Extracting logical structures from HTML tables |
| |
Authors: | Yeon-Seok Kyong-Ho |
| |
Affiliation: | aDepartment of Computer Science, Yonsei University, 134, Shinchon-dong, Sudaemoon-ku, Seoul, 120-749, South Korea |
| |
Abstract: | While HTML is mainly designed for the visual rendering of Web documents, XML is widely accepted as a standard format to process and manage information. In particular, it can embed the information of logical structures. However, in order to utilize XML, the logical structures of HTML tables should first be extracted and transformed into XML representations. This paper presents an efficient method for the process, which consists of two phases: area segmentation and structure analysis. The area segmentation cleans up tables and segments them into attribute and value areas by checking visual and semantic coherency. The hierarchical structure between attribute and value areas is then analyzed and transformed into an XML representation using a proposed table model. Experimental results with 1180 HTML tables show that the proposed method performs better than conventional methods, resulting in an average accuracy of 86.7%. |
| |
Keywords: | HTML table Structure analysis Attribute-value relations Information extraction XML |
本文献已被 ScienceDirect 等数据库收录! |
|