1 / 38

Answering WHY questions in Closed Domain from a Discourse Model

This research paper discusses the representation and implementation of a discourse model using GETARUNS for answering WHY questions in closed domain scenarios. It covers topics such as discourse entities, infons, spatio-temporal locations, and semantic processing.

ckrebs
Download Presentation

Answering WHY questions in Closed Domain from a Discourse Model

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Answering WHY questions in Closed Domain from a Discourse Model Rodolfo Delmonte University Ca’ Foscari - Venice Emanuele Pianta FBK - Trento

  2. OUTLINE • Representing a Discourse Model • Inds, Sets, Class, Infons, Locs, Card • Building a Discourse Model with GETARUNS • System Architecture • Asserting Discourse Entities • Questions/Answering from a DM • Entity properties pool • Wh-questions • Yes-No questions • Why-questions

  3. Discourse Model • A set of entities and relation between them, as “specified” in a discourse. • Discourse Entities can be used as Discourse Referents. • Entities and relation in a Discourse Model can be interpreted as representations of the cognitive objects of a mental model (cfr. Johnson-Laird) • Representation inspired to Situation Semantics. • Implemented as prolog facts.

  4. Representing a Discourse Model • Any piece of information is added to the DM as an infon. • An infon consists of a relation name, its arguments, a polarity (yes/no), and a couple of indexes anchoring the relation to a spatio-temporal location. • EX: meet, (arg1:john, arg2:mary), yes, 22-sept-2008, venice • Each infon has a unique identifier and can be referred to by other infons. • Infons are implemented as prolog facts (infon/6) • EX: infon(1, meet, [john, mary], 1, 22-sept-2008, venice).

  5. Kinds of Infons • Full infons • Situations: sit/6 • Facts: fact/6 • Complex infons: have other sit/fact as argument • Simplified infons • Entities: ind/2, set/2, class/2 • Cardinalities: card/3 • Membership: in/3 • Spatio-temporal rels: includes/2, during/2, …

  6. Entities, Cardinalities, Membership • Entities are represented in the DM without any commitment about their “existence” in reality. • Individual entities (“John”): ind(infon1, id5). • Extensional plural entities (“his kids”): set(infon2, id6). • Intensional plural entities (“lions”): class(…, id7). • Cardinality (only for sets: “four kids”) • card(…, id6, 5). • Membership (between individual and sets: “one of them”) • in(…, id5, id6).

  7. State of Affairs: sit vs fact • A sit is an abstraction (or representation or mental construct): no commitment is made with regard to correspondence with reality • May be used to represents objects of propositional attitudes: “He would like to sleep”, “When she sleeps, she’s happy” • A fact is like a sit, but accompanied with a commitment about correspondence to reality • May be used to represent “objective” statements: “He slept all night” • Arguments of facts and sits are labeled by their semantic role: • fact(inf4, sleep, [theme:id43], 1, time2, loc5)

  8. Spatio-temporal locations • Infons are “situated” in spatio-temporal locations • Related by a number of specific relations • cfr. Allen temporal relations • A special univ location is used to represent the universal location (including all other locations) • EG: instance-of relations or generic statements are situated in univ locations

  9. An Example “John slept in Venice” id(inf1, id1). id(inf2, loc1). name(inf3, id1, “John”). name(inf4, loc1, “Venice”). fact(inf3, sleep, [theme:id1], 1, time1, loc1). before(time1, speaker_time).

  10. Complex Infons “John wants to go to Venice tomorrow” id(inf1, id1). loc(inf2, loc1). name(inf3, id1, “John”). name(inf4, loc1, “Venice”). fact(inf3, want, [theme:id1, prop:inf4], 1, time1, loc2). sit(inf4, go, [theme:id1, goal:loc1], 1, time2, loc3). meets(time1, speaker_time). after(time2, time1). NB: loc2 and loc3 are left undefined

  11. SEMANTIC PROCESSING • Subdivision of Tasks • Referring Expressions • Clause Level Properties • External Pronouns + Definite Expressions • To check for disjointness • Informational Structure at Propositional level • Factitivity, Discourse Relations, Relevance, Subjectivity

  12. SEMANTIC PROCESSING • Spatiotemporal Locations • Producing Main Spatial Location • Updating Spatial Location • Whenever a new location is asserted either as argument or adjunct of main clause • Inferring same location • Use of pronominal deictics • Inferential processes to derive semantic relations • Meronimic or Hyponimic relations prevent update

  13. SEMANTIC PROCESSING • Spatiotemporal Locations • Producing Main Temporal Location • Updating Temporal Location • Whenever a new location is asserted either as argument or adjunct of main clause • This produces a new TIME FOCUS • New temporal locations must be lexically expressed: tense is not sufficient and only constitutes a local temporal relation • Inferring Same Location • Use of pronominal deictics • Inferential processes to derive semantic relations • Meronimic or Hyponimic relations prevent update

  14. SEMANTIC PROCESSING • Centering and Topic Hierarchy • External Pronouns • DPV may decide coreference • Creation of New Semantic Ids • Individuals (Ind) for singular new entities • Sets (Set) for plural new entities • Classes (Class) for generic new entities • Locations (Loc) for spatiotemporal main locations

  15. SEMANTIC PROCESSING • Indefinite NPs • Are treated as Ind if not in opaque contexts • Zero or bare singular/plural NPs • Are treated as Class or Sets (with a fixed cardinality of 5, or more if knowledge of the world is available) depending on whether they have an arbitrary or generic reading • Computed on the basis of tense, mood, modality, adjunct temporal modifiers, etc. • Definite NPs

  16. SEMANTIC PROCESSING • Definite NPs • if part of the scenery and belong to Mutual Knowledge or Generics or Common Knowledge of the World (see maintainance, instruction manuals) • Collective, group singular Definite NPs are computed as sets with a given cardinality (the army)

  17. SEMANTIC PROCESSING • Logical Form Mapping from DAGs • Conjoined wffs with syntactic indices • One for each f-structure • Recursively at clause level • Eventuality davidsonian structure • Tripartipe temporal structure • Mapping from LF into situation semantic structures • Conjoined wffs with semantic indices • Recursively headed by a situation operator

  18. Entity Property Pool • At the end of the computation each Entity has been associated to a certain number of properties and relations • This is what we call the EPP which is created automatically by collecting all relations, properties and other facts from the discourse model which carry the same semantic ID or are included or include that ID.

  19. Question Answering • Answering questions from the DM using the EPPs • As a first step we produce a new DM for the question where the facts are labeled as q_facts • Then we extract the main relation and the Focus attribute/s • These attributes will determine the type of answer and the type of search in the EPPs

  20. Question Answering • As a first step we look for a semantically similar/identical relation in the EPP • Then according the question type we extract arguments or adjuncts of the main predicate • Eventually we look for properties/attributed asserted in the question and try to match them with the properties associated to the current entity found, directly or inherited

  21. Question Answering • The focus item is recovered by means of the instantiation of the variable associated to the following q_facts: q_fact(K,focus,[arg:Id],1,_,_), q_fact(_,isa,[_:Id,_:Focus],1,A,B), • Where it can be noticed that polarity is forced to be equal to 1, that is positive

  22. Answer Generation • The first predicate fired by the system is, get_focus_arg(Focus, Pred, Args, Answer, True-NewFact), • which will give back the contents of the answer in the variable Answer and the governing predicate in Pred. These are then used to generate the actual surface form of the answer. Args and True-NewFact are used in case the question is a complete or yes/no question.

  23. Answer Generation • In order to generate the answer, tense and mood are searched in the DM; then a logical form is build as required by the generator, and the predicate build_reply is fired, get_focus_tense(T, M), Form=[Pred,T,M,P,[D]], build_reply(Out,Focus, Form), !. This predicate will actually generate the answer.

  24. Answer Generation • We will present general wh- questions at first. They include all types of factoid questions and also "How" questions. The main predicate looks for an appropriate linguistic description to substitute the wh- word argument position in the appropriate PAS, get_focus_arg(who, Pred, Ind, D1, NewP):- q_getevents(A,Pred), q_fact(X,Pred,Args,1,_,L), q_role(Y,X,Z,Role), answer_buildarg(Role, Pred, [Idx:Z], D, Facts), select_from_pred(Pred,Role,Facts,NewP,D1), !.

  25. Answer Generation • We use a different procedure in case the question governing predicate is a copulative verb, because we have to search for the associated property in the QDM, as follows, • copulative(Pred), q_fact(X,Pred,[prop:Y],1,_,_), q_fact(Y,Prop,[_:K,Role:Type],1,_,_), q_fact(_,inst_of,[_:K,_:Z],P,T,S), q_get_ind_des(K,Propp,Ty), • Copulative predicates have a proposition as their argument and the verb itself is not useful being semantically empty.

  26. Answer Generation The predicate corresponding to the proposition is searched through the infon "Y" identifying the fact. When we have recovered the Role and the linguistic description of the property "Propp" indicated by the wh- question, we pass them to the following predicate and search the associated individual in the DM, answer_buildarg(Role,Pred,[Idx:Propp],Answer,Facts)

  27. Answer Generation • Suppose the wh- question is a "Where" question with a copulative verb, the role will be a location and the Propp will be "in". "How" copulative questions will search for "class" properties, i.e. not for names or individuals, q_fact(X,how,[_:Y],1,_,_), q_fact(Q,isa,[_:K,class:Pred],1,_,_), q_fact(_,inst_of,[_:K,_:Z],P,T,S) • Semantic roles are irrelevant in this latter case: the only indication we use for the search is a dummy "prop" role.

  28. Answer Generation • On the contrary, when lexical verbs are governing predicates, we need to use the PAS and the semantic role associated to the missing argument to recover the appropriate answer in all other cases. Here we should also use different semantic strategy in case an argument is questioned and there is another argument expressed in the question – what, whom, who. Or else an adjunct is questioned – where, when, how, etc. – or the predicate is intransitive, an argument is questioned and there is no additional information available.

  29. Answer Generation • Consider a typical search for the answer argument, answer_buildarg(Role, Pred, Tops, Answer, Facts):- on(Ind:Prop, Tops), entity(Type,Id,Score,facts(Facts)), extract_properties(Type,Ind,Facts,Def,Num, NProp,Cat), select_allrole_facts(Role,Ind,Facts,Pred, PropLoc), Answer=[Def,nil,Num,NProp,Cat,PropLoc], !. • Here, "extract_properties" checks for the appropriate semantic type and property by picking one entity and its properties at the time. When it succeeds, the choice is further checked and completed by the call "select_allrole_facts".

  30. Answer Generation • extract_properties searches for individuals or sets filling a given semantic role in the predicate-argument structure associated to the governing predicate. • In addition, it has the important task of setting functional and semantic features for the generator, like Gender and Number. This is paramount when a pronoun has to be generated instead of the actual basic linguistic description associated to a given semantic identifier. In particular, Gender may be already explicitely associated in the DM to the linguistic description of a given entity or it may be derived from WordNet or other linguistic processors that looks for derivational morphology.

  31. Answer Generation • The call topichood_stack looks for static definiteness information associated to the linguistic description in the DM. Proper names are always "definite". On the contrary, common nouns may be used in definite or indefinite ways. This information may be modified by the dialogue intervening between user and system and be recorded in the user model. The decision is ultimately taken by "set_def" procedure which looks into the question-answering user model knowledge base where previous mentions of the same entity might have been recorded. Or else it does it - by means of update_user_model - to be used in further user-system interactions. If the entity semantic identifier is already present Def will be set to "definite", otherwise it will remain as it has been originally set in the DM.

  32. Computing Answers to WHY questions • Why question are usually answered by events, i.e. complete propositions. They would in general constitute cases of rhetorical clause pairs labeled either as a Motivation-Effect or a Cause-Result. In their paper [Delmonte et al. 2007], causal relations have been further decomposed into the following finer-grained subprocesses: • - Cause-Result • - Rationale-Effect • - Purpose-Outcome • - Circumstance-Outcome • - Means-Outcome

  33. Computing Answers to WHY questions • Consider now the pieces of knowledge needed to build the appropriate answer to the question "why is the tree called sugar maple tree?". Sentences involved to reconstruct the answer are, Maple syrup comes from sugar maple trees. At one time, maple syrup was used to make sugar. This is why the tree is called a "sugar" maple tree. • In order to build the appropriate answer, the system should be able to build an adequate semantic representation for the discourse anaphora "This", which is used to relate the current sentence to the event chain of the previous sentence.

  34. Computing Answers to WHY questions Eventually, the correct answer would be, "Because maple syrup was used to make sugar" which as can be easily gathered is the content of the previous complex sentence. Here below is the portion of the DM representation needed to reconstruct the answer, ind(infon19, id8) fact(infon20,inst_of,[ind:id8,class:edible_animal],1,univ, univ) fact(infon21, isa,[ind:id8,class:[maple_syrup]],1, id1, id7) set(infon23, id9) card(infon24, id9, 5) fact(infon25, sugar_maple, [ind:id10], 1, id1, id7) fact(infon26, of, [arg:id10, specif:id9], 1, univ, univ) fact(infon27,inst_of,[ind:id9,class:plant_life],1,univ, univ) fact(infon28, isa, [ind:id9, class:tree], 1, id1, id7)

  35. class(infon43, id13) fact(infon44,inst_of,[ind:id13,class:substance],1,univ, univ) fact(infon45, isa, [ind:id13, class:sugar], 1, id1, id7) fact(id14,make,[agent:id8,theme_aff:id13],1, tes(finf_m3), id7) fact(infon48,isa,[arg:id14,arg:ev],1,tes(finf_m3), id7) fact(infon49, isa, [arg:id15, arg:tloc], 1, tes(finf_m3), id7) fact(infon50, pres, [arg:id15], 1, tes(finf_m3), id7) fact(infon51,time,[arg:id14,arg:id15], 1, tes(finf_m3), id7) fact(id16,use,[theme_unaff:id8,prop:id14], 1, tes(sn5_m3), id7) _______________________ fact(id21,call,[actor:id9, theme_bound:id9], 1, tes(f1_m4), id7) ent(infon61, id18) fact(infon62,prop,[arg:id18, disc_set:[id16:use:[theme_unaff:id8, prop:id14]]], 1, id1, id7) ind(infon63, id19) fact(infon66, inst_of, [ind:id19, class:abstract], 1, univ, univ) fact(infon67, isa, [ind:id19, class:reason], 1, id1, id7) fact(infon81, in, [arg:id21, nil:id19], 1, tes(f1_m4), id7) fact(infon83, reason, [nil:id18, arg:id19], 1, id1, id7) fact(id23, be, [prop:infon83], 1, tes(sn10_m4), id7)

  36. Conclusions • Answering from a Discourse Model is very precise and very simple • Except for WHY question which requires special provision to encode event coreference • The use of Semantic Roles is paramount… • But it makes the machinery somewhat brittle

More Related