Research And Implementation Of Heuristic Fast Personal Blog Clustering Technology

Blog has been the important platform for people to record and share their life and work. For the person, whether the blogs have been properly archive and describe is important for them. In this paper, we propose heuristic blog clustering technology, which is different from previous blogs cluster work, we focus on the personal characteristics, cluster the personal blogs, and describe them.We design a heuristic search for similar blogs according to the personal characteristics to add to content for clustering. Similar blogs is obtained based on the user similarity. We propose a new user similarity which is based on the user’s interesting sets and the interesting sequence related to time. And about the cluster description, We propose a semi-automatic method based on the self blog platform. We combine the artificial description and automatic description, Firstly, extract the semantic topic from the dataset, which comes from the self blog platform. Then give the topic artificial description, and match the topic to the cluster topic, then we can get the cluster topic.The experiment on real data sets confirms that in the background of personal blogs clustering, heuristic blogs clustering is better both in precision and recall than the tradition blogs clustering. The cluster description we propose also has a better readability on the base of precision.