AI helps you reading Science

AI generates interpretation videos

AI extracts and analyses the key points of the paper to generate videos automatically


pub
Go Generating

AI Traceability

AI parses the academic lineage of this thesis


Master Reading Tree
Generate MRT

AI Insight

AI extracts a summary of this paper


Weibo:
We propose an approach named correct propagation to improve the matching result with the confirmed error matches

Actively Learning Ontology Matching via User Interaction

International Semantic Web Conference, (2009): 585-600

Cited by: 65|Views66
WOS SCOPUS EI

Abstract

Ontology matching plays a key role for semantic interoperability. Many methods have been proposed for automatically finding the alignment between heterogeneous ontologies. However, in many real-world applications, finding the alignment in a completely automatic way is highly infeasible . Ideally, an ontology matching system would have an ...More

Code:

Data:

0
Introduction
  • The growing need of information sharing poses many challenges for semantic integration.
  • Ontology matching, aiming to obtain semantic correspondences between two ontologies, is the key to realize ontology interoperability [10].
  • With the success of many online social networks, such as Facebook, MySpace, and Twitter, a large amount of user-defined ontologies are created and published on the Social Web, which makes it much more challenging for the ontology matching problem.
  • Bernstein et al (Eds.): ISWC 2009, LNCS 5823, pp. 585–600, 2009. c Springer-Verlag Berlin Heidelberg 2009
Highlights
  • The growing need of information sharing poses many challenges for semantic integration
  • We propose a novel problem of ontology matching with active user interaction
  • We propose a correct propagation algorithm, which aims at detecting related error matches to the selected one
  • In this paper we propose an active learning framework for ontology matching
  • We propose an approach named correct propagation to improve the matching result with the confirmed error matches
  • We present a simple but effective method of selecting the threshold with user feedback, which is helpful for the error match selection
Methods
  • The authors present details of the experiments .

    5.1 Experiment Setup, Data, and Evaluation Methodology

    The authors implement all the algorithms using Java 2 JDK version 1.6.0 environment.
  • For the experiments of the first two groups, the authors use the OAEI 2008 30x benchmark [2].
  • The traditional matching results on these data sets is very high, it is very suitable for the first two experiments.
  • For the experiment of the correct propagation, the authors use part of the OAEI 2005 Directory benchmark [1], which consists of aligning web sites directory with more than two thousand elementary tests.
  • The reason the authors select this data set lies in its available ground truth and its low matching accuracy by the traditional methods [16]
Results
  • Experimental results on several public data sets show that the proposed approach can significantly improve the matching accuracy (+8.0% better than the baseline methods).
Conclusion
  • Conclusion and Future Work

    In this paper the authors propose an active learning framework for ontology matching.
  • The framework is just a shell, and what separates a successful instantiation from a poor one is the selection of matches to query and the approach to improve the traditional matching result with the confirmed matches by users.
  • The authors propose an approach named correct propagation to improve the matching result with the confirmed error matches.
  • The authors present a simple but effective method of selecting the threshold with user feedback, which is helpful for the error match selection.
Summary
  • Introduction:

    The growing need of information sharing poses many challenges for semantic integration.
  • Ontology matching, aiming to obtain semantic correspondences between two ontologies, is the key to realize ontology interoperability [10].
  • With the success of many online social networks, such as Facebook, MySpace, and Twitter, a large amount of user-defined ontologies are created and published on the Social Web, which makes it much more challenging for the ontology matching problem.
  • Bernstein et al (Eds.): ISWC 2009, LNCS 5823, pp. 585–600, 2009. c Springer-Verlag Berlin Heidelberg 2009
  • Methods:

    The authors present details of the experiments .

    5.1 Experiment Setup, Data, and Evaluation Methodology

    The authors implement all the algorithms using Java 2 JDK version 1.6.0 environment.
  • For the experiments of the first two groups, the authors use the OAEI 2008 30x benchmark [2].
  • The traditional matching results on these data sets is very high, it is very suitable for the first two experiments.
  • For the experiment of the correct propagation, the authors use part of the OAEI 2005 Directory benchmark [1], which consists of aligning web sites directory with more than two thousand elementary tests.
  • The reason the authors select this data set lies in its available ground truth and its low matching accuracy by the traditional methods [16]
  • Results:

    Experimental results on several public data sets show that the proposed approach can significantly improve the matching accuracy (+8.0% better than the baseline methods).
  • Conclusion:

    Conclusion and Future Work

    In this paper the authors propose an active learning framework for ontology matching.
  • The framework is just a shell, and what separates a successful instantiation from a poor one is the selection of matches to query and the approach to improve the traditional matching result with the confirmed matches by users.
  • The authors propose an approach named correct propagation to improve the matching result with the confirmed error matches.
  • The authors present a simple but effective method of selecting the threshold with user feedback, which is helpful for the error match selection.
Related work
  • 6.1 Ontology Matching

    Many works have addressed ontology matching in the context of ontology design and integration [6][17][19][21]. Some of them use the names, labels or comments of elements in the ontologies to suggest the semantic correspondences. [7] gives a detailed compare of various string-based matching techniques, including edit-distance [13] and token-based functions, e.g., Jaccard similarity [25] and TF/IDF [22]. Many works do not deal with explicit notions of similarity. They use a variety of heuristics to match ontology elements [17][19].

    Some other works consider the structure information of ontologies. [15] uses the cardinalities of properties to match concepts. The method of similarity flooding is also an example using structure information [18]. Another type of method utilizes the background knowledge to improve the performance of ontology matching. For example, [4] proposes a similarity calculation method by using thesaurus WordNet. [12] presents a novel approximate method to discover the matches between concepts in directory ontology hierarchies. It utilizes information from Google search engine to define the approximate matches between concepts. [5] makes semantic mappings more amenable to matching through revising the mediated schema. Other methods based on instances of ontologies [28] or reasoning [27] also achieve good results.
Funding
  • The work is supported by the Natural Science Foundation of China (No 60703059), Chinese National Key Foundation Research (No 2007CB310803), National High-tech R&D Program (No 2009AA01Z138), and Chinese Young Faculty Research Fund (No 20070003093)
Reference
  • Oaei directory download site (2005), http://oaei.ontologymatching.org/2005/
    Findings
  • Ontology alignment evaluation initiative, http://oaei.ontologymatching.org/
    Findings
  • Baily, T., Jain, A.K.: A note on distance-weighted k-nearest neighbor rules. IEEE Transaction System Man Cybern 8(4), 311–313 (1978)
    Google ScholarLocate open access versionFindings
  • Budanitsky, A., Hirst, G.: Evaluating wordnet based measures of lexical semantic relatedness. Computational Linguistics 32(1), 13–47 (2006)
    Google ScholarLocate open access versionFindings
  • Chai, X., Sayyadian, M., Doan, A., Rosenthal, A., Seligman, L.: Analyzing and revising mediated schemas to improve their matchability. In: Proceedings of the VLDB Conference 2008 (2008)
    Google ScholarLocate open access versionFindings
  • Chalupsky, H.: Ontomorph: A translation system for symbolic knowledge. In: Principles of Knowledge Representation and Reasoning, pp. 471–482 (2000)
    Google ScholarLocate open access versionFindings
  • Cohen, W., Ravikumar, P., Fienberg, S.: A comparison of string metrics for matching names and records. In: Proceedings of the IJCAI 2003 Workshop on Information Integration on the Web, pp. 73–78 (2003)
    Google ScholarLocate open access versionFindings
  • Doan, A., Madhavan, J., Domingos, P., Halevy, A.: Ontology matching: A machine learning approach. Springer, Heidelberg (2003)
    Google ScholarFindings
  • Ehrig, M., Staab, S., Sure, Y.: Bootstrapping ontology alignment methods with APFEL. In: Gil, Y., Motta, E., Benjamins, V.R., Musen, M.A. (eds.) ISWC 2005. LNCS, vol. 3729, pp. 186–200. Springer, Heidelberg (2005)
    Google ScholarLocate open access versionFindings
  • Euzenat, J., Shvaiko, P.: Ontology Matching. Springer, Heidelberg (2007)
    Google ScholarFindings
  • Fedorov, V.: Theory of Optimal Experiments. Academic Press, London (1972)
    Google ScholarFindings
  • Gligorov, R., Aleksovski, Z., Kate, W., Harmelen, F.: Using google distance to weight approximate ontology matches. In: Proceedings of the 16th International World Wide Web Conference (WWW), pp. 767–776 (2007)
    Google ScholarLocate open access versionFindings
  • Gusfield, D.: Algorithms on strings, trees, and sequences. Cambridge University Press, Cambridge (1997)
    Google ScholarFindings
  • Kapoor, A., Horvitz, E., Basu, S.: Selective supervision: Guiding supervised learning with decision-theoretic active learning. In: Proceedings of the IJCAI Conference 2007, pp. 877–882 (2007)
    Google ScholarLocate open access versionFindings
  • Lee, M., Yang, L., Hsu, W., Yang, X.: Xclust: Clustering xml schemas for effective integration. In: Proceedings of the 11th International Conference on Information and Knowledge Management (CIKM), pp. 292–299 (2002)
    Google ScholarLocate open access versionFindings
  • Li, Y., Zhong, Q., Li, J., Tang, J.: Result of ontology alignment with rimom at oaei07. OM, 1 (2007) 17. McGuinness, D., Fikes, R., Rice, J., Wilder, S.: The chimaera ontology environment. In: Proceedings of the 17th National Conference on Articial Intelligence (AAAI), pp. 1123–1124 (2000)
    Google ScholarLocate open access versionFindings
  • 18. Melnik, S., Garcia-Molina, H., Rahm, E.: Similarity flooding: A versatile graph matching algorithm and its application to schema matching. In: Proceedings of 18th International Conference of Data Engineering (ICDE), pp. 117–128 (2002)
    Google ScholarLocate open access versionFindings
  • 19. Mitra, P., Wiederhold, G., Jannink, J.: Semi-automatic integration of knowledge sources. In: Proceedings of the 2nd International Conference On Information FUSION (1999)
    Google ScholarLocate open access versionFindings
  • 20. Muslea, I.: Active learning with multiple views. PhD thesis, Department of Computer Science, University of Southern California (2002)
    Google ScholarFindings
  • 21. Noy, N., Musen, M.: Prompt: Algorithm and tool for automated ontology merging and alignment. In: Proceedings of the National Conference on Articial Intelligence (AAAI), pp. 450–455 (2000)
    Google ScholarLocate open access versionFindings
  • 22. Salton, G., Buckley, C.: Term-weighting approaches in automatic text retrieval. In: Information Processing and Management, pp. 513–523 (1988)
    Google ScholarLocate open access versionFindings
  • 23. Sarawagi, S., Bhamidipaty, A.: Interactive deduplication using active learning. In: Proceedings of the 8th ACM SIGKDD international conference on Knowledge discovery and data mining, pp. 269–278 (2002)
    Google ScholarLocate open access versionFindings
  • 24. Shvaiko, P., Euzenat, J.: Ten challenges for ontology matching. In: On the Move to Meaningful Internet Systems, OTM (2008)
    Google ScholarLocate open access versionFindings
  • 25. Tan, P., Steinbach, M., Kumar, V.: Introduction to Data Mining (2005)
    Google ScholarFindings
  • 26. Tang, J., Li, J., Liang, B., Huang, X., Li, Y., Wang, K.: Using bayesian decision for ontology mapping. Web Semantics 4(4), 243–262 (2006)
    Google ScholarLocate open access versionFindings
  • 27. Udrea, O., Getoor, L., Miller, R.J.: Leveraging data and structure in ontology integration. In: Proceedings of the 26th International Conference on Management of Data (SIGMOD 2007), pp. 449–460 (2007)
    Google ScholarLocate open access versionFindings
  • 28. Wang, S., Englebienne, G., Schlobach, S.: Learning concept mappings from instance similarity. In: Sheth, A.P., Staab, S., Dean, M., Paolucci, M., Maynard, D., Finin, T., Thirunarayan, K. (eds.) ISWC 2008. LNCS, vol. 5318, pp. 339–355. Springer, Heidelberg (2008)
    Google ScholarLocate open access versionFindings
Your rating :
0

 

Tags
Comments
小科