A Method for Information Extraction from the Web

Many data sources are available today on the Web like product catalogs, conference, and multiple directories. The extraction of information from the content is a hard task since they are heterogeneous and dynamic. This paper presents a new method for extracting wrappers and relations from the Web that combines the discovery of similarities in the structures of data that a user wishes to extract from a given Web page, and the generalization of contexts of the extracted data. This method is now implemented in MDSManager our system for data sources Fusion