摘要
本文用改进的Apriori算法来减少基于输入输出的最小决策规则集,它由混合模糊聚类自动生成的隶属函数和规则构建而成。自适应模糊推理引擎采用最小二乘法和共轭梯度下降算法以实现用最少的规则集,从而得到更好的性能。
The present paper proposes a hybrid Modified Adaptive Fuzzy Inference Engine（MAFIE） for pattern classification. It is constructed by the automatic generation of membership functions and rules by the hybrid fuzzy clustering and Apriori algorithm technique, The generated adaptive fuzzy inference engine is adjusted by the least-squares fit and a conjugate gradient descent algorithm towards better performance with a minimal set of rules.
出处
《科技信息》
2013年第1期125-126,共2页
Science & Technology Information
