An Ant-Based Fast Text Clustering Approach Using Pheromone

Fast, effective text clustering has been playing an important role in organizing web text information. Due to its self-organization, robustness, flexibility, and visualization, the ant-based clustering approach has been applied to text clustering. In this approach, however, the ant´s moving is random, which leads to the convergence speed too slow. Aims at above mentioned problem, an ant-based fast text clustering approach (AFTC) is presented. This approach utilizes pheromone left by ants to avoid ant´s moving randomicity, which can make the ant move towards direction which has high pheromone concentration a teach step, and the direction of moving is the orientation where the text vectors are relatively concentration. Compared with the K-means algorithm,the ant-based fast text clustering approach performs the self-organization clustering process without setting clustering center number in advance, which can achieve better performance. The experimental results illustrate that the average time of ant-based fast text clustering approach is faster than that of traditional ant-based text clustering.