English
◆ Annual Review 2001

Database Systems Laboratory


Sabhash Bhalla
Associate Professor

Nadia L. Berthouze
Assistant Professor

Many types of new data types are commonly used within the web data resources. This year the laboratory plans to organize a workshop on "Databases inNetworked Information System (DNIS 2002)" with international participation. Many aspects of web related research activity wll be discussed at the workshop. In addition to complexity in features such as multiple attributed data, many new types of raw data are emerging that need to be captured by DBMSs for information extraction. Many research e??orts are being made to make suitable Object-Relational architectures for spatial databases.

Most of the advances in techniques concentrate on capturing more meaning within data. Anumber of researchers are actively developing improved data management strategies usingXML (extensive markup language) technology. This provides a challenging area for study. The domains of activity include :

  • Mobile computing and databases for electronic commerce;
  • LargeNetworkManagement DatabaseManagement Systems, for information exchange through XML; and
  • Multi-media Synchronization applications
  • Data Mining and Data Warehousing applications

The Database Systems Laboratory has research activity concentrating on data modeling as well as transaction processing activity.

Data Modeling for Spatial Objects

Information Processing Systems of future will be a combination of integrated components. There will be components for intelligent problem solving, or decision making, components for specialized data processing andcomponents for shared information management. The applications will utilize a shared base of information. Some examples are - business automation, industrial automation, computer-aided design and manufacture, and cartography.

Workflow and Electronic Commerce

Traditional approaches to transaction management introduce elements of unpredictable delays during transaction processing. Thus, making these not suitable for adoption in new application environments. The techniques for time-critical transactions are applicable to Mobile databases and multimedia databases.

Digital Library Studies

The goal of this research is to study the feasibility to endow a computational system with the ability to process information. As an experimental platform the laboratory is building a web retrieval system that enables users to retrieve multimedia information carrying a given subjective (Kansei) content. Such system will prove useful in various areas of information technology such as online shopping, advertising industry, entertainment, design, wearable computers. The modeling issues are reformulated from the complementary points of view:

  • A multimedia data mining approach to address the mapping between impression (or Kansei) words and multimedia impressions,
  • An approach that involves the user in the modeling process.

An early evaluation of the extensions to retrieving data for music is under study.

Referred Journal Papers
[bhalla-001:2001]S. Bhalla. The Performance of an EAEcient Distributed Synchronization and Recovery Algorithm. The Journal of Supercomputing, 19(2):199-219, 2001.
Concurrency control based on distributed locking, introduces delays due to local processing and message communication. With improvements in processor speeds, there is a simultaneous increase in contrast, between these delays. In this study, anew approach has been proposed to reduce delays by enhancing local processing, and reducing the inter-site communication delays. The proposal presents an eAEcient synchronization algorithm for exchange of precedence in case of a con??ict. The technique is efficient in global time-stamp management, and incurs a minimal overall synchronization overhead.
[bhalla-002:2001]S. Bhalla and S.E.Madnick. Parallel On-the-fly Reading of an Entire Database Copy. International Journal of Computer Research, 10(4):99-108, 2001.
Databases support transaction update activities on continuous and on-going bases. Reading an entire database is needed for many applications. Existing techniques introduce blocking. These restrict the transaction update activity. We propose an algorithm that reads current database entities without interference with update activity. The algorithm is simple to implement as compared with previous proposals. It assigns a color to eachentity readby the global-read. Normal transactions commit by declaring a color for the co mmitted updates. Color markings are used to update and for generation of a consistent copy of the entire database, in an off-line manner.
[nadia-001:2001]N. Bianchi-Berthouze. Mining Multimedia Subjective Feedback. Journal of Intelligent Information Systems, Special Issue on Multimedia Data Mining, Kluwer, 19(1):43-59, 2002.
Understanding the mechanisms underlying visual impressions would benefit the development of personalized web-search engines. However, those mechanisms are difficult to explicit because they are intrisincally variable. We propose a methodology based on multimedia datamining techniques to construct models of the mapping between visual information and impression word.The continuous adaptation of themodel is supported byboth traditional relevance feedback and externalization processes expressed in a conceptual space.
[nadia-002:2001]N. Bianchi-Berthouze and L. Berthouze. Exploring Kansei in Multimedia Information. International Journal on Kansei Engineering, 2(2):1-10, 2001.
With information technologies now reaching our daily-life, it is necessary that systems can deal with users' sensitivity (or Kansei). Rather than modeling the biology of users' Kansei, we suggest a functional approach by modelling the translation process between di??erent modalities of expression of a same Kansei concept. Grounding this mapping into the categorization of users' perceptions, we propose computational agents that are able to learn categories in visual perceptions and evolve a translation language between categories and impression words, through interaction with the user. The methodology is succesfully applied to the filtering of images against a given user Kansei impression.
Referred Proceeding Papers
[nadia-003:2001]N. Bianchi-Berthouze. An Interactive Environment for Kansei Data Mining. In O.R. Zaiane and Editor S.J. Simo??, editors, Proceeding of International workshop on multimedia data mining, in conjunction with the Seventh ACMSIGKDD International Conference on Knowledge Discovery & Data Mining, pages 58-67, S.Francisco, CA, USA, August 2001. ACM-SIGKDD, Quality Color Press Inc Edmonton.
We discuss the modelling of visual impressions from the point of view of multimedia data mining.We hypothesize that visual impressions are highly subjective and intrinsically variable because of the high dimensionality of the information contained in visual perceptions. To support the modeling process, we propose an interactive environment integrating different techniques of multimedia datamining. The user is led to assume an active role in the mining activity, through externalization processes.
[nadia-004:2001]
Unrefereed Papers
N. Bianchi-Berthouze. OnKansei-BasedMultimediaInformation Systems. In Riccardo Leonardi, editor, Proceedings of the International Workshop Content-Based Multimedia Indexing, pages 333-340, Brescia, Italy, September 2001. with the support of IEEE Signal Processing Society, DEA.
In this paper, we present a case-study on the interpretation of user Kansei messages encoded in multimedia information. We present an environment in which computation agent and user cooperate in the construction of anadaptive Kansei user model for image retrieval. While the experimental case-studyfocuses on visual perceptions, the framework is applicable to other modalities such as auditory perceptions.
Chapters in Book
[bhalla-003:2001]S. Bhalla and S.E.Madnick. Parallel On-the-??y Reading of an Entire Database Copy, pages 149-162. Number 9 in Practical Parallel Computating. Nova Science Publishers, New York, 2001.
[nadia-005:2001]N. Bianchi-Berthouze and L. Berthouze. Approach Dynamique de la Cognition Artificielle, chapter Development de la communication subjective. Hermes Science Press, Paris, France, to be published. editor:A. Guillot and E. Dauce
Grants
[nadia-006:2001]N. Berthouze and T. Izumita. Affectively-Capable Personal Digital Assistants, 2001-2002.
Ph.D and Other Thesis
[bhalla-004:2001]Tomoyuki Hosozawa. Graduation Thesis: Adopting XML for Mapsin aWebBasedGeographic Information System, UniversityofAizu, 2002.
Thesis Advisor: Subhash Bhalla.
[bhalla-005:2001]Hajime Sasaki. Graduation Thesis: Supporting Dynamic Contents in Web Based Geographic Information Systems, University of Aizu, 2002.
Thesis Advisor: Subhash Bhalla.
[bhalla-006:2001]KeisukeWatanabe. Graduation Thesis: Data Management System for a Web Based Geographic Information System, University of Aizu, 2002.
Thesis Advisor: Subhash Bhalla.
[bhalla-007:2001]Toru Sato. Graduation Thesis: Improving EAEciency of a Reference for a Dictionary of Japanese Characters and Words, University of Aizu, 2002.
Thesis Advisor: Subhash Bhalla.
[nadia-007:2001]Hasegawa Ryo. Graduation Thesis: A Dynamic Topology Classifier for Kansei User Modeling, University of Aizu, 2002.
Thesis Advisor: William Martens.
[nadia-008:2001]Hayashi Tomofumi. Graduation Thesis: Database Technology to Support Kansei Data Mining, University of Aizu, 2002.
Thesis Advisor: Subbash Bhalla.
[nadia-009:2001]Akio Sato. Graduation Thesis: Clustering Techniques for Word Nuances Detection, University of Aizu, 2002.
Thesis Advisor: William Martens.
Others
[nadia-010:2001]N. Berthouze. Member of the Organizing Commitee for the International Workshop onContext-Based Multimedia Indexing, September 2001.
Asian Liason
[nadia-011:2001]Interviewed by J@pan Inc Magazine. Interviewer: Sam Joseph. both hardcopy and electronic version, November 2001.
http://www.japaninc.com/article.php?articleID=514&page=3