DOC PREVIEW
MSU CSE 842 - Natural Language Processing
Course Cse 842-
Pages 10

This preview shows page 1-2-3 out of 10 pages.

Save
View full document
View full document
Premium Document
Do you want full access? Go Premium and unlock all 10 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 10 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 10 pages.
Access to all documents
Download any document
Ad free experience
Premium Document
Do you want full access? Go Premium and unlock all 10 pages.
Access to all documents
Download any document
Ad free experience

Unformatted text preview:

3/2/2011 CSE842, Spring 2011, MSU 1CSE842: Natural Language ProcessingLecture 14: Meaning Representation 3/2/2011 CSE842, Spring 2011, MSU 2Announcement• Homework 3 will be posted after the class– Due date: March 23• Start thinking about the final project– You can work on your own or as a team of two members. – You are encouraged to come up with your own topics relevant to your respective research fields that need NLP. – A list of potential topics will be made available later this week.3/2/2011 CSE842, Spring 2011, MSU 3Meaning• So far, we have focused on the structure of language – not on what things mean• We have seen that words have different meaning, depending on the context in which they are used• Everyday language tasks that require some semantic processing– Answering an essay question on an exam– Deciding what to order at a restaurant by reading a menu– Realizing that you’ve been insulted–…3/2/2011 CSE842, Spring 2011, MSU 4Meaning• Now, look at meaning representations --representations that link linguistic forms to knowledge of the world.• We are going to cover:– What is the meaning of a word– How can we represent the meaning– What formalisms can be used3/2/2011 CSE842, Spring 2011, MSU 5Common Meaning RepresentationsI have a car•FOPC/FOL:• Semantic Net: )(),()Speaker,()(, yCaryeHadThingeHavereHavingye∧∧∧∃havinghaver thing hadspeaker car3/2/2011 CSE842, Spring 2011, MSU 6Common Meaning Representations• Conceptual Dependency Diagram:Car ⇑ Poss-BySpeaker• Frame-based RepresentationsHavingHaver: SpeakerHadThing: Car3/2/2011 CSE842, Spring 2011, MSU 7Commonalities Between RepresentationsThey all share a common foundation:• meaning representation consists of structures composed of sets of symbols.• all represent the meaning of a particular linguistic input• All represent the state of affair in some world.3/2/2011 CSE842, Spring 2011, MSU 8What Can Serve as a Meaning Representation?• Anything that serves the core practical purposes of a program that is doing semantic processing ...– Answer questions (What is the tallest building in the world? )– Determining truth (Is the blue block on the red block? )– Drawing inferences (If the blue block is on the red block and the red block is on the tallest building in the world, then the blue block is on the tallest building in the world)• What are basic requirements of meaning representation3/2/2011 CSE842, Spring 2011, MSU 9What requirements must meaning representations fulfill?• Verifiability: The system’s ability to compare the state of affairs described by a representation to the state of affairs in some world as modeled in the knowledge base Does Maharani serve vegetarian food?Serves(Maharani, vegetarian food)3/2/2011 CSE842, Spring 2011, MSU 10What requirements must meaning representations fulfill?• Ambiguity: The system should allow us to represent meanings unambiguously– I wanna eat someplace that’s close to ICSI. • Vagueness: The system should allow us to represent vagueness– I want to eat Italian food.(pasta? spaghetti? lasagna?)3/2/2011 CSE842, Spring 2011, MSU 11Canonical Form• Distinct inputs could have the same meaning – Does Maharani serve vegetarian dishes?– Do they have vegetarian food at Maharani?– Are vegetarian dishes served at Maharani?– Does Maharani serve vegetarian fare? • Alternative: – Four different semantic representations– Problem with matching KB3/2/2011 CSE842, Spring 2011, MSU 12Canonical Form• Solution: inputs that mean the same thing should have the same meaning representation– Vegetarian dishes, vegetarian food, vegetarian fare– Have, serve• Relations among objects to be identical, how?->syntactic role analysis (e.g., subjects and objects)Maharani serves vegetarian dishesVegetarian dishes are served by Maharani3/2/2011 CSE842, Spring 2011, MSU 13Inference• Consider a more complex requestCan vegetarians eat at Maharani?It would be a mistake to invoke the canonical form to force the system to assign the same representation to this request as those of:Does Maharani serve vegetarian food?• Why are they result in the same answer? 3/2/2011 CSE842, Spring 2011, MSU 14Inference• Inference: system’s ability to draw valid conclusions based on the meaning representation of inputs and the background knowledge• The system must draw conclusions about the truth of propositions that are not explicitly represented in the KB, but that are logically derivable from the propositions that are present. 3/2/2011 CSE842, Spring 2011, MSU 15Expressiveness• Must accommodate wide variety of meanings• First-order Logic is expressive enough to handle many of the NLP needs3/2/2011 CSE842, Spring 2011, MSU 16Model-Theoretic Semantics• Model: a formal construct of particular states of the world we are trying to represent– Representing objects, properties of objects, and relations among objects. – Expressions in a meaning representation language can then be mapped to the elements of the model• Domain: a set of objects being represented for an application• Interpretation: function that maps meaning representation to the proper denotations in the model.3/2/2011 CSE842, Spring 2011, MSU 17We can evaluate a representation claim: Franco likes Frasca, Med serves ItalianBut what about: Not everybody likes Frasca, Katie and Caroline like the same restaurants3/2/2011 CSE842, Spring 2011, MSU 18First-Order Logic• FOL: provides a sound computational basis for the verifiability, inference, and expressiveness requirements– Supports the determination of truth– Supports compositionality of meaning– Supports representation of variables– Supports inference3/2/2011 CSE842, Spring 2011, MSU 19Basic Elements of FOL • Terms: constants, functions, variables– Constants: objects in the world, e.g. Maharani– Functions: concepts, e.g. LocationOf(Maharani)– Variables: x, e.g. LocationOf(x)• Predicates: symbols that refer to relations that hold among objects in some domain or properties Serves(Maharani, VegetarianFood)Restaurant(Maharani)3/2/2011 CSE842, Spring 2011, MSU 20FOL Syntax3/2/2011 CSE842, Spring 2011, MSU 21• Logical connectives permit compositionality of meaningI only have five dollars and I don’t have a lot of timeHave(Speaker, FiveDoallars) ∧¬Have(Speaker, LotofTime)FOL Syntax3/2/2011 CSE842, Spring 2011, MSU 22• Sentences in FOL can be assigned truth


View Full Document

MSU CSE 842 - Natural Language Processing

Course: Cse 842-
Pages: 10
Download Natural Language Processing
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view Natural Language Processing and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view Natural Language Processing 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?