Linguistic Resources for the 2013 TAC KBP Temporal SF Evaluation Joe Ellis (presenter), Jeremy Getman, Jonathan Wright, Stephanie Strassel Linguistic Data.

Slides:



Advertisements
Similar presentations
It’s a feeling…. An opinion…. An attitude!
Advertisements

University of Sheffield NLP Exercise I Objective: Implement a ML component based on SVM to identify the following concepts in company profiles: company.
Overview of the TAC2013 Knowledge Base Population Evaluation: Temporal Slot Filling Mihai Surdeanu with a lot help from: Hoa Dang, Joe Ellis, Heng Ji,
Text Analysis Conference Knowledge Base Population 2013 Hoa Trang Dang National Institute of Standards and Technology Sponsored by:
Overview of the TAC2013 Knowledge Base Population Evaluation: English Slot Filling Mihai Surdeanu with a lot help from: Hoa Dang, Joe Ellis, Heng Ji, and.
Time Management English lesson 7 Form “New Matrix”
Bibliography Cards Note Cards
Distant Supervision for Knowledge Base Population Mihai Surdeanu, David McClosky, John Bauer, Julie Tibshirani, Angel Chang, Valentin Spitkovsky, Christopher.
CALCULATING THE AGE OF A STUDENT FOR ASSESSMENT NATIONAL ASSOCIATION OF SPECIAL EDUCATION TEACHERS.
 TDT 2003 Evaluation Workshop, NIST, November 17-18, 2003 Creating the Annotated TDT-4 Y2003 Evaluation Corpus Stephanie Strassel, Meghan Glenn Linguistic.
Tri-lingual EDL Planning Heng Ji (RPI) Hoa Trang Dang (NIST) WORRY, BE HAPPY!
Overview of the KBP 2013 Slot Filler Validation Track Hoa Trang Dang National Institute of Standards and Technology.
Linguistic Resources for the 2013 TAC KBP Sentiment SF Evaluation Joe Ellis (presenter), Jeremy Getman, Jonathan Wright, Stephanie Strassel Linguistic.
TAC 2012 Cold Start Knowledge Base Population James Mayfield Javier Artiles Hoa Trang Dang Special thanks to: Brendan Callahan Bonnie Dorr Joe Ellis Tim.
Entity-oriented filtering of large streams John R. Frank Ian Soboroff Max Kleiman-Weiner Dan A. Roberts.
Linguistic Resources for the 2013 TAC KBP Slot Filling Evaluations Joe Ellis (presenter), Jeremy Getman, Jonathan Wright, Stephanie Strassel Linguistic.
Ang Sun Ralph Grishman Wei Xu Bonan Min November 15, 2011 TAC 2011 Workshop Gaithersburg, Maryland USA.
1 Adaptive Management Portal April
April 23, 2001LBSC 878 Text Data Mining Douglas W. Oard.
Event Extraction: Learning from Corpora Prepared by Ralph Grishman Based on research and slides by Roman Yangarber NYU.
1 COS 425: Database and Information Management Systems XML and information exchange.
Automatically Constructing a Dictionary for Information Extraction Tasks Ellen Riloff Proceedings of the 11 th National Conference on Artificial Intelligence,
Enhance legal retrieval applications with an automatically induced knowledge base Ka Kan Lo.
NCSRA Assignor Training Module For USSF North Carolina Arbiter Site Section 4 of 5 © Copyright July 2005 by Paul James, all rights reserved.
Leeds University Library How information literate are our incoming undergraduates? Angela Newton.
Basic Application Assessment Alan Jolliffe Team 3 Lower North Eastern Region.
1 NC WISE Parent Assistant A user-friendly web application to help parents track their children’s progress in school.
Empirical Methods in Information Extraction Claire Cardie Appeared in AI Magazine, 18:4, Summarized by Seong-Bae Park.
RESEARCHING AND CITING SOURCES WHAT INFORMATION WILL I HAVE TO RESEARCH? You will need to have researched information on the following areas: 
Nurse Home Visitor Program Revising your FY 11/12 CDPHE Funding NHVP Budget due to state budget reductions Presented on September 29, 2011 by Julie Becker.
Overview of the KBP 2012 Slot-Filling Tasks Hoa Trang Dang (National Institute of Standards and Technology Javier Artiles (Rakuten Institute of Technology)
Self-Service Finance Requisition, Encumbrance Query & View Document By Gina Bazile.
1 Automating Slot Filling Validation to Assist Human Assessment Suzanne Tamang and Heng Ji Computer Science Department and Linguistics Department, Queens.
Semantic Technologies & GATE NSWI Jan Dědek.
Creating a Digital Classroom. * Introduction * The Student Experience * Schoology’s Features * Create a Course & Experiment.
1 Where to look? –MGTC24 website has links on how to use search engines How to decide you have good sources? Doing Library Research.
Entity-oriented Filtering of Large Streams TREC KBA 2013 John R. Frank Ian Soboroff Max Kleiman-Weiner
IT-522: Web Databases And Information Retrieval By Dr. Syed Noman Hasany.
Linguistic Resources for the 2013 TAC KBP Entity Linking Evaluation Joe Ellis (presenter), Justin Mott, Xuansong Li, Jeremy Getman, Jonathan Wright, Stephanie.
PRIS at Slot Filling in KBP 2012: An Enhanced Adaboost Pattern-Matching System Yan Li Beijing University of Posts and Telecommunications
TimeML compliant text analysis for Temporal Reasoning Branimir Boguraev and Rie Kubota Ando.
©2003 Paula Matuszek CSC 9010: Text Mining Applications Dr. Paula Matuszek (610)
Indicator Management System (IMS) Relationships with other systems parts Indicator = Specification + assessment Data visualization (with metadata) Graphs.
RESEARCH POSTER PRESENTATION DESIGN © Triggers in Extraction 5. Experiments Data Development set: KBP SF 2012 corpus.
 TDT 2004 Evaluation Workshop, NIST, December 2-3, 2004 Creating the TDT5 Corpus and 2004 Evaluation Topics at LDC Stephanie Strassel, Meghan Glenn, Junbo.
Starter: reminder of the AS exam structure Paper 1: 3 questions assessing AOs 1, 3 and 4. – 2 questions on how language is used to create meanings and.
Linguistic Resources for the 2013 TAC KBP Cold Start Evaluation Joe Ellis (presenter), Jeremy Getman, Jonathan Wright, Stephanie Strassel Linguistic Data.
Case Study SummaryChallenges Novartis, a Swiss multinational pharmaceutical company, needed a translation, cognitive debriefing and linguistic validation.
CSE 454 Advanced Internet Systems Project Logistics Dan Weld.
AQUAINT AQUAINT Evaluation Overview Ellen M. Voorhees.
Extending ICT Research Co-operation between the European Union, Eastern Europe and the Southern Caucasus EECA FP7 Help-Desk EXTEND Training Workshop Baku,
What’s an HRIS? Any system that supports any aspect of the HR function Primary function: –Documenting the employment relation in all it’s complexity Basic.
GT Frankenstein Drills Quarter Drill 1 11/7  Take out phrase activity  Homework: Comma HO Comma quiz 11/11  Objective: SWBAT read.
Information Retrieval Quality of a Search Engine.
Lesson 3.5. ACT Practice Questions 1)Get out a piece of paper, and put your name on it. 2)I am going to show you 3 practice ACT math questions, and I.
Populating and Evaluating Knowledge Bases
PearsonAccess EOC Training
Tri-lingual EDL for 2017 and Beyond
From Strings to Things: KELVIN in 2016 KBP and EDL
Editing Your Website on SharePoint 2013
From Strings to Things: KELVIN in TAC KBP and EDL
Building the Localization Web
Free and open to the public.
Show the URL for the first location.
Reading & Writing to Succeed on the EAS
1. What Google app created this assessment?
Data Director 4.0 Creating a CFA Key J. Martin • Hemet USD August 2012.
PCS CONTENT ON CMS.
The ultimate in data organization
Warm-Up Problems Unit 0.
Presentation transcript:

Linguistic Resources for the 2013 TAC KBP Temporal SF Evaluation Joe Ellis (presenter), Jeremy Getman, Jonathan Wright, Stephanie Strassel Linguistic Data Consortium University of Pennsylvania, USA

2013 Source Corpus LanguageGenreDocuments English Newswire1,000,257 Web Text 999,999 Discussion Forums 99,063 TAC KBP Evaluation Workshop – NIST, November 18-19, 2013

Query Selection  Select queries and reference docs Temporal Slot Filling queries comprised of Entity – Slot – Filler Rich queries (2-3 pieces of temporal info) Evenly represent seven TSF slots per:spouse, per:title, per:employee_or_member_of, per:cities_of_residence, per:statesorprovinces_of_residence, per:countries_of_residence, org:top_employee_members  Link namestrings to KB or mark as NIL TAC KBP Evaluation Workshop – NIST, November 18-19, 2013

Query Selection  Select queries and reference docs Temporal Slot Filling queries comprised of Entity – Slot – Filler Rich queries (2-3 pieces of temporal info) Evenly represent seven TSF slots per:spouse, per:title, per:employee_or_member_of, per:cities_of_residence, per:statesorprovinces_of_residence, per:countries_of_residence, org:top_employee_members  Link namestrings to KB or mark as NIL TAC KBP Evaluation Workshop – NIST, November 18-19, 2013

Query Selection  Select queries and reference docs Temporal Slot Filling queries comprised of Entity – Slot – Filler Rich queries (2-3 pieces of temporal info) Evenly represent seven TSF slots per:spouse, per:title, per:employee_or_member_of, per:cities_of_residence, per:statesorprovinces_of_residence, per:countries_of_residence, org:top_employee_members  Link namestrings to KB or mark as NIL TAC KBP Evaluation Workshop – NIST, November 18-19, 2013

Annotation  For each query annotator spends up to 2 hours searching for temporal information for targeted relation  Annotators exhaustively add temporal information to system answers marked correct during assessment TAC KBP Evaluation Workshop – NIST, November 18-19, 2013 Ronnie James Dio per:employee_or_member_of: Black Sabbath BEGIN 1979-XX-XX END 1982-XX-XX

Annotation  For each query annotator spends up to 2 hours searching for temporal information for targeted relation  Annotators exhaustively add temporal information to system answers marked correct during assessment TAC KBP Evaluation Workshop – NIST, November 18-19, 2013 Ronnie James Dio per:employee_or_member_of: Black Sabbath BEGIN 1979-XX-XX END 1982-XX-XX

Assessment  Assess validity of fillers & justification from humans & systems Filler Correct – meets the slot requirements and entity supported in document Wrong – doesn’t meet slot requirements and/or entity not supported in doc Predicate Correct, Wrong, Inexact-Short, Inexact-Long Subject/Object Correct, Wrong, Inexact  In TSF only the entity must be supported by the source document, not the string No ‘Inexact’ fillers in TSF TAC KBP Evaluation Workshop – NIST, November 18-19, 2013

Justification  Justification is the string(s) of text that show a relation is true Predicate: Includes all three pieces of information necessary to justify the entity/slot/filler relation Subject: proves the entity’s involvement in the relation Object: proves the filler’s involvement in the relation Each part can be comprised of up to two, discontiguous strings Predicate 1: Dio died at his home in Los Angeles Predicate 2: Los Angeles, the second-most populous city in the US TAC KBP Evaluation Workshop – NIST, November 18-19, 2013 New in 2013: Ronnie James Dio - per:employee_or_member_of:

2014 Discoveries  Some restrictions lossy given prevalence of certain features in data Age used to resolve dates invalid Ronnie James Dio was born on July 10, 1942 […] […] Dio joined Black Sabbath when he was 37 Dates with unspecified years invalid XXXX  Some easy-to-add features could benefit data Offsets for temporal information provenance justification for temporal information TAC KBP Evaluation Workshop – NIST, November 18-19, 2013

Delivered 2013 Resources TAC KBP Evaluation Workshop – NIST, November 18-19, 2013 Corpus TitleTypeLDC CatalogLanguageSize TAC 2013 KBP English Temporal Slot Filling Training Queries and Annotations TrainingLDC2013E82English7 Queries TAC 2013 KBP English Temporal Slot Filling Evaluation Queries and Annotations EvaluationLDC2013E86English273 Queries TAC 2013 KBP English Temporal Slot Filling Evaluation Assessment Results EvaluationLDC2013E99English 4,376 Assessments