Processing of unstructured data for information extraction

Unstructured data are those that have no predetermined form or structure and are full of textual data. It does not fit well into relational tables. Most enterprise data today can actually be considered unstructured. Typical unstructured systems include emails, reports, contracts, transcripts of telephone conversations, and other communications. Web pages also contain links and references to External, often unstructured content such as images, XML files, animations and databases. This paper focuses on extracting features in html pages by using tokenization and Non matrix factorization. Classification of text is done using bag of words approach. The workbench is dataset collected in university domain web pages.