Computer Models of Interactive Narrative
Computer Models of Interactive Narrative CSC 582
Popular in Course
Popular in ComputerScienence
This 43 page Class Notes was uploaded by Jaden Jakubowski on Thursday October 15, 2015. The Class Notes belongs to CSC 582 at North Carolina State University taught by Staff in Fall. Since its upload, it has received 34 views. For similar materials see /class/223827/csc-582-north-carolina-state-university in ComputerScienence at North Carolina State University.
Reviews for Computer Models of Interactive Narrative
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 10/15/15
Discourse Barbara Grosz Martha Pollack Candace Sidner Presented by Samuel Manilla Outline Early Systems Discourse Structure Linguistic Indicators Plan Recognition Early Systems 0 LUNAR Woods et a1 1972 7 Provided Mechanisms for resolving pronouns 7 Each questionanswer pair is an independent exchange SHRDLU Winograd 1971 7 Engaged in dialogue about toy blocks 7 Could interpret some pronouns Next Generation Systems 0 SAM Cullingford 1977 amp GUS Bobrow 1977 7 Limited domain to a small class of actions 0 Task Dialogue Understanding System Grosz et al 1977 7 Distinguished domain knowledge discourse information amp intention recognition 7 First system to consider the interaction between the structure of discourse amp the interpretation of expressions in the discourse Discourse Structure 0 Segmentation and Structure 7 Linde amp Polanyi 7 How different parts of the discourse relate 7 Models based on surface relations Discourse tree constructed from discourse grammars 0 Rhetorical Relations 7 ReichmanAdar 7 Focused on linguistic level 7 Argument Claim followed by support Discourse Structure 0 Intentions amp Attention 7 Levy amp Grosz and Sidner 7 Plans amp Intentions are at the root of discourse structure 7 Pushdown stack model Stack grows as new segmenw are introduced Stack shrinks as intentions are resolved Linguistic Indicators 0 Cue Phrases 7 ie now in the first place by the way 7 Do not directly create meaning 7 Used to convey information about the structure 0 Intonation amp Gesture 7 Pitch Range Accent Tune 7 Gestures indicate segment boundries Phraselevel Phenomena 0 Pronouns 7 Interpretation based on focus or scope 7 Generally refer to the most recent applicable object that was focused on Reference to events Proverbs do it do that The do form and the bare pronoun behave differently John Runs John runs everyday of the week That s his main form of exercise 39that John running every day John runs everyday of the week He does it to stay healthy 39does it the act of John running Discourse Ellipsis 0 The omission of a syntactically required phrase needed to form an interpretation 7 Fred kissed his mother 7 John did i too 7 W kissed his own mother or Fred39s mother Plan Recognition 0 The use of language depends heavily on intention 39 Part of understanding is guessing the other39s plan Q Do you know when the next train leaves Without plan recognition Yes With plan recognition 1230 Sometimes larger plans need to be inferred as well Plan Recognition cont 0 A basic system 7 STRIPSlike representation 7 Plan recognition rules 7 Discourse is encoded into operations When does the train to Detroit leave REQUESTS H INFORMS H time1 7 The system uses these encoding to derive intention Discussion Pragmati s and Natural Language Generation JE thULiu IXdL 1H1 About the Author 1wa L i w LL Emmy 21 1de 39n mm mm L 13mgme M if Wahmm ngmh l iff 39ziimikxm m Ckqupmug gramme 1139 1 mw Wigwam and 1 t Pragmatics Semantics IRela ons between expressions Pragmatics I Context I Interlocutor PAULINE mm mm 01mm Lmngum In Natural Emimnmts Setting and Interlocutor Characteristics a Cmversational atmosphere time home comditiom a meledge interest opinim emotim E Heater Same as Speaker Language ability a SpeakerHeater Relationship Acquaimamef Social status ema m M FL Vft 4 InterpersonalGoals El Bearer I Affect knowledge Emmion goals I Involve bearer E Speaker39me Relationship 39 Affect hearet s emotim toward Speaker 39 Affect relative social status Affect Werperml distance MUmu Decisions a TopimCoIlec m Topic Orgmim m a Sentence Organization m 3th and Organization of Smtence Part3 a Word Choice 1 1171an Discoveries around Planning Formality a Afewoftherulesfor fmmality Include topics to make long sentences I Organize into complex sentmaes Include adverbial groups toward sentence beginning I No slang or comradions r3 Set meality based an other goals Formahtylmlloquial when Acquahtance riends Interpersonal mam step closer t0 callaun Inven ifeffect on heater is dislike or anger 1w ram 10 Detail E1 Dai ui M1ng Ffujm gnaw L m h ts Immiig 111 w 39w 2 m m m aka rm I Hm e jmmr m Par tiality M All natdi Sl anting Techniques QI39D IET39I JTYLE FS EEW 1 N39LG Where is it now a NEG Am m imdlm rm mud qg llmm if the 325 mt ij u seALg a Edam Batzemm gam 52 1 117139vrr l VlT Discussion Emits mag Why Emu VW E th YV x hmm Em hjgs parry ljpjm EEG i U mud g tmuc Vv I bumme Jame15 R Mash m Alok Baikadi Septem ber 24 2008 I 7m wr Outline 0 Architecture 0 Plans 9 States 0 Maps 0 Goals Lesso n s Q Conclusion Delta Act 0 Defined in terms of Goal States 0 4 main ones 0 DELTAPROX Proximity 5 DELTACONTROL Control a DELTAKNOW Knowledge 0 DELTATELL Sharing Knowledge 0 Defined in terms of PrePost conditions Example Act Planbox 1 X tries to move Y to Z Pre conditions X is selv movable If Xis different from Y then DPROXXXY and DOGRASP Xy DKNOW X wher is Z DKNOW X where is X DLINK X ocZ Act DO PTRANS XY ocZ Post contition is Y really at Z DKNOW could have goofed Postact If X is Different from Y then DO NEG GRASPX Y Sigma States o Describes Bodily Needs 0 Examples Hunger Thirst Rest Sex 0 Requires knowledge about how these are satisfied a Top level goals that the system works on Relationships o 7 axes of relationships 0 Competition 9 Dominance 9 Familiarity 0 Affection 6 Trust 0 Deceit 0 Indebtedness a Only exists as subgoals o No guaranteed procedures Personalities o 4 Personality traits 0 Kindness 0 Vanity 9 Honesty 9 Intelligence 9 Focus on individuals 9 Similar to relationships but no procedures for attaining o Heirarchical representation of geographical data a More detailed submap when the character is more familiar with the region 0 Doors are conceptual connections between submaps o Blueprints contain abstractions of maps Both temporary and permanent goal structures Permanent goals are called pi states Maintains subgoal relationships Some goals may be recoverable Frist Example One day Joe Bear was hungry He asked his friend Irving Bird where some honey was vring told him there was a beehive in the oak tree Joe threatened to hit Irving if he didn39t tell him where some honey was Second Example One day Joe Bear was hungry He asked his friend Irving Bird where some honey was Ivring told him there was a beehive in the oak tree Joe walked to the oak tree He ate the beehive Third Example Henry Ant was thirsty He walked over to the river bank where his good friend Bill Bird was sitting Henry slipped and fell in the river He was unable to call for help He drowned Fourth Example Henry Ant was thirsty He walked over to the river bank where his good friend Bill Bird was sitting Henry slipped and fell in the river Gravity having neither legs wings nor friends drowned Fifth Example Once upon a time there was a dishonest fox and a vain crow One day the crow was sitting in his tree holding a piece of cheese in his mouth He noticed that he was holding the piece of cheese He became hungry and swallowed the cheese The fox walked over to the crow The end Lessons Learned a What lessons should we take away from the paper a How can this system take advantage of new reasoning and planning advances
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'