Natural Language and Logic: Some Difficulties John Barnden Professor of Artificial Intelligence School of Computer Science

Slides:



Advertisements
Similar presentations
Basic Terms in Logic Michael Jhon M. Tamayao.
Advertisements

Artificial Intelligence
1 Knowledge Representation Introduction KR and Logic.
Week 2, Lecture 3 Dualism: mental events, substance vs. property dualism, four arguments.
The Subject-Matter of Ethics
Semantics (Representing Meaning)
Artificial Intelligence Chapter 21 The Situation Calculus Biointelligence Lab School of Computer Sci. & Eng. Seoul National University.
Semantics and Inference Part I Johan Bos. Overview of this lecture Inferences on the sentence level –Entailment –Paraphrase –Contradiction Using logic.
SEMANTICS.
Speak Symbols and Themes.
© Michael Lacewing Hume’s scepticism Michael Lacewing
Useful Expressions for Meeting & Negotiation Group 1 TiffanyTina LindaJodie JennyDaphne.
PARTS OF SPEECH  noun  pronoun  verb  adjective  adverb  preposition  conjunction  interjection!
Knowledge Representation Methods
Knowledge Representation I Suppose I tell you the following... The Duck-Bill Platypus and the Echidna are the only two mammals that lay eggs. Only birds.
Introduction to Semantics To be able to reason about the meanings of utterances, we need to have ways of representing the meanings of utterances. A formal.
Introduction to AI & AI Principles (Semester 1) WEEK 8 (07/08) [Barnden’s slides only] John Barnden Professor of Artificial Intelligence School of Computer.
Introduction to AI & AI Principles (Semester 1) WEEK 7 (07/08) [Barnden’s slides only] John Barnden Professor of Artificial Intelligence School of Computer.
Introduction to AI & AI Principles (Semester 1) WEEK 7 John Barnden Professor of Artificial Intelligence School of Computer Science University of Birmingham,
Synonymy of Syntactic Constructions
Introduction to AI & AI Principles (Semester 1) WEEK 6 – Wednesday Introduction to AI & AI Principles (Semester 1) WEEK 6 – Wednesday (2008/09) John Barnden.
Introduction to AI & AI Principles (Semester 1) WEEK 4 (07/08) John Barnden Professor of Artificial Intelligence School of Computer Science University.
Copyright © Cengage Learning. All rights reserved.
Highly successful people have the ability to think "outside the box" to see new and creative solutions to problems. Experiences with art help nurture creativity,
February 2009Introduction to Semantics1 Logic, Representation and Inference Introduction to Semantics What is semantics for? Role of FOL Montague Approach.
SAT Prep: Improving Paragraphs AVID III Spring 2012.
Artificial Intelligence Lecture No. 9 Dr. Asad Ali Safi ​ Assistant Professor, Department of Computer Science, COMSATS Institute of Information Technology.
CMPT 880/890 Arguments. Outline Arguments What should be an argument and what shouldn’t? Claims Reasons Evidence.
INFERENCES WHAT ARE INFERENCES ?. Inference Background Knowledge (schema) Background Knowledge (schema) Making Connections Making Connections Questioning.
Natural Language and Logic: Some Difficulties John Barnden Professor of Artificial Intelligence School of Computer Science
November 2003CSA4050: Semantics I1 CSA4050: Advanced Topics in NLP Semantics I What is semantics for? Role of FOL Montague Approach.
1. Make a rule that everyone in school should absolutely follow, without exception. 2. Make a rule that everyone in the world should absolutely follow.
Curly Questions By Clarissa Suchanek. Do you think you can ever lie to yourself? I don’t think I could ever lie to myself because even if I was capable.
Knowledge, Representation, and Reasoning CSC 244/444: Logical Foundations of Artificial Intelligence Henry Kautz.
Pronouns A pronoun is a word that takes the place of a noun.
Maniac Magee Literary Elements.
Kind honest brave loyal happy wise strong beautiful handsome rich smart funny What should a good friend be like? What qualities should a good friend.
Artificial Intelligence: Natural Language
LITERARY ANALYSIS Recap & Revision. YOUR LITERARY ANALYSIS NEEDS TO BE… BALANCED – both questions are equally important ACCURATE – your points need to.
CSE467/567 Computational Linguistics Carl Alphonce Computer Science & Engineering University at Buffalo.
Presentation about pragmatic concepts Implicatures Presuppositions
Artificial Intelligence Chapter 15 The Predicate Calculus Artificial Intelligence Chapter 15 The Predicate Calculus Biointelligence Lab School of Computer.
Slide 1 Improving your Persuasion and Influencing Skills for better negotiated outcomes Presented by Katrena Friel March 2009.
For Wednesday Read chapter 9, sections 1-3 Homework: –Chapter 7, exercises 8 and 9.
For Friday Read chapter 8 Homework: –Chapter 7, exercises 2 and 10 Program 1, Milestone 2 due.
STUDY SKILLS AGENDA -Subject/Predicate -Phrases -Independent and Dependent Clauses.
ARTIFICIAL INTELLIGENCE [INTELLIGENT AGENTS PARADIGM] Professor Janis Grundspenkis Riga Technical University Faculty of Computer Science and Information.
Artificial Intelligence Chapter 23 Multiple Agents Biointelligence Lab School of Computer Sci. & Eng. Seoul National University.
1 UNIT-3 KNOWLEDGE REPRESENTATION. 2 Agents that reason logically(Logical agents) A Knowledge based Agent The Wumpus world environment Representation,
Science Making measurements Accuracy and Precision.
Daniel Kroening and Ofer Strichman 1 Decision Procedures An Algorithmic Point of View Basic Concepts and Background.
How can we become good learners?
NATURAL LANGUAGE PROCESSING
Pronouns. What is a pronoun?  A pronoun takes the place of a noun.  Pronouns can be used in the following ways: Subject Predicate noun or adjective.
Ask for Advice: What should I do? Ask for advice: What should I do?
AS Computer Studies Algorithms & Problem Solving 1 & 2.
WHAT IS PEER PRESSURE? Pressure from people of one’s own age to behave in away that is similar or acceptable of them.
Artificial Intelligence Logical Agents Chapter 7.
The Scientific Method
CS621: Artificial Intelligence
Persuasion Defined Persuasion is the process of changing or reinforcing attitudes, beliefs, values, or behaviors. In a persuasive speech, the speaker explicitly.
Semantics (Representing Meaning)
The function of knowledge representation scheme is
Descartes’ conceivability argument for substance dualism
Writing an informal (describing an event)
Cavendish School Lincoln Close, Runcorn, Cheshire, England
What point is it trying to make?
Q3,J5.
Representations & Reasoning Systems (RRS) (2.2)
Presentation transcript:

Natural Language and Logic: Some Difficulties John Barnden Professor of Artificial Intelligence School of Computer Science

Substances §Do you like peanut butter? What exactly is it you like?  likes(student123, peanut-butter) ???  .x ( likes(student123, x)  is-peanut-butter(x) ) ???

Object Identity §What is a river? The water is constantly exchanged. And are the banks included? § So what would a logic constant like  the-nile stand for??? §Lincoln’s ax(e): Repaired bit by bit over years. Is it the same axe at the end?  What would lincolns-ax stand for?

Subtlety of Common Actions § “Tesco sells pineapples” § What does this mean, exactly? §And does it actually matter?  sells(Tesco, pineapples) ????? §Simple representation, but what inferences can be drawn ??

Subtlety of Prepositions §“There’s a banana in the bowl” §The banana need not be within the volume of the bowl. §“There’s a mirror on my ceiling” §The mirror is below the ceiling! §“Vanessa Granola is at her desk”

Context Often Needed for Precise Meaning: Some Examples §Pronouns. Ambiguous words. §Prepositional phrase attachment. “Hank saw Vanessa with a telescope” Did Hank use a telescope? Or did Vanessa have a telescope?

Quantification: Context-Sensitive §“When Hank arrived everyone laughed” §…. .x (is-person(x)  laughed(x)) would be wrong §“When Hank arrived everyone sat down to dinner.” §“Hank doesn’t believe anything Vanessa tells him”

Vague Quantification §Most, a few, several, many, ….  Most.x (is-person(x)  anxious(x)) ???? §But what inferences could you draw, and how?

Embedding of Propositions, Situations, etc. §“Vanessa fell over because Hank bumped into her” §In ordinary first-order logic, can’t write things like cause( bumpinto(H,V), fallover(V) )  because formulas (e.g. bumpinto(H,V) ) can’t be arguments in applications of predicate symbols etc.

§One method: take situations, events, etc. to be objects, just as e.g. people are.   f,b (is-fallover-event(f,V)  is-bumpinto-event(b,H,V)  cause(b,f))  Situations, events, etc. are treated as objects in ordinary language anyway.

Embedding contd.: Another example -- Belief §“Vanessa believes that Hank is lying.”  Can’t write following in ordinary first-order logic if lying is a predicate symbol: believes(Vanessa, lying(Hank))

“Honorary” Object Types §fake X, alleged X, imitation X, plastic tree, toy X, model boat,... § a fake gun is not actually a gun, so it would be bad to write something like fake (g)  is-gun(g) §but it would be nasty to have to use fake-gun(g)

Figurative Language §Metonymy: “He was listening to Bach” §Metaphor: “The suspicion grabbed me by the back of my neck.”

Variety of Types of Difficulty ¶May need context in order to pin precise meaning down. ·Precise meaning may be difficult to pin down even when context fully known. ¸ Precise meaning may be difficult to express in logic, or to do so usefully.

Final Remarks § This presentation has shown just a selection of the problems of expressing the meaning of natural language utterances in logic. §There are many approaches to the problems, but no-one has a complete solution to all of them and some remain puzzling. §Feel like doing a PhD on the issues??!