自组织映射SOM的主要目的是将任意维数的输入信号模式转变为一维或者二维的离散映射，并且以拓扑有序的方式自适应实现这个变换。呈现给网络的每一个输入模式，通常包含面对平静背景的一个局部化活动区域或“点”。这个点的位置和性质通常随输入模式的实现不同而不同。因此输入网络中所有神经元应经历输入模式的足够次数的不同实现，确保有机会完成恰当的自组织过程。自组织映射的形成主要有三个过程：竞争、合作、突触调节。-The main purpose of self-organizing map SOM is the input signal pattern of arbitrary dimension into a one-dimensional or two-dimensional discrete map, and in an orderly manner adaptive topology to achieve this transformation. Each input pattern is presented to the network, usually containing the face of a localized activity area or " point"　calm background. The location and nature of the point input mode is typically achieved with a different type. Therefore, the input of all the neurons in the network input mode should be a sufficient number of different implementations of experience to ensure that the opportunity to complete the appropriate self-organization process. Self-organizing map form three main processes: competition, cooperation and synaptic regulation.