Abstract
The World Wide Web (WWW) is emerging as the Web of Data, providing information on various domains. A vast number of scientific documents such as books, articles and journals can be found through many publisher's websites, portals and XML exports. The challenge here is that the data about these scientific documents can not be explored collectively, as they are published as a bounded group of sources organized by different publishers. To address these limitations, we have developed a generic framework termed as Linked Open Publications Data Framework (LOPDF) that facilitates crawling, processing, extracting and producing machine-processable data that is open and linked to other open datasets. We also demonstrate the RDF datasets produced by using LOPDF framework and describe statistics of different datasets entities.