Improved SOM algorithm-EDSOM applied in text clustering

SOM neural network is one of the most commonly used clustering algorithom in the text clustering field, but the learning strategies of SOM Network Clustering does not pursue the goal of the text clustering - the smallest deviation of the clusters, so it is very difficult to get the clusters with the smallest deviation. According to the principle of equal deviation, this paper presents an improved learning strategy: introduce equal cluster deviation theory into the learning process of SOM neural network, guide SOM neural network learning through adjusting the cluster deviation to be equal, with an expect to get clusters with the smallest deviation. The experimental results show that: by the measurement of the average accuracy, the algorithm shows a good performance.