2004 Fiscal Year Final Research Report Summary
Study on Methods of Kansei Data Analysis
| Project/Area Number |
14580414
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Intelligent informatics
|
|---|
| Research Institution | Japan Advanced Institute of Science and Technology |
|---|
Principal Investigator |
NAKAMORI Yoshiteru Japan Advanced Institute of Science and Technology, School of Knowledge Science, Professor, 知識科学研究科, 教授 (30148598)
|
|---|
| Project Period (FY) |
2002 – 2004
|
| Keywords | Kansei Data / Multivariable Analysis / Knowledge Representation / Fuzzy Sets / Ensemble Model / Context Model |
|---|
| Research Abstract |
Sets of subjective evaluation values of objects by the subjective evaluation standard are called "sensibility data" or "kansei data" here. For instance, in case of the evaluation of products we often use the word "externals" or "gorgeous", in case of the evaluation of environments we often say "I feel comfortable", or "I feel touch with nature", and in case of the evaluation of people we often say "he is reliable" or "she is pleasant", etc. The sensibility data is usually given as evaluation values of five or seven levels. It is an extremely vague value invented by the emergence function in the brain that the individual doesn't understand well either. "Sensitivity data" contain "the complexity in the evaluation object itself" in addition to "the complexity in recoginition of people".
In this research, we have developed the identification method of "context model (context model)" that pays attention to vagueness in the evaluation, and "ensemble model" that pays attention to vagueness in the context. An "ensemble model" is a model that expresses the ensemble of sensibilities of people caught in relating to the whole, and a "context model" is a model that expresses the background, the scene, the situation, and the context in the evaluation. The "ensemble model" is an original model of treating the whole at the same time with the individual in one model. On the other hand, if the "context model" reads the context in a different way as the condition, it is a commonsense idea in the field of the artificial intelligence. It is thought that this research can contribute to the knowledge science in which a main topic is to transform tacit knowledge into explicit one.
|
