Implementation of a weblog extraction system with an improved template extraction technique

Purpose: The objectives of this study are to explore an effective technique to extract information from weblogs and develop an experimental system to extract structured information as much as possible with this technique. The system will lay a foundation for evaluation, analysis, retrieval, and utilization of the extracted information.Design/methodology/approach: An improved template extraction technique was proposed.Separate templates designed for extracting blog entry titles, posts and their comments were established, and structured information was extracted online step by step. A dozen of data items, such as the entry titles, posts and their commenters and comments, the numbers of views, and the numbers of citations were extracted from eight major Chinese blog websites,including Sina, Sohu and Bokee.Findings: Results showed that the average accuracy of the experimental extraction system reached 94.6%. Because the online and multi-threading extraction technique was adopted, the speed of extraction was improved with the average speed of 15 pages per second without considering the network delay. In addition, entries posted by Ajax technology can be extracted successfully.Research limitations: As the templates need to be established in advance, this extraction technique can be effectively applied to a limited range of blog websites. In addition, the stability of the extraction templates was affected by the source code of the blog pages.Practical implications: This paper has studied and established a blog page extraction system,which can be used to extract structured data, preserve and update the data, and facilitate the collection, study and utilization of the blog resources, especially academic blog resources.Originality/value: This modified template extraction technique outperforms the Web page downloaders and the specialized blog page downloaders with structured and comprehensive data extraction.

2D Correlative-Chain Conditional Random Fields for Semantic Annotation of Web Objects

Semantic annotation of Web objects is a key problem for Web information extraction. The Web contains an abundance of useful semi-structured information about real world objects, and the empirical study shows that strong two-dimensional sequence characteristics and correlative characteristics exist for Web information about objects of the same type across different Web sites. Conditional Random Fields （CRFs） are the state-of-the-art approaches taking the sequence characteristics to do better labeling. However, as the appearance of correlative characteristics between Web object elements, previous CRFs have their limitations for semantic annotation of Web objects and cannot deal with the long distance dependencies between Web object elements efficiently. To better incorporate the long distance dependencies, on one hand, this paper describes long distance dependencies by correlative edges, which are built by making good use of structured information and the characteristics of records from external databases; and on the other hand, this paper presents a two-dimensional Correlative-Chain Conditional Random Fields （2DCC-CRFs） to do semantic annotation of Web objects. This approach extends a classic model, two-dimensional Conditional Random Fields （2DCRFs）, by adding correlative edges. Experimental results using a large number of real-world data collected from diverse domains show that the proposed approach can significantly improve the semantic annotation accuracy of Web objects.