Graph-Based Knowledge Representation Model and Pattern Retrieval

Knowledge representation and pattern retrieval are the basis of knowledge discovery and reasoning. Different from many knowledge representation models such as production rules, graph model used to present context information in text has been envisioned as an appropriate solution to solve complex relevance more acceptably by the user. In this paper, a novel graph model, feature event dependency graph (FEDG) is proposed. FEDG emphasizes on representing the fact level knowledge compressively without losing important information. Meanwhile, based on this model, we propose retrieval and rank strategies for knowledge pattern retrieval which is meaningful for effective reasoning and latent knowledge discovery on large volumes of text knowledge. Extensive experiments on real knowledge sets, containing hundreds of domain specific rule based knowledge, demonstrate the feasibility and effectiveness of the proposed scheme.

[1]  William J. Clancey,et al.  Conceptual structures — Information processing in mind and machine J.F. Sowa, systems programming series (Addison-Wesley, Reading, MA, 1984); 481 pages, indices, appendices , 1985 .

[2]  Tetsuya Nasukawa,et al.  Text analysis and knowledge mining system , 2001, IBM Syst. J..

[3]  Wesley W. Chu,et al.  The phrase-based vector space model for automatic retrieval of free-text medical documents , 2007, Data Knowl. Eng..

[4]  Gerard Salton,et al.  A vector space model for automatic indexing , 1975, CACM.

[5]  John F. Sowa,et al.  Conceptual Graphs for a Data Base Interface , 1976, IBM J. Res. Dev..

[6]  John F. Sowa,et al.  Conceptual Structures: Information Processing in Mind and Machine , 1983 .

[7]  Kamal Kant Bharadwaj.  A parellel genetic algorithm approach for automated discovery of censored production rules , 2007, Artificial Intelligence and Applications.

[8]  G.H. Gillespie,et al.  Knowledge rule base for the beam optics program TRACE 3-D , 1993, Proceedings of International Conference on Particle Accelerators.

[9]  Janis Stirna,et al.  Introducing a pattern based knowledge management approach: the Verbundplan case , 2003, 14th International Workshop on Database and Expert Systems Applications, 2003. Proceedings..

[10]  Alexander F. Gelbukh,et al.  Information Retrieval with Conceptual Graph Matching , 2000, DEXA.

[11]  Georgios Meditskos,et al.  A Rule-Based Object-Oriented OWL Reasoner , 2008, IEEE Transactions on Knowledge and Data Engineering.

[12]  Raymond J. Mooney,et al.  Relational Learning of Pattern-Match Rules for Information Extraction , 1999, CoNLL.

[13]  Thiago Tinoco Pires.  A knowledge representation model for the nuclear power generation domain , 2007, Data Knowl. Eng..

[14]  Abraham Kandel,et al.  Classification of Web documents using a graph model , 2003, Seventh International Conference on Document Analysis and Recognition, 2003. Proceedings..