NYU: Description of the Proteus/PET System as Used for MUC-7 ST Roman Yangarber & Ralph Grishman Presented by Jinying Chen 10/04/2002.

Slides:



Advertisements
Similar presentations
Describing Process Specifications and Structured Decisions Systems Analysis and Design, 7e Kendall & Kendall 9 © 2008 Pearson Prentice Hall.
Advertisements

NYU ANLP-00 1 Automatic Discovery of Scenario-Level Patterns for Information Extraction Roman Yangarber Ralph Grishman Pasi Tapanainen Silja Huttunen.
1 Words and the Lexicon September 10th 2009 Lecture #3.
Use Case Diagram © copyright 2001 SNU OOPSLA Lab..
Information Retrieval in Practice
Software Testing and Quality Assurance
Event Extraction: Learning from Corpora Prepared by Ralph Grishman Based on research and slides by Roman Yangarber NYU.
Ch 10 Part-of-Speech Tagging Edited from: L. Venkata Subramaniam February 28, 2002.
Chapter 6 Methodology Conceptual Databases Design Transparencies © Pearson Education Limited 1995, 2005.
Basi di dati distribuite Prof. M.T. PAZIENZA a.a
Using Information Extraction for Question Answering Done by Rani Qumsiyeh.
Empirical Methods in Information Extraction - Claire Cardie 자연어처리연구실 한 경 수
Marakas: Decision Support Systems, 2nd Edition © 2003, Prentice-Hall Chapter Chapter 1: Introduction to Decision Support Systems Decision Support.
Marakas: Decision Support Systems, 2nd Edition © 2003, Prentice-Hall Chapter Chapter 7: Expert Systems and Artificial Intelligence Decision Support.
Exploiting diverse knowledge source via Maximum Entropy in Name Entity Recognition Author: Andrew Borthwick John Sterling Eugene Agichtein Ralph Grishman.
1 Information Retrieval and Extraction 資訊檢索與擷取 Chia-Hui Chang, Assistant Professor Dept. of Computer Science & Information Engineering National Central.
Information Retrieval and Extraction 資訊檢索與擷取 Chia-Hui Chang National Central University
Automatic Acquisition of Lexical Classes and Extraction Patterns for Information Extraction Kiyoshi Sudo Ph.D. Research Proposal New York University Committee:
Automatically Constructing a Dictionary for Information Extraction Tasks Ellen Riloff Proceedings of the 11 th National Conference on Artificial Intelligence,
Overview of Search Engines
Artificial Intelligence Research Centre Program Systems Institute Russian Academy of Science Pereslavl-Zalessky Russia.
Chapter 4 Query Languages.... Introduction Cover different kinds of queries posed to text retrieval systems Keyword-based query languages  include simple.
Lecture 1, 7/21/2005Natural Language Processing1 CS60057 Speech &Natural Language Processing Autumn 2005 Lecture 1 21 July 2005.
Chapter 6: The Traditional Approach to Requirements
©Ian Sommerville 2004Software Engineering, 7th edition. Chapter 18 Slide 1 Software Reuse.
Computational Methods to Vocalize Arabic Texts H. Safadi*, O. Al Dakkak** & N. Ghneim**
Information Extraction
Empirical Methods in Information Extraction Claire Cardie Appeared in AI Magazine, 18:4, Summarized by Seong-Bae Park.
Chapter 14: Artificial Intelligence Invitation to Computer Science, C++ Version, Third Edition.
Avalanche Internet Data Management System. Presentation plan 1. The problem to be solved 2. Description of the software needed 3. The solution 4. Avalanche.
Semantic Grammar CSCI-GA.2590 – Lecture 5B Ralph Grishman NYU.
Methodology - Conceptual Database Design Transparencies
Methodology Conceptual Databases Design
9/14/2012ISC329 Isabelle Bichindaritz1 Database System Life Cycle.
Interpreting Dictionary Definitions Dan Tecuci May 2002.
Querying Structured Text in an XML Database By Xuemei Luo.
1/26/2004TCSS545A Isabelle Bichindaritz1 Database Management Systems Design Methodology.
Methodology: Conceptual Databases Design
Machine Translation  Machine translation is of one of the earliest uses of AI  Two approaches:  Traditional approach using grammars, rewrite rules,
Conceptual Database Design
Methodology - Conceptual Database Design
1 Two Applications of Information Extraction to Biological Science Journal Articles: Enzyme Interactions and Protein Structures Kevin Humphreys, George.
BioRAT: Extracting Biological Information from Full-length Papers David P.A. Corney, Bernard F. Buxton, William B. Langdon and David T. Jones Bioinformatics.
Artificial Intelligence Research Center Pereslavl-Zalessky, Russia Program Systems Institute, RAS.
IBM Research © Copyright IBM Corporation 2005 | A Development Environment for Configurable Meta-Annotators in a Pipelined NLP Architecture Youssef Drissi,
CPSC 871 John D. McGregor Module 3 Session 1 Architecture.
Using Semantic Relations to Improve Passage Retrieval for Question Answering Tom Morton.
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology Mining Logs Files for Data-Driven System Management Advisor.
MedKAT Medical Knowledge Analysis Tool December 2009.
4. Relationship Extraction Part 4 of Information Extraction Sunita Sarawagi 9/7/2012CS 652, Peter Lindes1.
Shallow Parsing for South Asian Languages -Himanshu Agrawal.
CS 4705 Lecture 17 Semantic Analysis: Robust Semantics.
Acquisition of Categorized Named Entities for Web Search Marius Pasca Google Inc. from Conference on Information and Knowledge Management (CIKM) ’04.
Using Wikipedia for Hierarchical Finer Categorization of Named Entities Aasish Pappu Language Technologies Institute Carnegie Mellon University PACLIC.
Exploiting Named Entity Taggers in a Second Language Thamar Solorio Computer Science Department National Institute of Astrophysics, Optics and Electronics.
Concepts and Realization of a Diagram Editor Generator Based on Hypergraph Transformation Author: Mark Minas Presenter: Song Gu.
A Brief Maximum Entropy Tutorial Presenter: Davidson Date: 2009/02/04 Original Author: Adam Berger, 1996/07/05
ICE: a GUI for training extraction engines CSCI-GA.2590 Ralph Grishman NYU.
Chapter – 8 Software Tools.
Overview of Statistical NLP IR Group Meeting March 7, 2006.
1 7 Systems Analysis and Design in a Changing World, 2 nd Edition, Satzinger, Jackson, & Burd Chapter 7 The Object-Oriented Approach to Requirements.
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology 1 Chinese Named Entity Recognition using Lexicalized HMMs.
Mastering the Pipeline CSCI-GA.2590 Ralph Grishman NYU.
Information Retrieval in Practice
Kenneth Baclawski et. al. PSB /11/7 Sa-Im Shin
Social Knowledge Mining
Overview of Entity‐Relationship Model
presented by Thomas L. Packer
CS246: Information Retrieval
Presentation transcript:

NYU: Description of the Proteus/PET System as Used for MUC-7 ST Roman Yangarber & Ralph Grishman Presented by Jinying Chen 10/04/2002

Outline Introduction Proteus IE System PET User Interface Performance on the Launch Scenario

Introduction Problem : portability and customization of IE engines at the scenario level To address this problem –NYU built a set of tools, which allow the user to adapt the system to new scenarios rapidly through example-based learning –The present system operates on two tiers: Proteus & PET

Introduction (Cont.) Proteus –Core extraction engine, an enhanced version of the one employed at MUC-6 PET –GUI front end, through which the user interacts with Proteus –The user provide the system examples of events in text, and examples of associated database entries to be created

Proteus IE System Modular design –Control is encapsulated in immutable, domain- independent core components –Domain-specific information resides in the knowledge bases

Proteus IE System (Cont.) Lexical analysis module –Assign each token a reading or a list of alternative readings by consulting a set of on- line dictionaries Name Recognition –Identify proper names in the text by using local contextual cues

Proteus IE System (Cont.) Partial Syntax –Find small syntactic units, such as basic NPs and VPs –Marks the phrase with semantic information, e.g. the semantic class of the head of the phrase Scenario Patterns –Find higher–level syntactic constructions using local semantic information: apposition, prepositional phrase attachment, limited conjucntions, and clausal constructions.

Proteus IE System (Cont.) Note: –The above three modules are Pattern matching phrases, they operate by deterministic, bottom-up, partial parsing or pattern matching. –The output is a sequence of LFs corresponding to the entities, relationships, and events encountered in the analysis.

Figure 2: LF for the NP: “a satellite built by Loral Corp. of New York for Intelsat”

Proteus IE System (Cont.) Reference Resolution (RefRes) –Links anaphoric pronouns to their antecedents and merges other co-referring expressions Discourse Analysis –Uses higher-level inference rules to build more complex event structures –E.g. a rule that merges a Mission entity with a corresponding Launch event. Output Generation

PET User Interface A disciplined method of customization of knowledge bases, and the pattern base in particular Organization of Patterns –The pattern base is organized in layers –Proteus treats the patterns at the different levels differently –Acquires the most specific patterns directly from user, on a per-scenario basis

clausal patterns that capture events (scenario-specific) find relationships among entities, such as between persons and organizations perform partial syntactic analysis most general patterns, capture the most basic constructs, such as proper names, temporal expressions etc. Domain- dependent Domain- independent Core part of System user Pattern Lib

PET User Interface (Cont.) Pattern Acquisition –Enter an example –Choose an event template –Apply existing patterns (step 3) –Tune pattern elements (step 4) –Fill event slots (step 5) –Build pattern –Syntactic generalization

Step 3 Step 4 Step 5

Performance on the Launch Scenario Scenario Patterns –Basically two types: launch events and mission events –In cases there is no direct connection between these two events, the post-processing inference rules attempted to tie the mission to a launch event Inference Rules –Involve many-to-many relations (e.g. multiple payloads correspond to a single event) –Extending inference rule set with heuristics, e.g. find date and site

Conclusion: Example-based pattern acquisition is appropriate for ST-level task, especially when training data is quite limited Pattern editing tools are useful and effective

NYU: Description of the MENE Named Entity System as Used in MUC-7 Andrew Borthwick, John Sterling etc. Presented by Jinying Chen 10/04/2002

Outline Maximum Entropy MENE’s Feature Classes Feature Selection Decoding Results Conclusion

Maximum Entropy Problem Definition The problem of named entity recognition can be reduced to the problem of assigning 4*n+1 tags to each token –n: the number of name categories, such as company, product, etc. For MUC-7, n=7 –4 states: x_start, x_continue, x_end, x_unique –other : not part of a named entity

Maximum Entropy (cont.) Maximum Solution –compute p(f | h), where f is the prediction among the 4*n+1 tags and h is the history –the computation of p(f | h) depends on a set of binary-valued features, e.g.

Maximum Entropy (cont.) –Given a set of features and some training data, the maximum entropy estimation process produces a model:

MENE’s Feature Classes Binary Features Lexical Features Section Features Dictionary Features External Systems Features

Binary Features Features whose “history” can be considered to be either on or off for a given token. Example: –The token begins with a capitalized letter –The token is a four-digit number

Lexical Features Example:

Section Features Features make predictions based on the current section of the article, like “Date”, “Preamble”, and “Text”. Play a key role by establishing the background probability of the occurrence of the different futures (predictions).

Dictionary Features Make use of a broad array of dictionaries of useful single or multi-word terms such as first names, company names, and corporate suffixes. Require no manual editing

External Systems Features MENE incorporates the outputs of three NE taggers –a significantly enhanced version of the traditional, hand-coded “Proteus” named-entity tagger –Manitoba –IsoQuest

Example:

Feature Selection Simple Select all features which fire at least 3 times in the training corpus

Decoding Simple –For each token, check all the active features for this token and compute p(f | h) –Run a viterbi search to find the highest probability coherent path through the lattice of conditional probabilities

Results Training set: 350 aviation disaster articles (consisted of about 270,000 words) Test set: –Dry run : within-domain corpus –Formal run : out-of-domain corpus

Result (cont.)

Conclusion A new, still immature system. Can improve the performance by: –Adding long-range reference-resolution features –Exploring compound features –Sophisticated methods of feature selection Highly portable An efficient method to combine NE systems