the rule extraction of neural networks is discussed, which is an effective method to avoid shortcoming of being “ black boxes ” . techniques based on decompositional and input-output mapping approaches are studied, their fundamental concepts and pedagogical rule extraction are compared in detail 神經(jīng)網(wǎng)絡中的規(guī)則提取方法是解決“黑箱問題”的有效手段,論文分析了基于結構分解和基于輸入輸出映射的神經(jīng)網(wǎng)絡規(guī)則提取的基本思想和對應的各種算法,并對它們的性能進行比較。