5th IFIP International Conference On

Intelligent Information Processing

18-22, October, 2008, Beijing, China

Sponsored by

    International Federation for Information Processing, IFIP TC12

Co-Sponsored by

Chinese Association of Artificial Intelligence

Institute of Computing Technology, Chinese Academy of Sciences

Lyon 3 University

Conference Organization

General Chairs

Danielle Boulanger (France) 

B. Wah(USA)

T. Nishida (Japan)

Program Chairs

Z. Shi (China)

E. Mercier-Laurent (France)

D. Leake (USA)

Plenary Speeches

Semantic Computing

Phillip C-y Sheu

EECS and Biomedical Engineering

University of California, Irvine

Abstract:  This talk highlights the past, present and future of semantic computing that brings together those disciplines concerned with connecting the (often vaguely-formulated) intentions  of humans with computational content. This connection can go both ways: retrieving,  using  and  manipulating  existing content according to user's goals ("do what the user means"); and creating, rearranging, and managing content that matches the  author's intentions ("do what the author means").

The content addressed in SC includes, but is not limited to, structured and semi-structured data, multimedia data, text, programs, services and, even, network behavior. This connection between content and the user is made via (1) Semantic  Analysis, which analyzes content with the goal of converting it to meaning (semantics); (2)Semantic Integration, which  integrates  content and  semantics  from multiple sources; (3)Semantic Applications, which utilize content and semantics to solve  problems;  and  (4)Semantic Interfaces, which attempt to interpret users' intentions expressed in natural language or other communicative forms.

The field Semantic Computing applies technologies in natural language processing, data and knowledge engineering, software engineering, computer systems and networks, signal processing and pattern recognition, and any combination of the above to extract, access, transform and synthesize the semantics as well as the contents of multimedia, texts, services and structured data.

Towards Brain-inspired Web Intelligence

Ning Zhong

Department of Life Science and Informatics

Maebashi Institute of Technology, Japan

Abstract:  Artificial Intelligence (AI) has been mainly studied within the realm of computer based technologies. Various computational models and knowledge based systems have been developed for automated reasoning, learning, and problem-solving.  However, there still exist several grand challenges.  The AI research has not produced major breakthrough recently due to a lack of understanding of human brains and natural intelligence.  In addition, most of the AI models and systems will not work well when dealing with large-scale, dynamically changing, open and distributed information sources at a Web scale.

  The next major advances in artificial intelligence and Web intelligence are most likely to be brought by an in-depth understanding of human intelligence and its application in the design and implementation of systems with human-level intelligence. To prepare us ready for the great opportunity, this talk outlines a unified framework for the study of brain inspired Web intelligence (WI) by exploring the latest results from brain informatics (BI). This leads to profound advances in the analysis and understanding of data, knowledge, intelligence and wisdom, as well as their inter-relationships, organization and creation process. The fast-evolving WI research and development initiatives are now moving towards understanding the multi-facet nature of intelligence in depth and incorporating it on a Web scale. The recently developed instrumentation (fMRI etc.) and advanced IT are causing an impending revolution in WI research and development, making it possible for us to pursue the new frontier of intelligence science and develop human-level Web intelligence. 

Data Mining Technologies Inspired from Visual Principle

Zongben Xu

Mathematics and computer science

Xi`an Jiaotong University, China

Abstract:  In this talk we review the recent work done by our group on data mining (DM) technologies deduced from simulating visual principle. Through viewing a DM problem as a cognition problems and treading a data set as an image with each light point located at a datum position, we developed a series of high efficient algorithms for clustering, classification and regression via mimicking visual principles. In pattern recognition, human eyes seem to possess a singular aptitude to group objects and find important structure in an efficient way. Thus, a DM algorithm simulating visual system may solve some basic problems in DM research. From this point of view, we proposed a new approach for data clustering by modeling the blurring effect of lateral retinal interconnections based on scale space theory. In this approach, as the data image blurs, smaller light blobs merge into large ones until the whole image becomes one light blob at a low enough level of resolution. By identifying each blob with a cluster, the blurring process then generates a family of clustering along the hierarchy. The proposed approach provides unique solutions to many long standing problems, such as the cluster validity and the sensitivity to initialization problems, in clustering. We extended such an approach to classification and regression problems, through combatively employing the Weber's law in physiology and the cell response classification facts. The resultant classification and regression algorithms are proven to be very efficient and solve the problems of model selection and applicability to huge size of data set in DM technologies. We finally applied the similar idea to the difficult parameter setting problem in support vector machine (SVM). Viewing the parameter setting problem as a recognition problem of choosing a visual scale at which the global and local structures of a data set can be preserved, and the difference between the two structures be maximized in the feature space, we derived a direct parameter setting formula for the Gaussian SVM. The simulations and applications show that the suggested formula significantly outperforms the known model selection methods in terms of efficiency and precision.

The advantages of the proposed approaches are: 1) The derived algorithms are computational stable and insensitive to initialization and they are totally free from solving difficult global optimization problems. 2) They facilitate the construction of new checks on DM validity and provide the final DM result a significant degree of robustness to noise in data and change in scale. 3) They are free from model selection in application. 4) The DM results are highly consistent with those perceived by our human eyes. 5) They provide unified frameworks for scale-related DM algorithms recently derived from many other fields such as estimation theory, recurrent signal processing, information theory and statistical mechanics, and artificial neural networks.