Mining Publication Records on Publication Pages based on Conditional Random Fields

• Main Authors: LIN, YA-HUEI, 林雅惠
• Other Authors: Hahn-Ming Lee
• Format: Others
• Language: en_US
• Published: 2012
• Online Access: http://ndltd.ncl.edu.tw/handle/bh8m2e

碩士 === 國立臺灣科技大學 === 資訊工程系 === 100 === A publication record is a list of semi-structured citation strings for publications of a research institute or an individual researcher. Publication records are integrated into a digital library which becomes an important knowledge base and thereby enables a variety of applications. A publication record is usually found among other information on a publication Web page (or ”publication page” for short). It is thus an interesting problem to extract publication record from such Web pages. The problem is difficult for several reasons, e.g., flexibility in formatting the metadata of a publication as a semi-structured citation string and flexibility in expressing the citation string visually presentation in HTML. Furthermore, two citation strings with a similar visual presentation on the same Web page may have different HTML constructs. In this paper, we present a content analysis approach, based on Conditional Random Fields and data region boundary analysis, the problem of automatically extracting publication records on a publication page. Experimental results show that our method performs well on a benchmark containing manually crafted publication pages. The precision rate and recall rate, and F-measure are 82.5%, 87.6%, and 85.0%, respectively. This is a significant improvement over previous researches.