Towards incremental fuzzy classifiers

被引:34
作者
Bouchachia, Abdelhamid [1 ]
Mittermeir, Roland [1 ]
机构
[1] Univ Klagenfurt, Dept Informat Syst, Klagenfurt, Austria
关键词
incremental fuzzy rule learning; incremental and supervised clustering; classification; incremental feature selection;
D O I
10.1007/s00500-006-0077-3
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Fuzzy classification systems (FCS) are traditionally built from observations (data points) in an off-line one shot-experiment. Once the learning phase is exhausted, the classifier is no more capable to learn further knowledge from new observations nor is it able to update itself in the future. This paper investigates the problem of incremental learning in the context of FCS. It shows how, in contrast to off-line or batch learning, incremental learning infers knowledge in the form of fuzzy rules from data that evolves over time. To accommodate incremental learning, appropriate mechanisms are applied in all steps of the FCS construction: (1) Incremental supervised clustering to generate granules in a progressive manner, (2) Systematic and automatic update of fuzzy partitions, (3) Incremental feature selection using an incremental version of Fisher's interclass separability criterion. The effect of incrementality on various aspects is demonstrated via a numerical evaluation.
引用
收藏
页码:193 / 207
页数:15
相关论文
共 33 条
  • [1] Supervised fuzzy clustering for the identification of fuzzy classifiers
    Abonyi, J
    Szeifert, F
    [J]. PATTERN RECOGNITION LETTERS, 2003, 24 (14) : 2195 - 2207
  • [2] [Anonymous], P 10 C INF PROC MAN
  • [3] [Anonymous], P INT WORKSH SOFT CO
  • [4] [Anonymous], Pattern Recognition With Fuzzy Objective Function Algorithms
  • [5] Rough fuzzy MLP: Knowledge encoding and classification
    Banerjee, M
    Mitra, S
    Pal, SK
    [J]. IEEE TRANSACTIONS ON NEURAL NETWORKS, 1998, 9 (06): : 1203 - 1216
  • [6] BARANYI P, 2000, FUZZY IF THEN RULES, P135
  • [7] CARAGEA D, 2001, ARCHITECTURES INTELL
  • [8] FUZZY ART - FAST STABLE LEARNING AND CATEGORIZATION OF ANALOG PATTERNS BY AN ADAPTIVE RESONANCE SYSTEM
    CARPENTER, GA
    GROSSBERG, S
    ROSEN, DB
    [J]. NEURAL NETWORKS, 1991, 4 (06) : 759 - 771
  • [9] Predictive learning models for concept drift
    Case, J
    Jain, S
    Kaufmann, S
    Sharma, A
    Stephan, F
    [J]. THEORETICAL COMPUTER SCIENCE, 2001, 268 (02) : 323 - 349
  • [10] CASTELLANO G, 2003, P 7 MULT SYST CYB IN