WPBL: A Webpage Block Labeling Based Approach for Web Information Extraction

摘要

Currently conventional web mining (e.g., information retrieval, information extraction) usually treat the whole webpage as a basic unit to process. This leads to some problems such as low extraction precision, the loss of semantic information, etc. In this paper, we present Webpage Block Labeling (WPBL) based algorithm. In our system, a webpage is firstly segmented into finer granularity-several structurally coherent blocks, and then they are automatically labeled. To identify the importance of different blocks, an effective ranking algorithm based on a block's location and its visual features on the webpage is proposed, which contributes to find important contents or links on a webpage. In addition, to demonstrate the prospects of WPBL, we apply WPBL to web information extraction. The experimental results show that the WPBL algorithm can significantly improve the performance of the information extraction.