Research On The Unified Platform For Access And Full Text Retrieval In Open Access Journals

Open access, as a new, free electronic academic resource, initiated the sharing of information and academic. It is being noticed and supported by more and more scholars. Now, there are about ten thousand ones in the word, and it has covered almost all the subjects. However, with the growing number of OA resource and the data being in different server, it is more difficult for user to get the articles. According to the survey, there are several unified platform for access OA resource, such as DOAJ, OpenDOAR, Open J-Gate and Socolar. They all garnered some OA-journals and OA-repository but limitd. In the retrieval services, OA resource websites have their own retrieval tools, but rarely full text retrival. There is also only field-retrieval in the unified platform for accessing OA resource. So conforming of open access resource and providing improved services, in particular full text retrieval, is imperative.With the current situation of OA-Journal retrieval platform, this paper proposed constructing of OA-Journal database and building unified platform for retrieval, including the achieveing of full text retrieval. In the building of database, the data is mainly from DOAJ, and they are all high-quality and peer review. Now, most of OA-Journals obey to OAI protocol, so this paper harvestes OA articles by OAI-PMH. Besides, the OAI-PMH dosen't offer the field of full text, full-text is downloaded by special software designed by own. This unified platform for retrieval provides two types of services, including journal browsing and journal retrieval, with emphases on achieving of full text retrieval. Then, this paper proposes and concretely implements the technology method of full text retrieval of OA-Journal using search engine as proxy agent automatically.Finally, the test results show that the system is practical, steady-going and satisfactory. It conforms to user's habits. And it gets satisfactory results in efficiency and quality test of journal retrieval, including field-retrieval (title, author, keyword, abstract) and full text retrival. But it is very slow in local full text retrieval, and searching quality is acceptable.