模糊模式識別是模糊集理論研究中的重要方向，神經(jīng)網(wǎng)絡是數(shù)據(jù)挖掘中的一種常用方
法。超圓神經(jīng)網(wǎng)絡的學習時間和網(wǎng)絡模型理解性都優(yōu)于BP 神經(jīng)網(wǎng)絡，它能以較少的數(shù)據(jù)量 蘊涵同樣的信息量。文章依據(jù)超圓神經(jīng)網(wǎng)絡模型思想，提出了一種新的基于模糊模式識別的 神經(jīng)網(wǎng)絡模型算法，該算法繼承了超圓神經(jīng)網(wǎng)絡的優(yōu)點，能有效地對樣本進行學習。
關健詞：模糊集合論；數(shù)據(jù)挖掘；模糊模式識別；超圓神經(jīng)網(wǎng)絡

Research of the Algorithm of Artificial Neural Network based on Fuzzy Pattern Recognition
Liu Bing-xiang，Li Hai-lin，Li Hui-ying (School of Information Engineering, JDZ Ceramic Institute, JingDeZhen, 333001) Abstract: Fuzzy Pattern Recognition is an important research field on the fuzzy theory, Artificial neural network (ANN) is a common method of data mining. The learning period and network model understand of the super circle covering ANN are better than BP neural network. According to its thought, the new Artificial Neural Network based on fuzzy pattern recognition will be introduced. Comparing with the super circle covering ANN, this algorithm has the same excellent? aspect, in practice it is useful to learn the samples.
Key words: fuzzy sets；data mining；fuzzy pattern recognition；super circle covering artificial
neural network