- Organization
| Contact Person |
Shinichi
Morishita |
| Affiliation |
Institute of Medical Science, University of Tokyo |
| Title |
Visiting Associate Professor |
| Research Area |
Database Technology, Data Mining |
| Last Graduated & Year Graduated |
University of Tokyo, 1983 |
| Degree |
Doctor of Science |
| Date of Birth |
September 18, 1960 |
| Research Partner |
Akihiro Nakaya |
| Affiliation |
Institute of Medical Science, University of Tokyo |
| Title |
Research Associate |
| Research Area |
Parallel Computing, Data Mining |
| Last Graduated & Year Graduated |
University of Tokyo, 1992 |
| Degree |
Master of Science |
| Year of Birth |
1970 |
- Project Name
Knowledge Discovery from Genome Databases
- Objective
Predicting the value of an unseen data has been a major research topic
of knowledge discovery and data mining. The crux of increasing the prediction
accuracy is to invent an intelligent and computationally efficient way
of splitting data. We have studied data splitting by range/region and conjunctions
of primitive tests, and in both cases we have implemented efficient algorithms.
In this project we will investigate how to parallelize those algorithms
invented so far on SMP architecture machines such as SUN E10000, NUMA architecture
machines such as SGI Origin 2000, and SMP clusters. Based on the technology,
we will focus on the discovery of multiple genes causing some common diseases.
- List of Selected Publications
- S. Morishita, "Avoiding Cartesian Products for Multiple Joins,"
Journal of the ACM, Volume 44, Number 1, pp. 57-85, January 1997
- S. Morishita, ``An Extension of Van Gelder's Alternating Fixpoint to
Magic Programs,'' (Invited Paper) Journal of Computer and System Sciences,
Academic Press, Volume 52, Number 3, pp. 506-521, June 1996
- Y. Morimoto, H. Ishii, and S. Morishita, "Efficient Construction
of Regression Trees with Range and Region Splitting," Proceedings
of VLDB'97, pages 166-175, Athens, Greece, August 1997
- K. Yoda, T. Fukuda, Y. Morimoto, S. Morishita, and T. Tokuyama, "Computing
Optimized Rectilinear Regions for Association Rules," Proceedings
of the Third Conference on Knowledge Discovery and Data Mining (KDD97),
pages 96-103, Los Angels, August 1997
- T. Fukuda, Y. Morimoto, S. Morishita, and T. Tokuyama, "Data Mining
Using Two-Dimensional Optimized Association Rules: Scheme, Algorithms,
and Visualization," Proceedings of the 1996 ACM SIGMOD International
Conference on Management of Data, pp. 13-23, Montreal, Canada, June 1996
|