Slide 1Slide 2Today’s AgendaParts of SpeechHow do we define POS?Parts of SpeechOpen Class POSNounsVerbsAdjectives and AdverbsClosed Class POSParticle vs. PrepositionsMore Closed Class POSClosed Class POS: ConjunctionsSlide 15Slide 16Slide 17Slide 18Slide 19Back to regularly scheduled programming…POS Tagging: What’s the task?Penn Treebank Tagset: 45 TagsPenn Treebank Tagset: ChoicesWhy do POS tagging?Why is it hard?Try your hand at tagging…Try your hand at tagging…Why is it hard?*Part-of-Speech TaggingIt’s all about the benjaminsEvolution of the EvaluationEvaluation MetricComponents of a Proper EvaluationPart-of-Speech TaggingAutomatic POS TaggingRule-Based POS TaggingEngCG ArchitectureEngCG: Sample Lexical EntriesEngCG: Constraint Rule ApplicationEngCG: EvaluationSupervised Machine LearningThree Laws of Machine LearningThree Pillars of Statistical NLPAutomatic POS TaggingLearn to automatically paint the next Cubist masterpieceTBL: TrainingTBL: TrainingTBL: TrainingTBL: TrainingTBL: TrainingTBL: TrainingTBL: TrainingTBL: TestingTBL: TestingTBL: TestingTBL: TestingTBL: TestingTBL: TestingTBL Painting AlgorithmTBL Painting AlgorithmTBL Painting AlgorithmTBL TemplatesTBL Example RulesTBL POS TaggingThree Pillars of Statistical NLPIn case you missed it…Penn Treebank TagsetTurkish MorphologyTurkish Morphological AnalyzerMorphology Annotation SchemeHow to tackle the problem…Learning Decision ListsResultsWhat we covered today…Part-of-Speech TaggingCMSC 723: Computational Linguistics I ― Session #4Jimmy LinThe iSchoolUniversity of MarylandWednesday, September 23, 2009Source: Calvin and HobbsToday’s AgendaWhat are parts of speech (POS)?What is POS tagging?Methods for automatic POS taggingRule-based POS taggingTransformation-based learning for POS taggingAlong the way…EvaluationSupervised machine learningParts of Speech“Equivalence class” of linguistic entities“Categories” or “types” of wordsStudy dates back to the ancient GreeksDionysius Thrax of Alexandria (c. 100 BC)8 parts of speech: noun, verb, pronoun, preposition, adverb, conjunction, participle, articleRemarkably enduring list!4How do we define POS?By meaningVerbs are actionsAdjectives are propertiesNouns are thingsBy the syntactic environmentWhat occurs nearby?What does it act as?By what morphological processes affect itWhat affixes does it take?Combination of the aboveUnreliable! Think back to the comic!Parts of SpeechOpen classImpossible to completely enumerateNew words continuously being invented, borrowed, etc.Closed classClosed, fixed membershipReasonably easy to enumerateGenerally, short function words that “structure” sentencesOpen Class POSFour major open classes in EnglishNounsVerbsAdjectivesAdverbsAll languages have nouns and verbs... but may not have the other twoNounsOpen classNew inventions all the time: muggle, webinar, ...Semantics:Generally, words for people, places, thingsBut not always (bandwidth, energy, ...)Syntactic environment:Occurring with determinersPluralizable, possessivizableOther characteristics:Mass vs. count nounsVerbsOpen classNew inventions all the time: google, tweet, ...Semantics:Generally, denote actions, processes, etc.Syntactic environment:Intransitive, transitive, ditransitiveAlternationsOther characteristics:Main vs. auxiliary verbsGerunds (verbs behaving like nouns)Participles (verbs behaving like adjectives)Adjectives and AdverbsAdjectivesGenerally modify nouns, e.g., tall girlAdverbsA semantic and formal potpourri…Sometimes modify verbs, e.g., sang beautifullySometimes modify adjectives, e.g., extremely hotClosed Class POSPrepositionsIn English, occurring before noun phrasesSpecifying some type of relation (spatial, temporal, …)Examples: on the shelf, before noonParticlesResembles a preposition, but used with a verb (“phrasal verbs”)Examples: find out, turn over, go onParticle vs. PrepositionsHe came by the office in a hurryHe came by his fortune honestlyWe ran up the phone billWe ran up the small hillHe lived down the blockHe never lived down the nicknames(by = preposition)(by = particle)(up = particle)(up = preposition)(down = preposition)(down = particle)More Closed Class POSDeterminersEstablish reference for a nounExamples: a, an, the (articles), that, this, many, such, …PronounsRefer to person or entities: he, she, itPossessive pronouns: his, her, itsWh-pronouns: what, whoClosed Class POS: ConjunctionsCoordinating conjunctionsJoin two elements of “equal status”Examples: cats and dogs, salad or soupSubordinating conjunctionsJoin two elements of “unequal status”Examples: We’ll leave after you finish eating. While I was waiting in line, I saw my friend.Complementizers are a special case: I think that you should finish your assignmentLest you think it’s an Anglo-centric world,It’s time to visit ......The (Linguistic) Twilight ZoneTurkishuygarlaştıramadıklarımızdanmışsınızcasına →uygar+laş+tır+ama+dık+lar+ımız+dan+mış+sınız+casınabehaving as if you are among those whom we could not cause to become civilizedDigressionThe (Linguistic)Twilight ZonePerhaps, not so strange…ChineseNo verb/adjective distinction!漂漂 : beautiful/to be beautifulTzeltal (Mayan language spoken in Chiapas)Only 3000 root forms in the vocabularyThe verb ‘EAT’ has eight variations:General : TUNBananas and soft stuff : LO’Beans and crunchy stuff : K’UXTortillas and bread : WE’Meat and Chilies : TI’Sugarcane : TZ’ULiquids : UCH’DigressionThe (Linguistic)Twilight ZoneRiau Indonesian/MalayNo ArticlesNo Tense Marking3rd person pronouns neutral to both gender and numberNo features distinguishing verbs from nounsDigressionThe (Linguistic)Twilight ZoneAyam (chicken) Makan (eat)The chicken is eatingThe chicken ateThe chicken will eatThe chicken is being eatenWhere the chicken is eatingHow the chicken is eatingSomebody is eating the chickenThe chicken that is eatingRiau Indonesian/MalayDigressionThe (Linguistic)Twilight ZoneBack to regularly scheduled programming…POS Tagging: What’s the task?Process of assigning part-of-speech tags to wordsBut what tags are we going to assign?Coarse grained: noun, verb, adjective, adverb, …Fine grained: {proper, common}
View Full Document