INCREMENTAL CONCEPT-FORMATION ALGORITHMS BASED ON GALOIS (CONCEPT) LATTICES

被引:299
|
作者
GODIN, R
MISSAOUI, R
ALAOUI, H
机构
[1] Département D'informatique, Université du Québec À Montréal, Montréal, Quebec, H3C 3P8, CP. 8888, Succursale “Centre Vtlle
关键词
CONCEPT LATTICE; GALOIS LATTICE; CONCEPTUAL CLUSTERING; INCREMENTAL ALGORITHM; CONCEPT FORMATION;
D O I
10.1111/j.1467-8640.1995.tb00031.x
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
The Galois (or concept) lattice produced from a binary relation has proved useful for many applications. Building the Galois lattice can be considered a conceptual clustering method because it results in a concept hierarchy. This article presents incremental algorithms for updating the Galois lattice and corresponding graph, resulting in an incremental concept formation method. Different strategies are considered based on a characterization of the modifications implied by such an update. Results of empirical tests are given in order to compare the performance of the incremental algorithms to three other batch algorithms. Surprisingly, when the total time for incremental generation is used, the simplest and less efficient variant of the incremental algorithms outperforms the batch algorithms in most cases. When only the incremental update time is used, the incremental algorithm outperforms all the batch algorithms. Empirical evidence shows that, on the average, the incremental update is done in time proportional to the number of instances previously treated. Although the worst case is exponential, when there is a fixed upper bound on the number of features related to an instance, which is usually the case in practical applications, the worst-case analysis of the algorithm also shows linear growth with respect to the number of instances.
引用
收藏
页码:246 / 267
页数:22
相关论文
共 50 条
  • [31] Power contexts and their concept lattices
    Guo, Lankun
    Huang, Fangping
    Li, Qingguo
    Zhang, Guo-Qiang
    DISCRETE MATHEMATICS, 2011, 311 (18-19) : 2049 - 2063
  • [32] LOWER AND UPPER CONCEPT LATTICES
    Vaideanu, Cristian
    ANALELE STIINTIFICE ALE UNIVERSITATII AL I CUZA DIN IASI-SERIE NOUA-MATEMATICA, 2014, 60 (02): : 337 - 352
  • [33] Isomorphic generating of concept lattices
    Shen, XJ
    Liu, ZT
    Zhang, Q
    Li, Y
    Shi, BS
    2005 IEEE INTERNATIONAL CONFERENCE ON GRANULAR COMPUTING, VOLS 1 AND 2, 2005, : 249 - 252
  • [34] Incomplete information and concept lattices
    Krupka, Michal
    Lastovicka, Jan
    INTERNATIONAL JOURNAL OF GENERAL SYSTEMS, 2017, 46 (03) : 235 - 274
  • [35] Concept Formation and Concept Grounding
    Jörgen Sjögren
    Christian Bennet
    Philosophia, 2014, 42 : 827 - 839
  • [36] On efficient factorization of standard fuzzy concept lattices and attribute-oriented fuzzy concept lattices
    Konecny, Jan
    FUZZY SETS AND SYSTEMS, 2018, 351 : 108 - 121
  • [37] Decomposition of Relations and Concept Lattices
    Berghammer, Rudolf
    Winter, Michael
    FUNDAMENTA INFORMATICAE, 2013, 126 (01) : 37 - 82
  • [38] A new case-based classification using incremental concept lattice knowledge
    Muangprathub, Jirapond
    Boonjing, Veera
    Pattaraintakorn, Puntip
    DATA & KNOWLEDGE ENGINEERING, 2013, 83 : 39 - 53
  • [39] Multi-scaled concept lattices based on neighborhood systems
    Li Ma
    Ju-Sheng Mi
    Bin Xie
    International Journal of Machine Learning and Cybernetics, 2017, 8 : 149 - 157
  • [40] Multi-scaled concept lattices based on neighborhood systems
    Ma, Li
    Mi, Ju-Sheng
    Xie, Bin
    INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS, 2017, 8 (01) : 149 - 157