Presentation | 1999/5/28 Causal Modeling through Factorial Representation Kazunori Watase, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | Blalock proposed the methods that obtained a causal model through thinking experiment. However these methods become difficult to apply as the variables increase. Furthermore, it is impossible often to specify the model from only correlation coefficients. Therefore, the knowledge regarding the order relations between variables is required to make a model. In model research, the order relations have to be considered. So, in this paper, we represent a permutation by a factorial scale. AS a result, it is possible to search models remaining order relations and to enumerate permutations that satisfy order conditions. Also, we understood that the factorial representations could be applied in simulated annealing. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | causal modeling / factorial representation / enumeration / heuristic |
Paper # | AI99-19 |
Date of Issue |
Conference Information | |
Committee | AI |
---|---|
Conference Date | 1999/5/28(1days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | |
Chair | |
Vice Chair | |
Secretary | |
Assistant |
Paper Information | |
Registration To | Artificial Intelligence and Knowledge-Based Processing (AI) |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Causal Modeling through Factorial Representation |
Sub Title (in English) | |
Keyword(1) | causal modeling |
Keyword(2) | factorial representation |
Keyword(3) | enumeration |
Keyword(4) | heuristic |
1st Author's Name | Kazunori Watase |
1st Author's Affiliation | Nagasaki Institute of Applied Science() |
Date | 1999/5/28 |
Paper # | AI99-19 |
Volume (vol) | vol.99 |
Number (no) | 96 |
Page | pp.pp.- |
#Pages | 8 |
Date of Issue |