PROJECT TITLE :

Unsupervised Web Topic Detection Using A Ranked Clustering-Like Pattern Across Similarity Cascades

ABSTRACT:

Despite the huge growth of social media on the Web, the process of organizing, understanding, and monitoring user generated content (UGC) has become one among the foremost pressing problems in today’s society. Discovering topics on the web from a huge volume of UGC is one in every of the promising approaches to realize this goal. Compared with classical topic detection and tracking in news articles, identifying topics on the net is by no means easy because of the noisy, sparse, and fewer- constrained knowledge on the Net. During this paper, we have a tendency to investigate ways from the attitude of similarity diffusion, and propose a clustering-like pattern across similarity cascades (SCs). SCs are a series of subgraphs generated by truncating a similarity graph with a collection of thresholds, and then maximal cliques are used to capture topics. Finally, a subject-restricted similarity diffusion method is proposed to efficiently identify real topics from a large variety of candidates. Experiments demonstrate that our approach outperforms the state-of-the-art strategies on 3 public knowledge sets.