent-78, score-0.146]
33 3 Predicates To account for representation differences, a single Cyc predicate was mapped to a unique combination of Verbnet predicate and thematic role (ie. [sent-80, score-0.22]
34 Though far from exhaustive, these hand mappings represent many frequently used predicates in VerbNet. [sent-83, score-0.16]
35 Because the mappings were not exhaustive, a safety net automatically catches predicates that haven’t been mapped. [sent-85, score-0.184]
36 The VerbNet predicates Cause and InReactionTo corresponded to the Cyc predicates performedBy doneBy, and cause s -Underspeci fied. [sent-86, score-0.262]
37 These predicates were selected whenever the VerbNet predicates occurred with a theme role that was the subject of the sentence. [sent-87, score-0.346]
38 The cause s -Underspeci fied predicate was used in frames whose time modifiers suggested that they were continuous states. [sent-90, score-0.563]
39 The predicates patientGeneric and pat ientGeneri c- Di rect were used when a predicate was not found for a required object or oblique object. [sent-91, score-0.289]
40 Some Cyc templates don’t have predicates that reference the event. [sent-92, score-0.207]
41 Most verb frames have an associated collection of events of which each use is an instance. [sent-98, score-0.605]
42 The associated collection of the class frame templates was automatically selected using the common link that both resources share with WordNet (Fellbaum, 1998). [sent-99, score-0.321]
43 To do this, the WordNet synsets of the member verbs for a class were matched with their Cyc-WordNet s ynonymousExte rnalConcept assertion. [sent-100, score-0.178]
44 The most general collection out of the list of viable collections was chosen as the general class frame collection. [sent-102, score-0.332]
45 While the most general collection was used for the class semantic frame, at the level of individual verb frames the specific synset denoted collection was substituted for the more general one when applicable. [sent-105, score-0.851]
46 The general class level collection was used in cases where no Cyc-WordNet-VerbNet link existed. [sent-108, score-0.134]
47 If no verb had a synset in Cyc, the general collection Situation was used. [sent-109, score-0.145]
48 5 Subcategorization Frames Each syntactic frame is a subcategorization frame or a subset of one. [sent-111, score-0.294]
49 Specific verb semantic templates were created by inferring that each member verb of a VerbNet class participated Again, collections in every template were taken in a class. [sent-118, score-0.455]
50 FIRE could then be queried for implied verb templates which became the final list of verb templates. [sent-122, score-0.252]
51 Subclasses contain verbs that take all of the syntactic formats of the main class plus additional frames that verbs in the main class cannot. [sent-125, score-0.733]
52 Verbs in a subclass inherit frames from their superordinate classes. [sent-126, score-0.515]
53 If no subclass member had a Cyc denotation, then the main class collection was used. [sent-129, score-0.187]
54 5 Results The end result of this process was the creation of 27,909 verb semantic template assertions for 5,050 different verbs. [sent-130, score-0.217]
55 This substantially increases the number of frames for ResearchCyc and creates frames for OpenCyc. [sent-131, score-0.962]
56 The first was to compare our frames with the 139 hand-checked VxC matches by hand. [sent-133, score-0.505]
57 Of the 139 frames from VxC, 81 were qualified as “good” matches, and 58 as “maybe” (Trumbo, 2006). [sent-134, score-0.481]
58 Since these frames already existed in Cyc and were hand matched we used them as the current gold standard for what a VerbNet frame translated into Cyc should look like. [sent-135, score-0.64]
59 First was whether the frame had as good a syntactic parse as the manual version. [sent-137, score-0.162]
60 This was defined as having predicates that addressed all syntactic roles in the sentence or, if not enough, as many as the VxC match. [sent-138, score-0.161]
61 Because framespecific predicates were not created on a large scale, a frame was not rejected for using general predicates. [sent-141, score-0.318]
62 First, the VxC mappings included frames in Cyc that only partially matched more syntactically robust VerbNet frames. [sent-143, score-0.566]
63 Our frames were only included if they matched the intended VerbNet syntactic frame. [sent-144, score-0.538]
64 Because of this some of our frames beat the VxC gold standard for syntactic completeness. [sent-145, score-0.511]
65 The VxC frames also included multiple similar senses for an individual verb. [sent-146, score-0.503]
66 Our verbs had one denotation per class or subclass. [sent-147, score-0.147]
67 Thus in some cases our frames failed not from over generalizing but because they were only meant to represent one meaning per class. [sent-148, score-0.505]
68 Since the strength of our approach lies in generating a near exhaustive list of syntactic frames and not multiple word senses, these kinds of failures are not necessarily representative of the success of the frames as a whole. [sent-149, score-1.038]
69 9%) of the correct frames having a more complete syntactic parse than the manually mapped frame. [sent-152, score-0.551]
70 8%) of the collection 64 rejected frames had a more complete parse than their manual counterparts. [sent-155, score-0.565]
71 1%) were as syntactically correct or better than the existing Cyc frame mapped to that VerbNet frame. [sent-157, score-0.242]
72 The second test compared the results of a natural language understanding system using either ResearchCyc alone or a version of ResearchCyc with our frames substituted for theirs. [sent-167, score-0.481]
73 The test corpus was 50 randomly selected example sentences from the VerbNet frame examples. [sent-168, score-0.132]
74 A parse was judged correct if it returned a verb frame for the central verb of the example sentence that either wholly or in combination with preposition frames addressed the syntactic constituents of the sentence with an acceptable collection and acceptable predicates. [sent-171, score-0.883]
75 ResearchCyc got sixteen out of 50 frames correct (32%). [sent-173, score-0.501]
76 Eleven frames (22%) did not return a template but did return a denotation to a Cyc collection. [sent-174, score-0.558]
77 Twelve verbs (24%) retuned nothing, while eleven (22%) returned frames that were either not the correct syntactic frame or were a different sense of the verb. [sent-175, score-0.759]
78 EA NLU running the VerbNet generated frames got 26 out of 50 (52%) frames correct. [sent-176, score-0.962]
79 Four generated frames (8%) were either not the correct syntactic frame or were for a different sense of the verb. [sent-179, score-0.663]
80 Five (10%) parses using the VerbNet generated correct frames that were labeled as noisy. [sent-181, score-0.501]
81 Noisy frames had duplicate predicates or more general predicates in addition to the specific ones. [sent-182, score-0.791]
82 The Hold frames separated out in the VxC test are an example of noisy frames. [sent-183, score-0.481]
83 None of these frames were syntactically incorrect or contradictory. [sent-184, score-0.51]
84 The redundant predicates arise because the predicate safety net had to be greedy. [sent-185, score-0.237]
85 This was in the interest of capturing more complex frames that may have multiple relations for the same thematic role in a sentence. [sent-186, score-0.517]
86 This evaluation is based on parser recall and frame semantic accuracy only. [sent-187, score-0.174]
87 As would be expected, adding more frames to the knowledge base did result in more parser retrievals and possible interpretations. [sent-188, score-0.502]
88 To improve predicate specificity, the next phase of research with these frames will be to implement predicate strengthening methods that move down the hierarchy to find more specific predicates to replace the generalized ones. [sent-190, score-0.829]
89 Thus in the future precision both in terms of frame retrieval and predicate specificity will be a vital metric for evaluating success. [sent-191, score-0.214]
90 6 Discussion As has been demonstrated in this approach and in previous research like Curtis et al’s (2009) TextLearner, Cyc provides powerful reasoning capabilities that can be used to successfully infer more specific information from general existing facts. [sent-192, score-0.156]
91 While many of the frames are general, they provide a solid foundation for further research. [sent-195, score-0.481]
92 As they are now, the added 27,909 frames increase the language capabilities of OpenCyc which previously had none. [sent-196, score-0.506]
93 However, with 35% of frames in the VxC comparison and 16% in the parse test failing because of collections, and 10. [sent-200, score-0.481]
94 8% of the VxC comparison set and 10% of correct parses classified as noisy, these frames are not as precise as the existing frames. [sent-201, score-0.542]
95 The goal of these frames is not necessarily to replace the existing frames, but rather to extend coverage and provide a platform for further development whether by hand or through automatic methods. [sent-202, score-0.544]
96 Additionally, there is a tradeoff between the number of frames covered and efficiency of disambiguation. [sent-206, score-0.481]
97 More frame choices make it harder for parsers to choose the correct frame, but it will hopefully improve their handling of more complex sentence structures. [sent-207, score-0.152]
98 The class based approach makes it easy to separate verbs by types, such as verbs that relate to mechanical processes or emotion verbs. [sent-209, score-0.159]
99 One could use classes of frames to strengthen specific areas of parsing while choosing not to take verbs from a class covering a domain that the parser already performs strongly in. [sent-210, score-0.641]
100 Thus an approach to computational verb semantic representation that is rooted in classes can take advantage of modern reasoning sources like Cyc to efficiently create semantic knowledge. [sent-213, score-0.222]
wordName wordTfidf (topN-words)
[('cyc', 0.549), ('frames', 0.481), ('verbnet', 0.402), ('vxc', 0.178), ('researchcyc', 0.164), ('frame', 0.132), ('predicates', 0.131), ('curtis', 0.119), ('opencyc', 0.119), ('textlearner', 0.104), ('ion', 0.083), ('predicate', 0.082), ('act', 0.081), ('templates', 0.076), ('verb', 0.074), ('class', 0.063), ('assertions', 0.06), ('recipient', 0.06), ('cabral', 0.059), ('forbus', 0.059), ('performedby', 0.059), ('theme', 0.058), ('kipper', 0.056), ('obj', 0.052), ('collection', 0.05), ('al', 0.048), ('verbs', 0.048), ('collections', 0.045), ('trumbo', 0.045), ('reasoning', 0.042), ('semantic', 0.042), ('template', 0.041), ('existing', 0.041), ('member', 0.04), ('object', 0.04), ('matuszek', 0.039), ('thematic', 0.036), ('nlu', 0.036), ('denotation', 0.036), ('pat', 0.036), ('agent', 0.035), ('fire', 0.035), ('subclass', 0.034), ('rejected', 0.034), ('ontology', 0.033), ('crouch', 0.032), ('syntactic', 0.03), ('kb', 0.03), ('aaai', 0.03), ('cycl', 0.03), ('doneby', 0.03), ('frompo', 0.03), ('givee', 0.03), ('iontrans', 0.03), ('mostek', 0.03), ('obl', 0.03), ('ppcompframefn', 0.03), ('ramachandran', 0.03), ('reify', 0.03), ('ricdi', 0.03), ('tomai', 0.03), ('syntactically', 0.029), ('mappings', 0.029), ('levin', 0.029), ('wordnet', 0.029), ('became', 0.028), ('specific', 0.027), ('matched', 0.027), ('definitional', 0.026), ('eleven', 0.026), ('strengthening', 0.026), ('subject', 0.026), ('capabilities', 0.025), ('connections', 0.025), ('exhaustive', 0.025), ('kenneth', 0.024), ('safety', 0.024), ('giver', 0.024), ('baxter', 0.024), ('subclasses', 0.024), ('matches', 0.024), ('failed', 0.024), ('coverage', 0.022), ('twelve', 0.022), ('possession', 0.022), ('classes', 0.022), ('returned', 0.022), ('individual', 0.022), ('fill', 0.022), ('creating', 0.021), ('failures', 0.021), ('northwestern', 0.021), ('karin', 0.021), ('expressiveness', 0.021), ('usable', 0.021), ('base', 0.021), ('general', 0.021), ('correct', 0.02), ('spring', 0.02), ('mapped', 0.02)]
simIndex simValue paperId paperTitle
same-paper 1 1.0000002 108 acl-2010-Expanding Verb Coverage in Cyc with VerbNet
Author: Clifton McFate
Abstract: A robust dictionary of semantic frames is an essential element of natural language understanding systems that use ontologies. However, creating lexical resources that accurately capture semantic representations en masse is a persistent problem. Where the sheer amount of content makes hand creation inefficient, computerized approaches often suffer from over generality and difficulty with sense disambiguation. This paper describes a semi-automatic method to create verb semantic frames in the Cyc ontology by converting the information contained in VerbNet into a Cyc usable format. This method captures the differences in meaning between types of verbs, and uses existing connections between WordNet, VerbNet, and Cyc to specify distinctions between individual verbs when available. This method provides 27,909 frames to OpenCyc which currently has none and can be used to extend ResearchCyc as well. We show that these frames lead to a 20% increase in sample sentences parsed over the Research Cyc verb lexicon. 1
2 0.12689292 41 acl-2010-Automatic Selectional Preference Acquisition for Latin Verbs
Author: Barbara McGillivray
Abstract: We present a system that automatically induces Selectional Preferences (SPs) for Latin verbs from two treebanks by using Latin WordNet. Our method overcomes some of the problems connected with data sparseness and the small size of the input corpora. We also suggest a way to evaluate the acquired SPs on unseen events extracted from other Latin corpora.
3 0.11773949 85 acl-2010-Detecting Experiences from Weblogs
Author: Keun Chan Park ; Yoonjae Jeong ; Sung Hyon Myaeng
Abstract: Weblogs are a source of human activity knowledge comprising valuable information such as facts, opinions and personal experiences. In this paper, we propose a method for mining personal experiences from a large set of weblogs. We define experience as knowledge embedded in a collection of activities or events which an individual or group has actually undergone. Based on an observation that experience-revealing sentences have a certain linguistic style, we formulate the problem of detecting experience as a classification task using various features including tense, mood, aspect, modality, experiencer, and verb classes. We also present an activity verb lexicon construction method based on theories of lexical semantics. Our results demonstrate that the activity verb lexicon plays a pivotal role among selected features in the classification perfor- , mance and shows that our proposed method outperforms the baseline significantly.
4 0.10325532 120 acl-2010-Fully Unsupervised Core-Adjunct Argument Classification
Author: Omri Abend ; Ari Rappoport
Abstract: The core-adjunct argument distinction is a basic one in the theory of argument structure. The task of distinguishing between the two has strong relations to various basic NLP tasks such as syntactic parsing, semantic role labeling and subcategorization acquisition. This paper presents a novel unsupervised algorithm for the task that uses no supervised models, utilizing instead state-of-the-art syntactic induction algorithms. This is the first work to tackle this task in a fully unsupervised scenario.
5 0.088465139 121 acl-2010-Generating Entailment Rules from FrameNet
Author: Roni Ben Aharon ; Idan Szpektor ; Ido Dagan
Abstract: Idan Szpektor Ido Dagan Yahoo! Research Department of Computer Science Haifa, Israel Bar-Ilan University idan @ yahoo- inc .com Ramat Gan, Israel dagan @ c s .biu . ac . i l FrameNet is a manually constructed database based on Frame Semantics. It models the semantic Many NLP tasks need accurate knowledge for semantic inference. To this end, mostly WordNet is utilized. Yet WordNet is limited, especially for inference be- tween predicates. To help filling this gap, we present an algorithm that generates inference rules between predicates from FrameNet. Our experiment shows that the novel resource is effective and complements WordNet in terms of rule coverage.
6 0.082510263 49 acl-2010-Beyond NomBank: A Study of Implicit Arguments for Nominal Predicates
7 0.073028132 130 acl-2010-Hard Constraints for Grammatical Function Labelling
8 0.06958171 153 acl-2010-Joint Syntactic and Semantic Parsing of Chinese
9 0.065111749 94 acl-2010-Edit Tree Distance Alignments for Semantic Role Labelling
10 0.063945763 238 acl-2010-Towards Open-Domain Semantic Role Labeling
11 0.06192454 158 acl-2010-Latent Variable Models of Selectional Preference
12 0.06111028 17 acl-2010-A Structured Model for Joint Learning of Argument Roles and Predicate Senses
13 0.060476936 258 acl-2010-Weakly Supervised Learning of Presupposition Relations between Verbs
14 0.058822762 184 acl-2010-Open-Domain Semantic Role Labeling by Modeling Word Spans
15 0.056615558 139 acl-2010-Identifying Generic Noun Phrases
16 0.052694045 6 acl-2010-A Game-Theoretic Model of Metaphorical Bargaining
17 0.051180478 247 acl-2010-Unsupervised Event Coreference Resolution with Rich Linguistic Features
18 0.050952222 198 acl-2010-Predicate Argument Structure Analysis Using Transformation Based Learning
19 0.048724514 216 acl-2010-Starting from Scratch in Semantic Role Labeling
20 0.047759734 168 acl-2010-Learning to Follow Navigational Directions
topicId topicWeight
[(0, -0.116), (1, 0.083), (2, 0.078), (3, 0.005), (4, 0.05), (5, 0.003), (6, -0.033), (7, 0.034), (8, -0.022), (9, -0.058), (10, 0.037), (11, 0.042), (12, -0.023), (13, 0.038), (14, 0.04), (15, 0.008), (16, 0.055), (17, 0.079), (18, 0.046), (19, 0.052), (20, 0.009), (21, 0.005), (22, 0.037), (23, -0.08), (24, 0.08), (25, -0.058), (26, -0.035), (27, 0.026), (28, 0.058), (29, -0.084), (30, 0.115), (31, -0.06), (32, 0.109), (33, -0.058), (34, 0.103), (35, -0.039), (36, -0.072), (37, 0.061), (38, 0.062), (39, -0.073), (40, -0.021), (41, 0.042), (42, -0.031), (43, 0.02), (44, 0.001), (45, -0.037), (46, -0.209), (47, 0.064), (48, -0.042), (49, -0.092)]
simIndex simValue paperId paperTitle
same-paper 1 0.95476073 108 acl-2010-Expanding Verb Coverage in Cyc with VerbNet
Author: Clifton McFate
Abstract: A robust dictionary of semantic frames is an essential element of natural language understanding systems that use ontologies. However, creating lexical resources that accurately capture semantic representations en masse is a persistent problem. Where the sheer amount of content makes hand creation inefficient, computerized approaches often suffer from over generality and difficulty with sense disambiguation. This paper describes a semi-automatic method to create verb semantic frames in the Cyc ontology by converting the information contained in VerbNet into a Cyc usable format. This method captures the differences in meaning between types of verbs, and uses existing connections between WordNet, VerbNet, and Cyc to specify distinctions between individual verbs when available. This method provides 27,909 frames to OpenCyc which currently has none and can be used to extend ResearchCyc as well. We show that these frames lead to a 20% increase in sample sentences parsed over the Research Cyc verb lexicon. 1
2 0.73590761 85 acl-2010-Detecting Experiences from Weblogs
Author: Keun Chan Park ; Yoonjae Jeong ; Sung Hyon Myaeng
Abstract: Weblogs are a source of human activity knowledge comprising valuable information such as facts, opinions and personal experiences. In this paper, we propose a method for mining personal experiences from a large set of weblogs. We define experience as knowledge embedded in a collection of activities or events which an individual or group has actually undergone. Based on an observation that experience-revealing sentences have a certain linguistic style, we formulate the problem of detecting experience as a classification task using various features including tense, mood, aspect, modality, experiencer, and verb classes. We also present an activity verb lexicon construction method based on theories of lexical semantics. Our results demonstrate that the activity verb lexicon plays a pivotal role among selected features in the classification perfor- , mance and shows that our proposed method outperforms the baseline significantly.
3 0.69095135 41 acl-2010-Automatic Selectional Preference Acquisition for Latin Verbs
Author: Barbara McGillivray
Abstract: We present a system that automatically induces Selectional Preferences (SPs) for Latin verbs from two treebanks by using Latin WordNet. Our method overcomes some of the problems connected with data sparseness and the small size of the input corpora. We also suggest a way to evaluate the acquired SPs on unseen events extracted from other Latin corpora.
4 0.61824822 126 acl-2010-GernEdiT - The GermaNet Editing Tool
Author: Verena Henrich ; Erhard Hinrichs
Abstract: GernEdiT (short for: GermaNet Editing Tool) offers a graphical interface for the lexicographers and developers of GermaNet to access and modify the underlying GermaNet resource. GermaNet is a lexical-semantic wordnet that is modeled after the Princeton WordNet for English. The traditional lexicographic development of GermaNet was error prone and time-consuming, mainly due to a complex underlying data format and no opportunity of automatic consistency checks. GernEdiT replaces the earlier development by a more userfriendly tool, which facilitates automatic checking of internal consistency and correctness of the linguistic resource. This paper pre- sents all these core functionalities of GernEdiT along with details about its usage and usability. 1
5 0.54254383 258 acl-2010-Weakly Supervised Learning of Presupposition Relations between Verbs
Author: Galina Tremper
Abstract: Presupposition relations between verbs are not very well covered in existing lexical semantic resources. We propose a weakly supervised algorithm for learning presupposition relations between verbs that distinguishes five semantic relations: presupposition, entailment, temporal inclusion, antonymy and other/no relation. We start with a number of seed verb pairs selected manually for each semantic relation and classify unseen verb pairs. Our algorithm achieves an overall accuracy of 36% for type-based classification.
6 0.48673004 121 acl-2010-Generating Entailment Rules from FrameNet
7 0.42478436 120 acl-2010-Fully Unsupervised Core-Adjunct Argument Classification
8 0.41759148 238 acl-2010-Towards Open-Domain Semantic Role Labeling
9 0.38640693 139 acl-2010-Identifying Generic Noun Phrases
10 0.36791047 216 acl-2010-Starting from Scratch in Semantic Role Labeling
11 0.35852906 49 acl-2010-Beyond NomBank: A Study of Implicit Arguments for Nominal Predicates
12 0.35325491 247 acl-2010-Unsupervised Event Coreference Resolution with Rich Linguistic Features
13 0.35285702 148 acl-2010-Improving the Use of Pseudo-Words for Evaluating Selectional Preferences
14 0.34842628 111 acl-2010-Extracting Sequences from the Web
15 0.34657419 165 acl-2010-Learning Script Knowledge with Web Experiments
16 0.32101098 128 acl-2010-Grammar Prototyping and Testing with the LinGO Grammar Matrix Customization System
17 0.31728902 35 acl-2010-Automated Planning for Situated Natural Language Generation
18 0.31641167 225 acl-2010-Temporal Information Processing of a New Language: Fast Porting with Minimal Resources
19 0.30807126 6 acl-2010-A Game-Theoretic Model of Metaphorical Bargaining
20 0.30637273 235 acl-2010-Tools for Multilingual Grammar-Based Translation on the Web
topicId topicWeight
[(14, 0.015), (25, 0.062), (39, 0.013), (42, 0.027), (44, 0.013), (59, 0.081), (73, 0.046), (76, 0.029), (78, 0.061), (81, 0.357), (83, 0.063), (84, 0.04), (98, 0.066)]
simIndex simValue paperId paperTitle
same-paper 1 0.73201257 108 acl-2010-Expanding Verb Coverage in Cyc with VerbNet
Author: Clifton McFate
Abstract: A robust dictionary of semantic frames is an essential element of natural language understanding systems that use ontologies. However, creating lexical resources that accurately capture semantic representations en masse is a persistent problem. Where the sheer amount of content makes hand creation inefficient, computerized approaches often suffer from over generality and difficulty with sense disambiguation. This paper describes a semi-automatic method to create verb semantic frames in the Cyc ontology by converting the information contained in VerbNet into a Cyc usable format. This method captures the differences in meaning between types of verbs, and uses existing connections between WordNet, VerbNet, and Cyc to specify distinctions between individual verbs when available. This method provides 27,909 frames to OpenCyc which currently has none and can be used to extend ResearchCyc as well. We show that these frames lead to a 20% increase in sample sentences parsed over the Research Cyc verb lexicon. 1
2 0.65759957 183 acl-2010-Online Generation of Locality Sensitive Hash Signatures
Author: Benjamin Van Durme ; Ashwin Lall
Abstract: Motivated by the recent interest in streaming algorithms for processing large text collections, we revisit the work of Ravichandran et al. (2005) on using the Locality Sensitive Hash (LSH) method of Charikar (2002) to enable fast, approximate comparisons of vector cosine similarity. For the common case of feature updates being additive over a data stream, we show that LSH signatures can be maintained online, without additional approximation error, and with lower memory requirements than when using the standard offline technique.
3 0.59358561 252 acl-2010-Using Parse Features for Preposition Selection and Error Detection
Author: Joel Tetreault ; Jennifer Foster ; Martin Chodorow
Abstract: Jennifer Foster NCLT Dublin City University Ireland j fo st er@ comput ing . dcu . ie Martin Chodorow Hunter College of CUNY New York, NY, USA martin . chodorow @hunter . cuny . edu We recreate a state-of-the-art preposition usage system (Tetreault and Chodorow (2008), henceWe evaluate the effect of adding parse features to a leading model of preposition us- age. Results show a significant improvement in the preposition selection task on native speaker text and a modest increment in precision and recall in an ESL error detection task. Analysis of the parser output indicates that it is robust enough in the face of noisy non-native writing to extract useful information.
4 0.39904246 158 acl-2010-Latent Variable Models of Selectional Preference
Author: Diarmuid O Seaghdha
Abstract: This paper describes the application of so-called topic models to selectional preference induction. Three models related to Latent Dirichlet Allocation, a proven method for modelling document-word cooccurrences, are presented and evaluated on datasets of human plausibility judgements. Compared to previously proposed techniques, these models perform very competitively, especially for infrequent predicate-argument combinations where they exceed the quality of Web-scale predictions while using relatively little data.
5 0.39244053 70 acl-2010-Contextualizing Semantic Representations Using Syntactically Enriched Vector Models
Author: Stefan Thater ; Hagen Furstenau ; Manfred Pinkal
Abstract: We present a syntactically enriched vector model that supports the computation of contextualized semantic representations in a quasi compositional fashion. It employs a systematic combination of first- and second-order context vectors. We apply our model to two different tasks and show that (i) it substantially outperforms previous work on a paraphrase ranking task, and (ii) achieves promising results on a wordsense similarity task; to our knowledge, it is the first time that an unsupervised method has been applied to this task.
6 0.39153221 120 acl-2010-Fully Unsupervised Core-Adjunct Argument Classification
7 0.3907066 211 acl-2010-Simple, Accurate Parsing with an All-Fragments Grammar
8 0.38937041 130 acl-2010-Hard Constraints for Grammatical Function Labelling
9 0.38934889 17 acl-2010-A Structured Model for Joint Learning of Argument Roles and Predicate Senses
10 0.38570717 10 acl-2010-A Latent Dirichlet Allocation Method for Selectional Preferences
11 0.38366008 23 acl-2010-Accurate Context-Free Parsing with Combinatory Categorial Grammar
12 0.38333362 160 acl-2010-Learning Arguments and Supertypes of Semantic Relations Using Recursive Patterns
13 0.38293442 169 acl-2010-Learning to Translate with Source and Target Syntax
14 0.38292742 59 acl-2010-Cognitively Plausible Models of Human Language Processing
15 0.38209406 71 acl-2010-Convolution Kernel over Packed Parse Forest
16 0.38206181 248 acl-2010-Unsupervised Ontology Induction from Text
17 0.37945345 65 acl-2010-Complexity Metrics in an Incremental Right-Corner Parser
18 0.37914714 153 acl-2010-Joint Syntactic and Semantic Parsing of Chinese
19 0.37913448 139 acl-2010-Identifying Generic Noun Phrases
20 0.37897596 162 acl-2010-Learning Common Grammar from Multilingual Corpus