Creation of English and Hindi Verb Hierarchies and
Creation of English and Hindi Verb Hierarchies and their Application to Hindi Word. Net Building and English-Hindi MT Debasri Chakrabarti, Gajanan Krishna Rane, Pushpak Bhattacharyya. Computer Science and Engineering Department, Indian Institute of Technology, Bombay, Mumbai, 40076, India. debasri, gkrane, pb@cse. iitb. ac. in 9/9/2020 C. F. I. L. T. , I. I. T. BOMBAY 1
Introduction n Verb hierarchy Ø creation of the verb hierarchy for English and Hindi verbs. organized according to semantics and syntax semantic hierarchy - through the super-ordinate terms and the Ø syntactic information- through UNL case relations Ø Ø n inbuilt ontology of the UNL KB. System is based on Ø Ø Ø English verb classes and their alternation (Levin) UNL System: UW Manual, Knowledge base (KB) & specification Semantic relations of English Word. Net 9/9/2020 C. F. I. L. T. , I. I. T. BOMBAY 2
Levin’s Class of English verbs n Classification of the English verbs n Details of Levin’s work Adopted from English Verb Classes and Alternation of Beth Levin’s classification of the English verb is the most significant and celebrated work. n Assumption underlying Levin’s work Syntactic behavior of a verb is semantically determined n Levin investigated and exploited this hypothesis for about 3200 English verbs. 9/9/2020 C. F. I. L. T. , I. I. T. BOMBAY 3
Details of Levin’s work Verb Classes n Preliminary Investigation considerable correlation between some facets of the semantics of verbs and their syntactic behavior n 200 semantic classes defined in Levin’s system each class share a number of alternations n Example of verb classes verbs of putting , verbs of communication, correspond verbs etc. 9/9/2020 C. F. I. L. T. , I. I. T. BOMBAY 4
The Universal Networking Language (UNL) n Universal Networking Language (UNL) electronic language for computers to express and exchange information n UNL system consists Universal words (UW) : Vocabulary of UNL Relations, attributes : Syntax of UNL knowledge base (KB): Semantics of UNL 9/9/2020 C. F. I. L. T. , I. I. T. BOMBAY 5
The Universal Networking Language n UNL represents information Ø Ø n sentence-by-sentence as a hyper-graph concepts as nodes and relations as arcs Sentence is a hyper-graph Ø Ø 9/9/2020 a node in the structure can itself be a graph the node is called a compound word (CW) C. F. I. L. T. , I. I. T. BOMBAY 6
Graphical representation in UNL eat (icl>do) agt obj @ entry @ present ins John (iof>person) rice (icl>food) spoon (icl>artifact) John eats rice with a spoon 9/9/2020 C. F. I. L. T. , I. I. T. BOMBAY 7
Verbal Concepts in UNL n Verbal concepts in the UNL system are organized into three categories Ø (icl>do) for defining the concept of an event which is caused by something or someone change (icl>do) : as in She changed the dress Ø (icl>occur) for defining the concept of an event that happens of its own accord change (icl>occur) : as in The weather will change Ø (icl>be) for defining the concept of a state verb remember (icl>be) : as in Do you remember me? 9/9/2020 C. F. I. L. T. , I. I. T. BOMBAY 8
Verbal Concepts in UNL do(agt>thing{, ^gol>thing, icl>do, ^obj>thing, ^ptn>thing, ^src>thing}) do(agt>volitional thing{, icl>do(agt>thing)}) do(agt>living thing{, icl>do(agt>volitional thing)}) do(agt>human{>living thing, icl>do(agt>living thing)}) do(agt>thing, gol>thing{, icl>do, ^obj>thing, ^ptn>thing, ^src>thing}) Partial hierarchical structure for do 9/9/2020 C. F. I. L. T. , I. I. T. BOMBAY 9
do in UNL KB n Semantic hierarchy in terms of the inbuilt ontology in KB do(agt>thing, gol>thing{, icl>do}, obj>thing{, ^ptn>thing, ^src>thing}) do({icl>do(}agt>thing{, gol>thing, obj>thing)}, gol>abstract thing, obj>abstract thing) do({icl>do(}agt>thing{, gol>abstract thing, obj>abstract thing)}, gol>custom{>abstract thing}, ob j>custom{>abstract thing}) do(gol>thing) 9/9/2020 do(gol>abstract thing) C. F. I. L. T. , I. I. T. BOMBAY do(gol>custom) 10
Creation of the verb hierarchy n n n First, a particular verb class is selected from Levin. Next the class is categorized according to the UNL format Parent node of a class is obtained through English wordnet and various dictionaries 9/9/2020 C. F. I. L. T. , I. I. T. BOMBAY 11
Creation of the verb hierarchy “put” ‘Put your clothes in the cupboard’. (to put something into a certain place) (icl>move(agt>person, obj>concrete thing, gol>place) (loc_prep{in/on/into/under/over}) [VTRANS, VOA-ACT] “hang” ‘He hanged the wallpaper on the wall’. (to suspend or fasten something so that it is held up from above and not supported from below) (icl>put{>move}(agt>person, obj>concrete thing, gol>place) (loc_prep{from/on}) [VTRANS, VOA-ACT] Partial hierarchy of the put class 9/9/2020 C. F. I. L. T. , I. I. T. BOMBAY 12
Verb Hierarchy in Hindi रखन ; rakhanaa; r «k. Hna ‘put’ ‘Put your things here. ’ (to put something into a certain place) (icl>act(agt>person, obj>concrete thing, gol>place) अपन स म न यह पर रख । ; ( «pna saman y «ha) p «r r «k. Ho) ; apanaa saamaana yahaa par rakho) {(adv_plc (यह /वह / ‘y «ha) / v «ha)’ loc_postp (पर ‘p «r’)} रखन , सज न ; r «k. Hna, s «jana; rakhanaa , sajaanaa; ‘arrange’ ‘he arranged the books here’. (to put into a proper or systematic manner) (icl>put{>act}(agt>person, obj>thing) उसन क त ब क यह पर सज कर रख । usne kitabo) ko y «ha) p «r s «jak «r r «k. Ha. ) (usne kitabo ko yahaa par sajaakar rakhaa. ) {(adv_man (सज कर , s «jak «r ; करम स , kr «m se))+ (adv_plc (यह /वह / ‘y «ha) / v «ha)’ ))+ loc_postp( पर ‘p «r’)} 9/9/2020 C. F. I. L. T. , I. I. T. BOMBAY 13
Verb Hierarchy in Hindi n Syntax frames specified for the put class in English Ø Ø n (adv_plc{here/there}) (loc_prep) Sentence frames for put in Hindi Ø Ø Ø 9/9/2020 English Hindi adv_plc (here / adv_man (सज कर , s «jak «r; करम स , there) kr «m se etc ) loc_prep (in, inside, on etc) adv_man adv_plc + adv_man loc_postp + adv_man C. F. I. L. T. , I. I. T. BOMBAY adv_plc(यह /वह / ‘y «ha) / v «ha)’) +loc_postp(पर ‘p «r)+adv_man (सज कर , s «jak «r ; करम स , kr «m se etc) loc_postp(क उपर, ke up «r etc)+adv_man (सज कर , s «jak «r etc) 14
Verb hierarchy and the Hindi Word. Net n Application of the hierarchy in the Hindi wordnet will help in determining Ø Ø n semantic relations like hypernymy and troponymy syntactic frames Application of the hierarchy in the Hindi wordnet revealed facts like Ø Ø 9/9/2020 difference in the representations for troponyms in Hindi and English reclassifications of the verbs in Hindi C. F. I. L. T. , I. I. T. BOMBAY 15
Representations of Troponyms English put sentence Hindi put your things here. रखन r «k. H na pile your books up on the shelves. ----- cram she cram the books into the suitcase. ----- 9/9/2020 C. F. I. L. T. , I. I. T. BOMBAY sentence अपन स म न यह पर रख । ; ( «pna s «man y «ha) p «r r «k. Ho) उसन ख न म एक क ऊपर एक स म न रख । ; ((((((usn e k. Hane me) ek ke Up «r ek saman r «k. Ha) उसन बकस क अनदर स र क त ब ठसकर रख । ; (usne 16
Classification of Hindi Verbs simple noun + verb 9/9/2020 conjunct adjective + verb C. F. I. L. T. , I. I. T. BOMBAY compound adverb + verb 17
Classification of the Hindi Verbs n n n Simple verbs ख न (k. Hana) ‘to eat’ Compound verbs ग र पडन Conjunct verbs Ø Ø Ø 9/9/2020 noun + verb adjective +verb down’ adverb + verb (gir p «ê na) ‘to fall down’ आरभ करन (ar «mb. H k «rna) ‘to start’ श त करन (Sant k «rna) ‘to calm उठ कर रखन (utak «r r «k. Hna) ‘to lift’ C. F. I. L. T. , I. I. T. BOMBAY 18
Reclassification of the Hindi verbs n Sentence frames of the verbs reveals Ø only noun+ verb conjunct is a true conjunct Hence, a re-classification of the verbs is needed 9/9/2020 C. F. I. L. T. , I. I. T. BOMBAY 19
Application in NLP n The application of the verb hierarchy in NLP Ø gives semantic hierarchy of a verbal concept Ø enumerates syntactic details of a verb Ø UNL based MT will be immensely benefited v 9/9/2020 possible UNL relations that appear with a concept is specified C. F. I. L. T. , I. I. T. BOMBAY 20
Application in MT Verb Sentence Frame UNL Relations fight Sam and Sue fought. conj_and agt>person fight Sam was fighting with Sue. prep_accompaniment{with} agt>person, ptn>person fight The tribesmen fought each other. -prep_with agt>person, obj>person 9/9/2020 C. F. I. L. T. , I. I. T. BOMBAY 21
Conclusion n System statistics Ø Ø n Common English verbs are dealt with Ø n n approximately 3000 English verbs approximately 5500 UWs tested against British National Corpus Coverage of both English and Hindi verbs is increasing everyday Visualizer and an application programming interface for the verb knowledge bases in both the languages are under construction 9/9/2020 C. F. I. L. T. , I. I. T. BOMBAY 22
- Slides: 22