Good Nouns, Bad Nouns What the corpus says and wha.ppt

《Good Nouns, Bad Nouns What the corpus says and wha.ppt》由会员分享，可在线阅读，更多相关《Good Nouns, Bad Nouns What the corpus says and wha.ppt（42页珍藏版）》请在淘文阁 - 分享文档赚钱的网站上搜索。

1、Good Nouns, Bad NounsWhat the corpus says and what native speakers thinkPhilip Dilts, University of Alberta Department of LinguisticsAACL 2008, Brigham Young University, Provo, UTMarch 14, 2008OverviewnSemantic Orientation Psychological measure of how good/bad a word isnSemantic Preference Corpus-ba

2、sed measure of collocational preferencenHow Semantic Orientation and Preference relate: The wrong story (abstract version) The right story (maybe?) How to get the right storySemantic Orientation Term from Hatzivassiloglou & McKeown (1997) How good or bad people feel about a wordSemantic Orientationf

3、unSemantic OrientationjailSemantic Orientation How good or bad people feel about a word:Out of context Psychologically realOsgood et al (1957) “Evaluative Factor” Experimentally determinedE.g., ANEW (Bradley & Lang 1999)“rate how you felt while reading each word: Happy vs. Unhappy”ANEW Questionnaire

4、Semantic PreferencenCollocational preference for a semantic category See, e.g, Partington (2004)nThe focus here: Preference for good/bad adjectives in adjective-noun sequencesSemantic PreferencethoughtsthoughtsthoughtsthoughtsthoughtsthoughtsthoughtsSemantic Preferencethoughtsthoughtsthoughtsthought

5、sthoughtsthoughtsthoughtssadcruelmorbid colddirtyi.e. nouns that collocate with many adjectives of one orientationSemantic Preferencethoughtsthoughtsthoughtsthoughtsthoughtsthoughtsthoughtssadcruelmorbid colddirtysweetsoothingi.e. nouns that collocate with many adjectives of one orientation.and rela

6、tively few adjectives of the other orientationSemantic PreferencenIn this case, we would say thoughts has a negative semantic preferencenHow closely do preference and orientation correspond? Do good nouns prefer to collocate with good adjectives? Can good nouns prefer to collocate with bad adjective

7、s?Does Preference Match Orientation?nDilts & Newman (2006) Start with adjectives rated as good/bad Pairs of adjectives from Osgood et al. (1957) Find nouns that follow many good or bad adjectives in a corpus (BNC) Approximately, nouns that are consistently modified by good or bad evaluationsDoes Pre

8、ference Match Orientation?nSurprising results: joke, grin, jokes, beast, thoughts, mistakes, laughter, head, shoulder, dreamsn10 strongest negative preferences in the BNCnAre these words consistently judged as negative by native speakers?Does Preference Match Orientation?nPilot orientation study (Di

9、lts & Newman 2008) Participants: 5 native English speakers Stimuli - 121 nouns: 25 nouns with most positive preference in BNC 25 nouns with most negative preference in BNC 71 filler nouns (strong preference by other methods) Method: ask participants to rate each noun as good/neutral/badDoes Preferen

10、ce Match Orientation?Speaker ratingsGoodBadCollocational preferencePositiveNegativeDoes Preference Match Orientation?conversationdevelopmentassessmentrangecommunityprogrammefamilysettingsupportsourceresponsealternativehomeconditionsmistakesdelusionbastardinjurieskillerbruiseblowscoughbeastmoodslumpm

11、omentlaughterdreamstruthslegacyjokeslogicjokegringodsSpeaker ratingsGoodBadCollocational preferencePositiveNegativeDoes Preference Match Orientation?conversationdevelopmentassessmentrangecommunityprogrammefamilysettingsupportsourceresponsealternativehomeconditionsmistakesdelusionbastardinjurieskille

12、rbruiseblowscoughbeastmoodslumpmomentlaughterdreamstruthslegacyjokeslogicjokegringodsSpeaker ratingsGoodBadCollocational preferencePositiveNegativeReinforcementDoes Preference Match Orientation?conversationdevelopmentassessmentrangecommunityprogrammefamilysettingsupportsourceresponsealternativehomec

13、onditionsmistakesdelusionbastardinjurieskillerbruiseblowscoughbeastmoodslumpmomentlaughterdreamstruthslegacyjokeslogicjokegringodsSpeaker ratingsGoodBadCollocational preferencePositiveNegativetransformationDoes Preference Match Orientation?conversationdevelopmentassessmentrangecommunityprogrammefami

14、lysettingsupportsourceresponsealternativehomeconditionsmistakesdelusionbastardinjurieskillerbruiseblowscoughbeastmoodslumpmomentlaughterdreamstruthslegacyjokeslogicjokegringodsSpeaker ratingsGoodBadCollocational preferencePositiveNegativetransformationTentative Conclusionsn Nouns that speakers rate

15、as good collocate with good adjectives, and vice versa i.e., nouns often collocate with adjectives that reinforce their orientationnMany good nouns attract bad adjectives Good nouns collocate with adjectives that transform their orientationnBUT: few bad nouns attract good adjectives Bad nouns dont c

16、ollocate (much) with adjectives that transform their orientationOrientation DOES NOT Match Preferencen Is this the right story? We looked at only 50 nouns with extreme collocational preferences Noun orientation ratings came from a small, biased sample 5 linguistics grad studentsnHow can we get a bet

17、ter story? Better measure of orientationANEWANEWn1034 words rated for pleasurenDo they agree with our raters?nDid our participants agree with ANEW?ANEW ratings vs. Collocational PreferencenDid our participants agree with ANEW?ANEW ratings vs. Collocational PreferencenDid our participants agree with

18、ANEW?ANEW ratings vs. Collocational PreferencetransformationconversationdevelopmentassessmentrangecommunityprogrammefamilysettingsupportsourceresponsealternativehomeconditionsmistakesdelusionbastardinjurieskillerbruiseblowscoughbeastmoodslumpmomentlaughterdreamstruthslegacyjokeslogicjokegringodsSpea

19、ker ratingsGoodBadCollocational preferencePositiveNegativetransformationANEW ratings vs. Collocational PreferenceANEW ratings vs. Collocational Preferencenbut only 10 of our 50 nouns were in ANEWnFor Dilts & Newman 2006, we established collocational preferences for 30,000 nouns More than 700 were ra

20、ted in ANEWnDoes the mismatch hold for these nouns too?n10 nouns from Pilot studyANEW ratings vs. Collocational Preferencen734 nouns from Dilts & Newman (2006)ANEW ratings vs. Collocational PreferenceANEW ratings vs. Collocational PreferenceANEW ratings vs. Collocational PreferenceANEW ratings vs. C

21、ollocational PreferencenEverything is drawn upwards Nouns tend to collocate with positive adjectivesANEW ratings vs. Collocational PreferencenMany bad nouns attracting good adjectivesMore transformation hereconversationdevelopmentassessmentrangecommunityprogrammefamilysettingsupportsourceresponsealt

22、ernativehomeconditionsmistakesdelusionbastardinjurieskillerbruiseblowscoughbeastmoodslumpmomentlaughterdreamstruthslegacyjokeslogicjokegringodsSpeaker ratingsGoodBadCollocational preferencePositiveNegativeMore transformation hereANEW ratings vs. Collocational PreferencenOpposite to the previous mism

23、atch!nFocusing on the extreme values selected outliersnNegative preference nouns had less extreme values e.g. minimum preference was -4, maximum preference was 8ANEW ratings vs. Collocational PreferenceConclusionsnThe right story: General tendency for positive collocational preference Noun orientati

24、ons are more often reinforced by collocating adjectives than transformed More bad nouns than good nouns are transformed by collocating adjectivesFuture WorknIs this the right story? Preference based on adjective pairs Negative adjectives were usually lower frequency e.g., pure (3301) vs. impure (68)

25、 (BNC) We controlled for low frequency, but not 0 frequency ie. nouns that collocated with pure but not impure got full positive preference scores for that pair But nouns are more likely to collocate only with pure because of its higher frequencyFuture WorknHow to get the right story: Select pairs o

26、f good/bad adjectives ANEW? MPQA? Matched for frequency Matched for goodness/badness Find nouns with strong collocational preferences for these adjectives Use Collostructional Analysis (Stefanowitsch & Gries 2003) ProfitReferencesnBradley, M.M., & Lang, P.J. (1999). Affective norms for English words

27、 (ANEW): Stimuli, instruction manual and affective ratings. Technical report C-1, Gainesville, FL. The Center for Research in Psychophysiology, University of Florida.nDilts, P. & J. Newman. 2008. Good Nouns, Bad Nouns, and the Company they Keep. Presentation at the 2008 Annual meeting of the LSA, Ch

28、icago, IL.nDilts, P. & J. Newman. 2006. A Note on Quantifying good and bad prosodies. Corpus Linguistics and Linguistic Theory 2(2):233-242.nOsgood, C. E., G.J. Suci & P.H. Tannenbaum. 1957. The Measurement of Meaning. Urbana: University of Illinois Press.nPartington, A. 2004. Utterly Content in Each Others Company. International Journal of Corpus Linguistics, 9:131-156nStefanowitsch, A. & S.Th. Gries. 2003. Collostructions. International Journal of Corpus Linguistics, 8:209-243Thanks!Members of the University of Alberta Department of LinguisticsYou!