Internet information source discovery based on multi-seeds cocitation

Abstract

The technology of Internet information source discovery on specific topic is the groundwork of information acquisition in current big data era. This paper presents a multi-seeds cocitation algorithm to find new Internet information sources. The proposed algorithm is based on cocitation, but what difference with the traditional algorithms is that we use multiple websites on specific topic as input seeds. Then we induce Combined Cocitation Degree(CCD) to measure the relevancy of newly found websites, which is that the new websites have higher combined cocitation degree and are more topic related. Finally a websites collection of the biggest CCD is referred to as the new Internet information sources on the specific topic. The experiments show that the proposed method outperforms traditional algorithms in the scenarios we tested.