Hoai-Viet To1, Ryutaro Ichise2, and Hoai-Bac Le1

Slides:



Advertisements
Similar presentations
EER to Relation Models Mapping
Advertisements

Texas Conference of Urban Counties TIJIS Update TCJUIG Conference May 4, 2011 Tarrant County 1/8/20141.
Previous training sessions by our Consultants 1/9/20141 Presented by Achieve Strategies International.
Monday, January 13, Instructor Development Unit 1 Instructional Responsibilities Ed Humphrey.
Monday, January 13, Instructor Development Strand 7 / Lesson 8.
Monday, January 13, Instructor Development Unit 1 Instructional Responsibilities Ed Humphrey.
Monday, January 13, Instructor Development Lesson 9.
1/15/ A car starts from rest and travel 10s with uniform acceleration 5m/s 2. Find out its final velocity. Here, u=0 m/s t=10s a= 5m/s 2 v=? We.
Dr. Peter OReilly Chairperson- ISM Services Group /23/20141 NAPM-AZ Presentation- March 2009.
Instituto Politécnico do Porto Escola Superior de Tecnologia e Gestão de Felgueiras Mestrado de Redes e Segurança de Computadores A. Paulo.
IEEE Symposium on Pre- University Teacher Training Teacher In-Service Program (TISP) Tampa, 2012 Teacher In Service Program in Uruguay Summary of Workshops.
National Seminar on Developing a Program for the Implementation of the 2008 SNA and Supporting Statistics in Turkey Arzu TOKDEMİR 10 September 2013 Ankara.
NAESB OASIS Recommendation
GOES-East Co-Location Operations with BRASILSAT-B1 Kevin T. Work ASRC Aerospace February 11, 2008.
State the purpose of the presentation. State the outcome of the presentation. (expectation – go/no decision) 2/14/20142St. Louis Public Schools.
2/14/20141 Wabasso: Utilizing the PACE Methodology As a Tool for Improving the Quality of Life in Communities Presentation by: Julianne Price Indian River.
BUS 220: ELEMENTARY STATISTICS
IEEE Chapter Symposium
U.S. ENVIRONMENTAL PROTECTION AGENCY For Conference Use Only Publishing EPA Registry Content as Linked Open Data Michael Pendleton Office of Environmental.
Identifying and Accessing Relevant Public and Private Databases: Trademarks Amanda Fila Myers Economist US Patent and Trademark Office
18/02/20141 ©Sustainable Cities Research Institute Conceptualising Urban-Rural Linkages for Sustainable Development Bob Evans Sustainable Cities Research.
Inclusions Regulation Testing Situations Decide if the situation is a violation or not a violation. Cite the page and paragraph number from the document(s)
Administration Code Testing Situations Decide if the situation is a violation or not a violation. Cite the page and paragraph number from the document(s)
3/28/20141 Professional Learning Communities Overview On-Site Professional Development.
4/6/20141 GC16/3011 Functional Programming Lecture 8 Designing Functional Programs (2)

Nordic Council of Ministers Friday, May 30, The Nordic Council of Ministers and the EU Baltic Sea Strategy.
Monday, June 02, 20141ABS Confence Beginning to Transform Assessment and Feedback for Institutional Change at MMU Rod Cullen and Rachel Forsyth, Manchester.
Attributes for Classifier Feedback Amar Parkash and Devi Parikh.
Oracle Rally Applications Modernization. 4 June About the Company Founded in 2002 Unites high-level information technology and organization architecture.
The First Few Pages By: Jeton Mehmeti Joshi Dibyadhar
First Experimental Tests 08/04/20141/18. First Experimental Tests Temperature sensors 08/04/20142/18.
© 2007 Cisco Systems, Inc. All rights reserved. 1 Valašské Meziříčí Networking Media.
6/10/20141 Top-Down Clustering Method Based On TV-Tree Zbigniew W. Ras.
Submission Writing Master Class Gerard Byrne B Comm FCPA FAIM Townsville, 17 April 2010 Thursday, June 12,
OMHARN Women's Health Panel Dr Olive Wahoush March 17 th /06/20141.
6/13/20141 Course Syllabus Service Operations Management (THM 348)
6/14/20141 A Cluster Formation Algorithm with Self-Adaptive Population for Wireless Sensor Networks Luis J. Gonzalez.
Intersection Schemas as a Dataspace Integration Technique 8/21/20141 Richard BrownlowAlex Poulovassilis.
SWFDP-SA: Evolution, challenges and successes Mark Majodina South African Weather Service 1 October 20141FCAST-PRE
10/4/20141 WP2 Discovery mechanism of the OpenKnowledge system (“Semantic routing”) (presented by Ronny Siebes) OpenKnowledge project review WP2 -Discovery.
10/8/20141 DV for Tax Module 5 Wage Item Validation.
More work with Weak Acids, Ka, and pH April 25, 2012.
New Successes – New Standards for English Learners March 22, 2014 TNTESOL Jan Lanier.
10/22/20141 International Trade and Exchange Rates Chapter 12.
10/22/20141 Money and Banking Chapter Outline The Functions of Money The Functions of Money The Components of Money Supply The Components of Money.
Exchange of Performance Assessments EPA Project State, Outlook Urs Hassler – Project leader ETH Zurich
Create Reports Individual Student Reports > 90 Days Old KDE:OAA:pp: 9/5/2014 1
Propositional Predicate
Maurice Hermans.  Ontologies  Ontology Mapping  Research Question  String Similarities  Winkler Extension  Proposed Extension  Evaluation  Results.
GENERATING AUTOMATIC SEMANTIC ANNOTATIONS FOR RESEARCH DATASETS AYUSH SINGHAL AND JAIDEEP SRIVASTAVA CS DEPT., UNIVERSITY OF MINNESOTA, MN, USA.
Watching Unlabeled Video Helps Learn New Human Actions from Very Few Labeled Snapshots Chao-Yeh Chen and Kristen Grauman University of Texas at Austin.
Predicting Missing Provenance Using Semantic Associations in Reservoir Engineering Jing Zhao University of Southern California Sep 19 th,
OMAP: An Implemented Framework for Automatically Aligning OWL Ontologies SWAP, December, 2005 Raphaël Troncy, Umberto Straccia ISTI-CNR
BACKGROUND KNOWLEDGE IN ONTOLOGY MATCHING Pavel Shvaiko joint work with Fausto Giunchiglia and Mikalai Yatskevich INFINT 2007 Bertinoro Workshop on Information.
Machine Learning Approach for Ontology Mapping using Multiple Concept Similarity Measures IEEE/ACIS International Conference on Computer and Information.
Active Learning An example From Xu et al., “Training SpamAssassin with Active Semi- Supervised Learning”
Training dependency parsers by jointly optimizing multiple objectives Keith HallRyan McDonaldJason Katz- BrownMichael Ringgaard.
Presenter: Lung-Hao Lee ( 李龍豪 ) January 7, 309.
Probabilistic Graphical Models for Semi-Supervised Traffic Classification Rotsos Charalampos, Jurgen Van Gael, Andrew W. Moore, Zoubin Ghahramani Computer.
Towards Distributed Information Retrieval in the Semantic Web: Query Reformulation Using the Framework Wednesday 14 th of June, 2006.
CoCQA : Co-Training Over Questions and Answers with an Application to Predicting Question Subjectivity Orientation Baoli Li, Yandong Liu, and Eugene Agichtein.
Number Sense Disambiguation Stuart Moore Supervised by: Anna Korhonen (Computer Lab)‏ Sabine Buchholz (Toshiba CRL)‏
Iterative similarity based adaptation technique for Cross Domain text classification Under: Prof. Amitabha Mukherjee By: Narendra Roy Roll no: Group:
Introduction to Machine Learning August, 2014 Vũ Việt Vũ Computer Engineering Division, Electronics Faculty Thai Nguyen University of Technology.
Coached Active Learning for Interactive Video Search Xiao-Yong Wei, Zhen-Qun Yang Machine Intelligence Laboratory College of Computer Science Sichuan University,
Websoft Research Group
Assoc. Prof. Dr. Syed Abdul-Rahman Al-Haddad
Actively Learning Ontology Matching via User Interaction
Hierarchical, Perceptron-like Learning for OBIE
Presentation transcript:

Hoai-Viet To1, Ryutaro Ichise2, and Hoai-Bac Le1 An Adaptive Machine Learning Framework with User Interaction for Ontology Matching Hoai-Viet To1, Ryutaro Ichise2, and Hoai-Bac Le1 1 Ho Chi Minh University of Science, Vietnam 2 National Institute of Informatics, Japan 3/27/2017

Ontology Matching (OM) Problem Ontology is a hierarchical structure used to organize concepts. Ontology plays an important role in semantic web development. Ontology matching finds correspondences between concepts from two ontologies. Ontology matching is an important process when we want to integrate heterogeneous information source in new semantic web environment. 3/27/2017

Machine Learning Framework for OM We introduced a machine learning framework for ontology matching problem in [Ichise, 2008] Our hypothesis: the use of semi-supervised learning method will reduce the manual annotation cost. Cb1 Ca1 Ca2 Ca3 Cb2 Cb3 ID Sim1 … Simn Class Ca1  Cb1 0.5 0.7 1 Ca1  Cb2 0.3 0.56 Pre-alignment Correct mapping: Ca1  Cb1 Ca2  Cb1 … Incorrect mapping Ca1  Cb2 Ca1  Cb3… 3/27/2017

Semi-supervised Learning with User Interaction Basic idea: propagate label through unlabeled data Problem: few samples of labeled data  low confidence prediction. ? Red ? Blue User Interaction 3/27/2017

Adaptive Machine Learning Framework Use multiple learning strategies + user interaction Ontology Storage learner training Initialize labeling User Interaction Ontology Parser Similarity Calculator learner Pre-alignment training 3/27/2017

Adaptive Machine Learning Framework Similarity measures are based on those used in machine learning framework proposed in [Ichise, 2008], which: include 24 string-based similarity measures calculate similarity between: concept feature, concept structure feature, and concept hierarchical feature. Our system: Machine Learning Framework for Ontology Matching with User Interaction (MalfomUI) 3/27/2017

Experiments Purpose: Compare the performance of our learning framework with other matching systems. General setting: Dataset from directory track of OAEI 2008’s campaign. [Caracciolo et. al., 2008] The dataset is constructed from three internet directories: Yahoo, Google, Looksmart. Simple equivalent relation. The dataset includes 4487 labeled matching tasks, in which there are 2160 positive samples and 2327 negative samples. Base learner: Naïve Bayes 3/27/2017

Experiments Pre-Experiment – Supervised Learning method: Used as baseline to compare with semi-supervised learning method. Study the effect of training-set size on the performance of the supervised learning method. 3/27/2017

Experimental Results MalUI-5 to MalUI-4000: Training set size Add training size Training set size 3/27/2017

Experiments Experiment – Semi-supervised learning with user interaction Study the performance of semi-supervised learning method with user interaction. User annotate 20 samples at initialize phase and then label 4 samples more in 2 feedback round. 3/27/2017

Comparison with other matching systems [Caracciolo et. al., 2008] Experimental Results MalUI-RF: Change the size Comparison with other matching systems [Caracciolo et. al., 2008] 3/27/2017

Experimental Results Semi-supervised learning with user feedback can reduce the cost of manual annotation. * In MalfomUI-RF experiment, users need to label 28 samples in total. MalUI -30 MalUI -100 MalUI -500 MalUI -4000 MalUI –RF Precision 0.56 0.62 0.68 0.61 Recall 0.59 0.63 0.74 0.75 0.73 F-Measure 0.58 0.71 0.67 Note the sample labeled in RF 3/27/2017

Conclusion Conclusions: Our adaptive machine learning framework is effective: it requires less annotation cost but gains approximately good performance. Machine learning approaches with user interaction are promising for ontology matching systems. Future works: Integrate more similarity measures to cover real datasets. Consider more complicate semi-supervised models. 3/27/2017