Presentation is loading. Please wait.

Presentation is loading. Please wait.

University of Texas at Austin Machine Learning Group Department of Computer Sciences University of Texas at Austin Learning for Semantic Parsing of Natural.

Similar presentations


Presentation on theme: "University of Texas at Austin Machine Learning Group Department of Computer Sciences University of Texas at Austin Learning for Semantic Parsing of Natural."— Presentation transcript:

1 University of Texas at Austin Machine Learning Group Department of Computer Sciences University of Texas at Austin Learning for Semantic Parsing of Natural Language July 31, 2005 Raymond J. Mooney Ruifang Ge, Rohit Kate, Yuk Wah Wong John Zelle, Cynthia Thompson

2 2 Syntactic Natural Language Learning Most computational research in natural-language learning has addressed “low-level” syntactic processing. –Morphology (e.g. past-tense generation) –Part-of-speech tagging –Shallow syntactic parsing –Syntactic parsing

3 3 Semantic Natural Language Learning Learning for semantic analysis has been restricted to relatively small, isolated tasks. –Word sense disambiguation (e.g. SENSEVAL) –Semantic role assignment (determining agent, patient, instrument, etc., e.g. FrameNet, PropBank) –Information extraction

4 4 Semantic Parsing A semantic parser maps a natural-language sentence to a complete, detailed semantic representation (logical form). For many applications, the desired output is immediately executable by another program. Two application domains: –CLang: RoboCup Coach Language –GeoQuery: A Database Query Application

5 5 CLang: RoboCup Coach Language In RoboCup Coach competition teams compete to coach simulated players The coaching instructions are given in a formal language called CLang Simulated soccer field Coach CLang If the ball is in our penalty area, then all our players except player 4 should stay in our half. ((bpos (penalty-area our)) (do (player-except our{4}) (pos (half our))) Semantic Parsing

6 6 GeoQuery: A Database Query Application Query application for U.S. geography database containing about 800 facts [Zelle & Mooney, 1996] User How many cities are there in the US? Query answer(A, count(B, (city(B), loc(B, C), const(C, countryid(USA))),A)) Semantic Parsing

7 7 Learning Semantic Parsers Manually programming robust semantic parsers is difficult due to the complexity of the task. Semantic parsers can be learned automatically from sentences paired with their logical form. NL  LF Training Exs Semantic-Parser Learner Natural Language Logical Form Semantic Parser

8 8 Engineering Motivation Most computational language-learning research strives for broad coverage while sacrificing depth. –“Scaling up by dumbing down” Realistic semantic parsing currently entails domain dependence. Domain-dependent natural-language interfaces have a large potential market. Learning makes developing specific applications more tractable. Training corpora can be easily developed by tagging existing corpora of formal statements with natural-language glosses.

9 9 Cognitive Science Motivation Most natural-language learning methods require supervised training data that is not available to a child. –General lack of negative feedback on grammar. –No POS-tagged or treebank data. Assuming a child can infer the likely meaning of an utterance from context, NL  LF pairs are more cognitively plausible training data.

10 10 Our Semantic-Parser Learners CHILL+WOLFIE (Zelle & Mooney, 1996; Thompson & Mooney, 1999, 2003) –Separates parser-learning and semantic-lexicon learning. –Learns a deterministic parser using ILP techniques. SILT (Kate, Wong & Mooney, 2005) –Learns symbolic transformation rules for mapping directly from NL to LF. SCISSOR (Ge & Mooney, 2005) –Integrates semantic interpretation into a statistical syntactic parser (Collins model-3).

11 11 CHILL (Zelle & Mooney, 1992-96) Semantic parser acquisition system using Inductive Logic Programming (ILP) to induce a parser written in Prolog. Starts with a parsing “shell” written in Prolog and learns to control the operators of this parser to produce the given I/O pairs. Requires a semantic lexicon, which for each word gives one or more possible semantic representations. Parser must disambiguate words, introduce proper semantic representations for each, and then put them together in the right way to produce a proper representation of the sentence.

12 12 CHILL Example U.S. Geographical database –Sample training pair Cuál es el capital del estado con la población más grande? answer(C, (capital(S,C), largest(P, (state(S), population(S,P))))) –Sample semantic lexicon cuál : answer(_,_) capital: capital(_,_) estado: state(_) más grande: largest(_,_) población: population(_,_)

13 13 WOLFIE (Thompson & Mooney, 1995-1999) Learns a semantic lexicon for CHILL from the same corpus of semantically annotated sentences. Determines hypotheses for word meanings by finding largest isomorphic common subgraphs shared by meanings of sentences in which the word appears. Uses a greedy-covering style algorithm to learn a small lexicon sufficient to allow compositional construction of the correct representation from the words in a sentence.

14 14 WOLFIE + CHILL Semantic Parser Acquisition NL  LF Training Exs WOLFIE Lexicon Learner Natural Language Logical Form Semantic Lexicon Semantic Parser CHILL Parser Learner

15 15 Semantic Parsing using Transformation Rules S ILT (Semantic Interpretation by Learning Transformations) Uses pattern-based transformation rules which map natural language phrases to formal language constructs Transformation rules are repeatedly applied to the sentence to construct its formal language expression

16 16 Formal Language Grammar NL: If our player 4 has the ball, our player 4 should shoot. CLang: ((bowner our {4}) (do our {4} shoot)) CLang Parse: Non-terminals: RULE, CONDITION, ACTION… Terminals: bowner, our, 4… Productions: RULE  CONDITION DIRECTIVE DIRECTIVE  do TEAM UNUM ACTION ACTION  shoot RULE CONDITIONDIRECTIVE doTEAMUNUMACTIONbownerTEAMUNUM our 4 4shoot

17 17 Transformation Rule Representation Rule has two components: a natural language pattern and an associated formal language template Two versions of S ILT : –String-based rules: used to convert natural language sentence directly to formal language –Tree-based rules: used to convert syntactic tree to formal language String-patternTEAM UNUM has [1] ball Template CONDITION  (bowner TEAM {UNUM}) Tree- pattern Template CONDITION  (bowner TEAM {UNUM}) NP VP VBZ NP DT NN theball has TEAMUNUM S word gap

18 18 Example of Semantic Parsing If our player 4 has the ball, our player 4 should shoot. our TEAM  our player 4 UNUM  4 shoot ACTION  shoot TEAM UNUM has [1] ball CONDITION  (bowner TEAM {UNUM}) TEAM UNUM should ACTION DIRECTIVE  (do TEAM {UNUM} ACTION) If CONDITION, DIRECTIVE. RULE  (CONDITION DIRECTIVE)

19 19 Example of Semantic Parsing If player 4 has the ball, player 4 should shoot. our TEAM  our player 4 UNUM  4 shoot ACTION  shoot TEAM UNUM has [1] ball CONDITION  (bowner TEAM {UNUM}) TEAM UNUM should ACTION DIRECTIVE  (do TEAM {UNUM} ACTION) If CONDITION, DIRECTIVE. RULE  (CONDITION DIRECTIVE) our TEAM our TEAM our

20 20 Example of Semantic Parsing If player 4 has the ball, player 4 should shoot. our TEAM  our player 4 UNUM  4 shoot ACTION  shoot TEAM UNUM has [1] ball CONDITION  (bowner TEAM {UNUM}) TEAM UNUM should ACTION DIRECTIVE  (do TEAM {UNUM} ACTION) If CONDITION, DIRECTIVE. RULE  (CONDITION DIRECTIVE) TEAM our TEAM our

21 21 Example of Semantic Parsing If has the ball, should shoot. our TEAM  our player 4 UNUM  4 shoot ACTION  shoot TEAM UNUM has [1] ball CONDITION  (bowner TEAM {UNUM}) TEAM UNUM should ACTION DIRECTIVE  (do TEAM {UNUM} ACTION) If CONDITION, DIRECTIVE. RULE  (CONDITION DIRECTIVE) TEAM our TEAM our player 4 UNUM 4 4

22 22 Example of Semantic Parsing If has the ball, should shoot. our TEAM  our player 4 UNUM  4 shoot ACTION  shoot TEAM UNUM has [1] ball CONDITION  (bowner TEAM {UNUM}) TEAM UNUM should ACTION DIRECTIVE  (do TEAM {UNUM} ACTION) If CONDITION, DIRECTIVE. RULE  (CONDITION DIRECTIVE) TEAM our TEAM our UNUM 4 4

23 23 ACTION shoot Example of Semantic Parsing If has the ball, should. our TEAM  our player 4 UNUM  4 shoot ACTION  shoot TEAM UNUM has [1] ball CONDITION  (bowner TEAM {UNUM}) TEAM UNUM should ACTION DIRECTIVE  (do TEAM {UNUM} ACTION) If CONDITION, DIRECTIVE. RULE  (CONDITION DIRECTIVE) TEAM our TEAM our UNUM 4 4 shoot

24 24 Example of Semantic Parsing If has the ball, should. our TEAM  our player 4 UNUM  4 shoot ACTION  shoot TEAM UNUM has [1] ball CONDITION  (bowner TEAM {UNUM}) TEAM UNUM should ACTION DIRECTIVE  (do TEAM {UNUM} ACTION) If CONDITION, DIRECTIVE. RULE  (CONDITION DIRECTIVE) TEAM our TEAM our UNUM 4 4 ACTION shoot

25 25 Example of Semantic Parsing If, should. our TEAM  our player 4 UNUM  4 shoot ACTION  shoot TEAM UNUM has [1] ball CONDITION  (bowner TEAM {UNUM}) TEAM UNUM should ACTION DIRECTIVE  (do TEAM {UNUM} ACTION) If CONDITION, DIRECTIVE. RULE  (CONDITION DIRECTIVE) TEAM our TEAM our UNUM 4 4 ACTION shoot has the ball CONDITION (bowner our {4})

26 26 Example of Semantic Parsing If, should. our TEAM  our player 4 UNUM  4 shoot ACTION  shoot TEAM UNUM has [1] ball CONDITION  (bowner TEAM {UNUM}) TEAM UNUM should ACTION DIRECTIVE  (do TEAM {UNUM} ACTION) If CONDITION, DIRECTIVE. RULE  (CONDITION DIRECTIVE) TEAM our UNUM 4 ACTION shoot CONDITION (bowner our {4})

27 27 Example of Semantic Parsing If,. our TEAM  our player 4 UNUM  4 shoot ACTION  shoot TEAM UNUM has [1] ball CONDITION  (bowner TEAM {UNUM}) TEAM UNUM should ACTION DIRECTIVE  (do TEAM {UNUM} ACTION) If CONDITION, DIRECTIVE. RULE  (CONDITION DIRECTIVE) TEAM our UNUM 4 ACTION shoot CONDITION (bowner our {4}) should DIRECTIVE (do our {4} shoot)

28 28 Example of Semantic Parsing If,. our TEAM  our player 4 UNUM  4 shoot ACTION  shoot TEAM UNUM has [1] ball CONDITION  (bowner TEAM {UNUM}) TEAM UNUM should ACTION DIRECTIVE  (do TEAM {UNUM} ACTION) If CONDITION, DIRECTIVE. RULE  (CONDITION DIRECTIVE) CONDITION (bowner our {4}) DIRECTIVE (do our {4} shoot)

29 29 Example of Semantic Parsing If,. our TEAM  our player 4 UNUM  4 shoot ACTION  shoot TEAM UNUM has [1] ball CONDITION  (bowner TEAM {UNUM}) TEAM UNUM should ACTION DIRECTIVE  (do TEAM {UNUM} ACTION) If CONDITION, DIRECTIVE. RULE  (CONDITION DIRECTIVE) CONDITION (bowner our {4}) DIRECTIVE (do our {4} shoot) RULE ((bowner our {4}) (do our {4} shoot))

30 30 Learning Transformation Rules S ILT induces rules from a corpora of NL sentences paired with their formal representations Patterns are learned for each production by bottom-up rule learning For every production: –Call those sentences positives whose formal representations’ parses use that production –Call the remaining sentences negatives

31 31 Rule Learning for a Production S ILT applies greedy-covering, bottom-up rule induction method that repeatedly generalizes positives until they start covering negatives The ball is in REGION, our player 7 is in REGION and no opponent is around our player 7 within 1.5 distance. If the ball is in REGION and not in REGION then player 3 should intercept the ball. During normal play if the ball is in the REGION then player 7, 9 and 11 should dribble the ball to the REGION. When the play mode is normal and the ball is in the REGION then our player 2 should pass the ball to the REGION. All players except the goalie should pass the ball to REGION if it is in RP18. If the ball is inside rectangle ( -54, -36, 0, 36 ) then player 10 should position itself at REGION with a ball attraction of REGION. Player 2 should pass the ball to REGION if it is in REGION. If our player 6 has the ball then he should take a shot on goal. If player 4 has the ball, it should pass the ball to player 2 or 10. If the condition DR5C3 is true, then player 2, 3, 7 and 8 should pass the ball to player 3. During play on, if players 6, 7 or 8 is in REGION, they should pass the ball to players 9, 10 or 11. If "Clear_Condition", players 2, 3, 7 or 5 should clear the ball REGION. If it is before the kick off, after our goal or after the opponent's goal, position player 3 at REGION. If the condition MDR4C9 is met, then players 4-6 should pass the ball to player 9. If Pass_11 then player 11 should pass to player 9 and no one else. CONDITION (bpos REGION) positives negatives

32 32 Generalization of String Patterns ACTION  (pos REGION) Pattern 1: Always position player UNUM at REGION. Pattern 2: Whenever the ball is in REGION, position player UNUM near the REGION. Find the highest scoring common subsequence:

33 33 Generalization of String Patterns ACTION  (pos REGION) Pattern 1: Always position player UNUM at REGION. Pattern 2: Whenever the ball is in REGION, position player UNUM near the REGION. Find the highest scoring common subsequence: Generalization: position player UNUM [2] REGION.

34 34 Generalization of Tree Patterns REGION  (penalty-area TEAM) Pattern 1: Pattern 2 Find common subgraphs. NP NN TEAMPOS penalty box ’s’s NP PRP$NN TEAM penaltyarea

35 35 Generalization of Tree Patterns REGION  (penalty-area TEAM) Pattern 1: Pattern 2 Find common subgraphs. NP NN TEAMPOS penalty box ’s’s NP PRP$NN TEAM penaltyarea NP TEAM NN penalty NN Generalization: *

36 36 Rule Learning for a Production If our player 6 has the ball then he should take a shot on goal. If player 4 has the ball, it should pass the ball to player 2 or 10. If the condition DR5C3 is true, then player 2, 3, 7 and 8 should pass the ball to player 3. During play on, if players 6, 7 or 8 is in REGION, they should pass the ball to players 9, 10 or 11. If "Clear_Condition", players 2, 3, 7 or 5 should clear the ball REGION. If it is before the kick off, after our goal or after the opponent's goal, position player 3 at REGION. If the condition MDR4C9 is met, then players 4-6 should pass the ball to player 9. If Pass_11 then player 11 should pass to player 9 and no one else. CONDITION  (bpos REGION) positives negatives Bottom-up Rule Learner ball is [2] REGION CONDITION  (bpos REGION) it is in REGION CONDITION  (bpos REGION) The ball is in REGION, our player 7 is in REGION and no opponent is around our player 7 within 1.5 distance. If the ball is in REGION and not in REGION then player 3 should intercept the ball. During normal play if the ball is in the REGION then player 7, 9 and 11 should dribble the ball to the REGION. When the play mode is normal and the ball is in the REGION then our player 2 should pass the ball to the REGION. All players except the goalie should pass the ball to REGION if it is in REGION. If the ball is inside REGION then player 10 should position itself at REGION with a ball attraction of REGION. Player 2 should pass the ball to REGION if it is in REGION.

37 37 Rule Learning for a Production If our player 6 has the ball then he should take a shot on goal. If player 4 has the ball, it should pass the ball to player 2 or 10. If the condition DR5C3 is true, then player 2, 3, 7 and 8 should pass the ball to player 3. During play on, if players 6, 7 or 8 is in REGION, they should pass the ball to players 9, 10 or 11. If "Clear_Condition", players 2, 3, 7 or 5 should clear the ball REGION. If it is before the kick off, after our goal or after the opponent's goal, position player 3 at REGION. If the condition MDR4C9 is met, then players 4-6 should pass the ball to player 9. If Pass_11 then player 11 should pass to player 9 and no one else. CONDITION  (bpos REGION) positives negatives Bottom-up Rule Learner ball is [2] REGION CONDITION  (bpos REGION) it is in REGION CONDITION  (bpos REGION) The CONDITION, our player 7 is in REGION and no opponent is around our player 7 within 1.5 distance. If the CONDITION and not in REGION then player 3 should intercept the ball. During normal play if the CONDITION then player 7, 9 and 11 should dribble the ball to the REGION. When the play mode is normal and the CONDITION then our player 2 should pass the ball to the REGION. All players except the goalie should pass the ball to REGION if CONDITION. If the CONDITION then player 10 should position itself at REGION with a ball attraction of REGION. Player 2 should pass the ball to REGION if CONDITION.

38 38 Rule Learning for All Productions Transformation rules for productions should cooperate globally to generate complete semantic parses Redundantly cover every positive example by β = 5 best rules Find the subset of these rules which best cooperate to generate complete semantic parses on the training data coverage accuracy

39 39 Based on a fairly standard approach to compositional semantics [Jurafsky and Martin, 2000] A statistical parser is used to generate a semantically augmented parse tree (SAPT) –Augment Collins’ head-driven model 2 (Bikel’s implementation, 2004) to incorporate semantic labels Translate SAPT into a complete formal meaning representation (MR) SCISSOR : Semantic Composition that Integrates Syntax and Semantics to get Optimal Representations MR: bowner(player(our,2)) ourplayer2has theball PRP$-teamNN-playerCD-unumVB-bowner DT-nullNN-null NP-null VP-bownerNP-player S-bowner ourplayer2has theball PRP$-teamNN-playerCD-unumVB-bowner DT-nullNN-null NP-null VP-bownerNP-player S-bowner

40 40 Overview of S CISSOR Integrated Semantic Parser SAPT Training Examples TRAINING SAPT ComposeMR MR NL Sentence TESTING learner

41 41 ComposeMR ourplayer 2 has theball teamplayerunumbowner null bownerplayer bowner

42 42 ComposeMR ourplayer 2 has theball teamplayer(_,_)unumbowner(_) null bowner(_)player(_,_) bowner(_)

43 43 ComposeMR ourplayer 2 has theball teamplayer(_,_)unumbowner(_) null player(_,_) bowner(_) null bowner(_) null player(_,_) player(team,unum) player(our,2) bowner(_) bowner(player) bowner(player(our,2))

44 44 Collins’ Head-Driven Model 2 A generative, lexicalized model Each node on the tree has a syntactic label, it is also lexicalized with its head word ourplayer2hastheball PRP$NNCDVBDTNN NP(ball) VP(has)NP(player) S(has)

45 45 Modeling Rule Productions as Markov Processes S(has) VP(has) P h (VP | S, has)

46 46 Modeling Rule Productions as Markov Processes S(has) VP(has) {NP }{ } P h (VP | S, has) × P lc ({NP} | S, VP, has) × P rc ({} | S, VP, has)

47 47 Modeling Rule Productions as Markov Processes S(has) VP(has)NP(player) {NP }{ } P h (VP | S, has) × P lc ({NP} | S, VP, has) × P rc ({} | S, VP, has) × P d (NP(player) | S, VP, has, LEFT, {NP})

48 48 Modeling Rule Productions as Markov Processes S(has) VP(has)NP(player) { } P h (VP | S, has) × P lc ({NP} | S, VP, has) × P rc ({} | S, VP, has) × P d (NP(player) | S, VP, has, LEFT, {NP})

49 49 Modeling Rule Productions as Markov Processes S(has) VP(has)NP(player) { } STOP P h (VP | S, has) × P lc ({NP} | S, VP, has) × P rc ({} | S, VP, has) × P d (NP(player) | S, VP, has, LEFT, {NP}) ×P d (STOP | S, VP, has, LEFT, {})

50 50 S(has) VP(has)NP(player) { } STOP P h (VP | S, has) × P lc ({NP} | S, VP, has) × P rc ({} | S, VP, has) × P d (NP(player) | S, VP, has, LEFT, {NP}) ×P d (STOP | S, VP, has, LEFT, {}) × P d (STOP | S, VP, has, RIGHT, {}) Modeling Rule Productions as Markov Processes

51 51 Integrating Semantics into the Model Use the same Markov processes Add a semantic label to each node Add semantic subcat frames –Give semantic subcategorization preferences –bowner takes a player as its argument ourplayer2hastheball PRP$-teamNN-playerCD-unumVB-bowner DT-null NN-null NP-null(ball) VP-bowner(has)NP-player(player) S-bowner(has ) ourplayer2hastheball PRP$NNCDVBDTNN NP(ball) VP(has)NP(player) S(has)

52 52 Adding Semantic Labels into the Model S-bowner(has) VP-bowner(has) P h (VP-bowner | S-bowner, has)

53 53 Adding Semantic Labels into the Model S-bowner(has) VP-bowner(has) P lc ({NP}-{player} | S-bowner, VP-bowner, has)× P rc ({}-{}| S-bowner, VP-bowner, has) P h (VP-bowner | S-bowner, has) × {NP}-{player}{ }-{ }

54 54 Adding Semantic Labels into the Model P d (NP-player(player) | S-bowner, VP-bowner, has, LEFT, {NP}-{player}) P lc ({NP}-{player} | S-bowner, VP-bowner, has)× P rc ({}-{}| S-bowner, VP-bowner, has) × P h (VP-bowner | S-bowner, has) × S-bowner(has) VP-bowner(has)NP-player(player) {NP}-{player}{ }-{ }

55 55 Adding Semantic Labels into the Model P d (NP-player(player) | S-bowner, VP-bowner, has, LEFT, {NP}-{player}) P lc ({NP}-{player} | S-bowner, VP-bowner, has)× P rc ({}-{}| S-bowner, VP-bowner, has) × P h (VP-bowner | S-bowner, has) × S-bowner(has) VP-bowner(has)NP-player(player) { }-{ }

56 56 Adding Semantic Labels into the Model P d (STOP | S-bowner, VP-bowner, has, LEFT, {}-{}) P d (NP-player(player) | S-bowner, VP-bowner, has, LEFT, {NP}-{player}) × P lc ({NP}-{player} | S-bowner, VP-bowner, has)× P rc ({}-{}| S-bowner, VP-bowner, has) × P h (VP-bowner | S-bowner, has) × S-bowner(has) VP-bowner(has)NP-player(player) { }-{ } STOP

57 57 P d (STOP | S-bowner, VP-bowner, has, RIGHT, {}-{}) S-bowner(has) VP-bowner(has)NP-player(player) { }-{ } STOP P d (STOP | S-bowner, VP-bowner, has, LEFT, {}-{}) × P d (NP-player(player) | S-bowner, VP-bowner, has, LEFT, {NP}-{player}) × P lc ({NP}-{player} | S-bowner, VP-bowner, has)× P rc ({}-{}| S-bowner, VP-bowner, has) × P h (VP-bowner | S-bowner, has) × Adding Semantic Labels into the Model

58 58 Smoothing Each label in SAPT is the combination of a syntactic label and a semantic label Increases data sparsity Use Bayes rule to break the parameters down P h (H | P, w) = P h (H syn, H sem | P, w) = P h (H syn | P, w) × P h (H sem | P, w, H syn ) Details in the paper

59 59 Experimental Corpora CLang –300 randomly selected pieces of coaching advice from the log files of the 2003 RoboCup Coach Competition –22.52 words on average in NL sentences –14.24 tokens on average in formal expressions GeoQuery [Zelle & Mooney, 1996] –250 queries for the given U.S. geography database –6.87 words on average in NL sentences –5.32 tokens on average in formal expressions

60 60 Experimental Methodology Evaluated using standard 10-fold cross validation Syntactic parses for tree-based version of SILT were obtained by training Collins’ parser [Bikel, 2004] on WSJ treebank and gold-standard parses of training sentences Correctness –CLang: output exactly matches the correct representation –Geoquery: the resulting query retrieves the same answer as the correct representation Metrics

61 61 Compared Systems C HILL [Tang & Mooney, 2001] –Learn control rules for parsing based on Inductive Logic Programming (ILP) S ILT [Kate, Wong, and Mooney, 2005] –Learns pattern-based transformation rules –S ILT -string –S ILT -tree S CISSOR [Ge and Mooney, 2005] –Integrated syntactic/semantic statistical parser G EOBASE –The original hand-built parser on Geoquery

62 62 Precision Learning Curve for CLang

63 63 Recall Learning Curve for CLang

64 64 F1 Measure Learning Curve for CLang

65 65 Precision Learning Curve for Geoquery

66 66 Recall Learning Curve for Geoquery

67 67 F1 Measure Learning Curve for Geoquery

68 68 Related Work [Zettlemoyer and Collins, 2005] use Combinatorial Categorial Grammar (CCG) formalism to learn a statistical semantic parser. –Requires an initial CCG syntactic grammar of English. P RECISE [Popescu, 2003] –Designed to work specially on NL database interfaces –Not a learning system [Miller et al., 1996; Miller et al., 2000] use an approach similar to S CISSOR to train a statistical parser integrating syntax and semantics. –Does not utilize subcat information –Task: Information Extraction

69 69 Future Work Explore methods that can automatically generate SAPTs to minimize the annotation effort for S CISSOR. Use of methods from statistical MT (e.g. word alignment) for semantic parsing. Learning semantic parsers just from sentences paired with “perceptual context.”

70 70 Contextually Ambiguous Sentence Meaning Sentences are uttered in complex situations composed of numerous potential meanings. Could assume each sentence is annotated with multiple possible meanings inferred from context (Siskind, 1996). –Multiple instance learning (Dietterich et al., 1997) Assuming context meaning is represented as a semantic network, sentence meaning could be assumed to be any connected subgraph of the context.

71 71 Sample Ambiguous Context Spot Dog Chewing Thing1 Bone isa agent patientMary “J uvenile caresses canine.” Child Petting agent patient isa HasColor Black obj attr Possess Barbie Doll HasColor Blonde obj attr HasPart Thing2 agent patient isa Hair isa obj part

72 72 Sample Ambiguous Context Spot Dog Chewing Thing1 Bone isa agent patientMary “J uvenile caresses canine.” Child Petting agent patient isa HasColor Black obj attr Possess Barbie Doll HasColor Blonde obj attr HasPart Thing2 agent patient isa Hair isa obj part

73 73 Sample Ambiguous Context Spot Dog Chewing Thing1 Bone isa agent patientMary “J uvenile caresses canine.” Child Petting agent patient isa HasColor Black obj attr Possess Barbie Doll HasColor Blonde obj attr HasPart Thing2 agent patient isa Hair isa obj part

74 74 Sample Ambiguous Context Spot Dog Chewing Thing1 Bone isa agent patientMary “J uvenile caresses canine.” Child Petting agent patient isa HasColor Black obj attr Possess Barbie Doll HasColor Blonde obj attr HasPart Thing2 agent patient isa Hair isa obj part

75 75 Sample Ambiguous Context Spot Dog Chewing Thing1 Bone isa agent patientMary “J uvenile caresses canine.” Child Petting agent patient isa HasColor Black obj attr Possess Barbie Doll HasColor Blonde obj attr HasPart Thing2 agent patient isa Hair isa obj part

76 76 Sample Ambiguous Context Spot Dog Chewing Thing1 Bone isa agent patientMary “J uvenile caresses canine.” Child Petting agent patient isa HasColor Black obj attr Possess Barbie Doll HasColor Blonde obj attr HasPart Thing2 agent patient isa Hair isa obj part

77 77 Sample Ambiguous Context Spot Dog Chewing Thing1 Bone isa agent patientMary “J uvenile caresses canine.” Child Petting agent patient isa HasColor Black obj attr Possess Barbie Doll HasColor Blonde obj attr HasPart Thing2 agent patient isa Hair isa obj part

78 78 Sample Ambiguous Context Spot Dog Chewing Thing1 Bone isa agent patientMary “J uvenile caresses canine.” Child Petting agent patient isa HasColor Black obj attr Possess Barbie Doll HasColor Blonde obj attr HasPart Thing2 agent patient isa Hair isa obj part

79 79 Sample Ambiguous Context Spot Dog Chewing Thing1 Bone isa agent patientMary “J uvenile caresses canine.” Child Petting agent patient isa HasColor Black obj attr Possess Barbie Doll HasColor Blonde obj attr HasPart Thing2 agent patient isa Hair isa obj part

80 80 Conclusions Learning semantic parsers is an important and challenging problem in natural-language learning. We have obtained promising results on several applications using a variety of approaches with different strengths and weaknesses. Not many others have explored this problem, I would encourage others to consider it. More and larger corpora are needed for training and testing semantic parser induction.

81 81 Thank You! Our papers on learning semantic parsers are on-line at: http://www.cs.utexas.edu/~ml/publication/lsp.html Our corpora can be downloaded from: http://www.cs.utexas.edu/~ml/nldata.html Questions??


Download ppt "University of Texas at Austin Machine Learning Group Department of Computer Sciences University of Texas at Austin Learning for Semantic Parsing of Natural."

Similar presentations


Ads by Google