Artificial Intelligence Natural Language Processing OUTLINE Overview of

Case Frame Examples “John broke the window with a hammer on Elm Street for Billy on Tuesday” “John broke the window with Sally” “Sally threw the ball at Billy” “Billy gave Sally the baseball bat” “Billy took the bat from his house to the playground”

Uninstantiated Case Frame [CASE-F: [HEADER [NAME: “move”] [PATTERN: ]] [OBJECT: [VALUE: _______ ] [POSITION: DO] [SEM-FILLER: | ]] [DESTINATION: [VALUE: _____ ] [MARKER: ] [SEM-FILLER: | ]] [SOURCE: [VALUE: _____ ] [MARKER: ] [SEM-FILLER: | ]]]

Case-Frame Grammar Fragments HEADER PATTERN determines which case frame to instantiate “move” | “transfer” | … “delete” | “erase” | “flush” | … LEXICAL MARKERS are prepositions++ that assign NPs to cases “to” | “into” | “onto” | … “from” | “in” | “that’s in” | … POSITIONAL INDICATORS also assign NPs to cases DO means “direct object position” (unmarked NP right of V) SUBJ means “subject position” (unmarked NP left of V)

Case Frame Instantiation Process Select which case-frame(s) match input string n Match header-patterns against input Set up constraint-satisfaction problem n SEM-FILLER, POSITION, MARKER constraints At-most one value per case constraint n Any required case must be filled constraint n At-most one case per input-substring constraint n Solve constraint-satisfaction problem n Use least-commitment, or satisfiability algorithm

Instantiated Case Frame S 1: “Please transfer foo. c from the diskette to my notes directory” [CASE-F: [HEADER [NAME: “move”] [VALUE: S 1]] [OBJECT: [VALUE: “foo. c” ]] [DESTINATION: [VALUE: “notes directory” ]] [SOURCE: [VALUE: “diskette” ]]]

Conceptual Dependency Canonical representation of NL developed by Schank Computational motivation—organization of inferences [ATRANS rel: POSSESSION actor: JOHN object: BALL source: JOHN recipient: MARY] "John gave Mary a ball" [ATRANS rel: POSSESSION actor: MARY object: BALL source: JOHN recipient: MARY] "Mary took the ball from John" [ATRANS rel: OWNERSHIP CAUSE rel: OWNERSHIP actor: JOHN actor: MARY object: APPLE object: 25 CENTS source: JOHN CAUSE source: MARY recipient: JOHN "John sold an apple to Mary for 25 Cents. "

Conceptual Dependency Other conceptual dependency primitive actions include: PTRANS--Physical transfer of location MTRANS--Mental transfer of information MBUILD--Create a new idea/conclusion from other info INGEST--Bring any substance into the body PROPEL--Apply a force to an object States and causal relations are also part of the representation: ENABLE (State enables an action) RESULT (An action results in a state change) INITIATE (State or action initiates mental state) REASON (Mental state is the internal reason for an action) [PROPEL [STATECHANGE actor: JOHN CAUSE state: PHYSICALINTEGRITY object: HAMMER object: WINDOW direction: WINDOW] endpoint: -10] "John broke the window with a hammer"

Robust Parsing Spontaneously generated input will contain errors and items outside an interface's grammar n Spelling errors tarnsfer Jim Smith from Econoics 237 too Mathematics 156 n Novel words transfer Smith out of Economics 237 to Basketwork 100 n Spurious phrases please enroll Smith if that's possible in I think Economics 237 n Ellipsis or other fragmentary utterances also Physics 314 n Unusual word order In Economics 237 Jim Smith enroll n Missing words enroll Smith Economics 237

What Makes MT Hard? Word Sense “Comer” [Spanish] eat, capture, overlook “Banco” [Spanish] bank, bench Specificity “Reach” (up) “atteindre” [French] “Reach” (down) “baisser” [French] 14 words for “snow” in Inupiac Lexical holes “Shadenfreuder” [German] happiness in the misery of others, no such English word Syntactic Ambiguity (as discussed earlier)

Bar Hillel's Argument 1. Text must be (minimally) understood before translation can proceed effectively. 2. Computer understanding of text is too difficult. 3. Therefore, Machine Translation is infeasible. - Bar Hillel (1960) Premise 1 is accurate Premise 2 was accurate in 1960 Some forms of text comprehension are becoming possible with present AI technology, but we have a long way to go. Hence, Bar Hillel's conclusion is losing its validity, but only gradually.

What Makes MT Hard? Word Sense “Comer” [Spanish] eat, capture, overlook “Banco” [Spanish] bank, bench Specificity “Reach” (up) “atteindre” [French] “Reach” (down) “baisser” [French] 14 words for “snow” in Inupiac Lexical holes “Shadenfreuder” [German] happiness in the misery of others, no such English word Syntactic Ambiguity (as discussed earlier)

Types of Machine Translation Interlingua Semantic Analysis Syntactic Parsing Source (Arabic) Sentence Planning Transfer Rules Direct: SMT, EBMT Text Generation Target (English)

Transfer Grammars: N(N-1) L 1 L 2 L 3 L 4

Interlingua Paradigm for MT (2 N) L 1 L 2 L 3 L 4 L 1 Semantic Representation aka “interlingua” L 2 L 3 L 4 For N = 72, T/G 5112 grammars, Interlingua 144

Beyond Parsing, Generation and MT Anaphora and Ellipsis Resolution n n "Mary got a nice present from Cindy. It was her birthday. " "John likes oranges and Mary apples. " Dialog Processing n n n "Speech Acts" (literal intended message) Social Role context s peech act selection "General" context sometimes needed Example 10 -year old: "I want a juicy Hamburger!" Mother: "Not today, perhaps tomorrow…" General: "I want a juicy Hamburger. " Aide: "Yes, sir!!" Prisoner 1: "I want a juicy Hamburger. " Prisoner 2: "Wouldn't that be nice for once. "

Social Role Determines Interpretation 10 -year old: “I want a juicy Hamburger!” Mother: “Not today, perhaps tomorrow…” General: Aide: “I want a juicy Hamburger!” “Yes, sir!!” Prisoner 1: “I want a juicy Hamburger!” Prisoner 2: “Wouldn't that be nice for once!”

Merit Cigarette Advertisement Merit Smashes Taste Barrier. -National Smoker Study ____________________ Majority of smokers confirm 'Enriched Flavor' cigarette matches taste of leading high tar brands. Why do we intepret barrier-smashing as good? [Metaphors, Metonomy, … other hard stuff]