Scalable discovery of informative structural concepts using domain knowledge

被引:12
作者
Cook, DJ [1 ]
Holder, LB [1 ]
Djoko, S [1 ]
机构
[1] BELL NO RES, SCI STAFF, RICHARDSON, TX USA
来源
IEEE EXPERT-INTELLIGENT SYSTEMS & THEIR APPLICATIONS | 1996年 / 11卷 / 05期
基金
美国国家航空航天局;
关键词
D O I
10.1109/64.539018
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
The subdue system evaluates the benefits of using domain knowledge to guide the discovery of repetitive, functional substructures in large structural databases. Results show that domain-specific knowledge improves the search for such substructures and enables greater data compression. © 1996 IEEE.
引用
收藏
页码:59 / 68
页数:10
相关论文
共 10 条
[1]   Inexact graph matching for structural pattern recognition [J].
Bunke, H. ;
Allermann, G. .
PATTERN RECOGNITION LETTERS, 1983, 1 (04) :245-253
[2]   MACHINE DISCOVERY OF PROTEIN MOTIFS [J].
CONKLIN, D .
MACHINE LEARNING, 1995, 21 (1-2) :125-150
[3]   Substructure Discovery Using Minimum Description Length and Background Knowledge [J].
Cook, Diane J. ;
Holder, Lawrence B. .
JOURNAL OF ARTIFICIAL INTELLIGENCE RESEARCH, 1993, 1 :231-255
[4]  
DJOKO S, 1995, THESIS U TEXAS ARLIN
[5]  
FAYYAD UM, 1996, ADV KNOWLEDGE DISCOV
[6]  
LEVINSON R, 1984, P NAT C ART INT, P203
[7]  
QUINLAN JR, 1989, INFORM COMPUT, V80, P227, DOI 10.1016/0890-5401(89)90010-2
[8]  
Rissanen J., 1989, STOCHASTIC COMPLEXIT
[9]  
SEGEN J, 1988, 5TH P INT C MACH LEA, P29
[10]  
THOMPSON K, 1991, CONCEPT FORMULATION