Abstract
In such areas as knowledge discovery, data mining and logical analysis of data, methodologies to nd relations among attributes are considered important. In this paper, given a data set (T;F) of a phenomenon, where T ⊆{0,1}n 1gn denotes a set of positive examples and F ⊆{0,1}ndenotes a set of negative examples, we propose a method to identify decomposable structures among the attributes of the data. Such information will reveal hierarchical structure of the phenomenon under consideration. We rst study computational complexity of the problem of nding decomposable Boolean extensions. Since the problem turns out to be intractable (i.e., NP-complete), we propose a heuristic algorithm in the second half of the paper. Our method searches a decomposable partition of the set of all attributes, by using the error sizes of almost-t decomposable extensions as a guiding measure, and then nds structural relations among the attributes in the obtained partition. The results of numerical experiment on synthetically generated data sets are also reported.
| Original language | English |
|---|---|
| Title of host publication | Computing and Combinatorics - 6th Annual International Conference, COCOON 2000, Proceedings |
| Editors | Peter Eades, Vladimir Estivill-Castro, Xuemin Lin, Arun Sharma, Ding-Zhu Du |
| Publisher | Springer Verlag |
| Pages | 396-406 |
| Number of pages | 11 |
| ISBN (Print) | 3540677879, 9783540677871 |
| Publication status | Published - Jan 1 2000 |
| Event | 6th Annual International Conference on Computing and Combinatorics, COCOON 2000 - Sydney, Australia Duration: Jul 26 2000 → Jul 28 2000 |
Publication series
| Name | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
|---|---|
| Volume | 1858 |
| ISSN (Print) | 0302-9743 |
| ISSN (Electronic) | 1611-3349 |
Other
| Other | 6th Annual International Conference on Computing and Combinatorics, COCOON 2000 |
|---|---|
| Country | Australia |
| City | Sydney |
| Period | 7/26/00 → 7/28/00 |
Fingerprint
All Science Journal Classification (ASJC) codes
- Theoretical Computer Science
- Computer Science(all)
Cite this
Logical analysis of data with decomposable structures. / Ono, Hirotaka; Makino, Kazuhisa; Ibaraki, Toshihide.
Computing and Combinatorics - 6th Annual International Conference, COCOON 2000, Proceedings. ed. / Peter Eades; Vladimir Estivill-Castro; Xuemin Lin; Arun Sharma; Ding-Zhu Du. Springer Verlag, 2000. p. 396-406 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 1858).Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Logical analysis of data with decomposable structures
AU - Ono, Hirotaka
AU - Makino, Kazuhisa
AU - Ibaraki, Toshihide
PY - 2000/1/1
Y1 - 2000/1/1
N2 - In such areas as knowledge discovery, data mining and logical analysis of data, methodologies to nd relations among attributes are considered important. In this paper, given a data set (T;F) of a phenomenon, where T ⊆{0,1}n 1gn denotes a set of positive examples and F ⊆{0,1}ndenotes a set of negative examples, we propose a method to identify decomposable structures among the attributes of the data. Such information will reveal hierarchical structure of the phenomenon under consideration. We rst study computational complexity of the problem of nding decomposable Boolean extensions. Since the problem turns out to be intractable (i.e., NP-complete), we propose a heuristic algorithm in the second half of the paper. Our method searches a decomposable partition of the set of all attributes, by using the error sizes of almost-t decomposable extensions as a guiding measure, and then nds structural relations among the attributes in the obtained partition. The results of numerical experiment on synthetically generated data sets are also reported.
AB - In such areas as knowledge discovery, data mining and logical analysis of data, methodologies to nd relations among attributes are considered important. In this paper, given a data set (T;F) of a phenomenon, where T ⊆{0,1}n 1gn denotes a set of positive examples and F ⊆{0,1}ndenotes a set of negative examples, we propose a method to identify decomposable structures among the attributes of the data. Such information will reveal hierarchical structure of the phenomenon under consideration. We rst study computational complexity of the problem of nding decomposable Boolean extensions. Since the problem turns out to be intractable (i.e., NP-complete), we propose a heuristic algorithm in the second half of the paper. Our method searches a decomposable partition of the set of all attributes, by using the error sizes of almost-t decomposable extensions as a guiding measure, and then nds structural relations among the attributes in the obtained partition. The results of numerical experiment on synthetically generated data sets are also reported.
UR - http://www.scopus.com/inward/record.url?scp=84949473307&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84949473307&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:84949473307
SN - 3540677879
SN - 9783540677871
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 396
EP - 406
BT - Computing and Combinatorics - 6th Annual International Conference, COCOON 2000, Proceedings
A2 - Eades, Peter
A2 - Estivill-Castro, Vladimir
A2 - Lin, Xuemin
A2 - Sharma, Arun
A2 - Du, Ding-Zhu
PB - Springer Verlag
ER -