Research Repository Design And Recommendation System

With the rapid development of Internet technology, information on the very fast growth of the Internet. Internet users can get all sorts of information from the network, using the Internet to learn new knowledge is now becoming more common. Build a comprehensive resource library is essential, first of all users can share their research results through the repository; Second, users can browse or download other people’s research repository. In this paper, the system also provides a repository personalized recommendation feature, can be provided for a personalized list of recommended users based on user behavior data and their own information, to facilitate the users to find resources.Some of the more mature domestic electronic resource database Articles, VIP, etc. These are some very professional electronic repository, providing a large number of professional research papers. There are some private companies or organizations to build some platforms. Academic as Baidu, Google and other academic and some professional forums, which relative to the Articles and VIP their late start, but there are also some of their own unique advantages. Such as Baidu or Google, their search engine technology is very mature, to help users quickly find the corresponding resources. Although a wide variety of electronic resources platform more and more, but they have their own drawbacks. For example, now appears on the network of various professional forums, electronic resources are some of the relevant professional who published a number of issues on their own knowledge or understanding and insight, understanding and insight that these experts have not been verification, so there may be some errors. Some of the more authoritative electronic repositories such as Articles and VIP electronic repository, each of these professional network platform provides electronic resources are reviewed by experts correctness relatively high. But these are some of the main forms of electronic resources in the form of papers appeared, so most are merely illustrative of the new results of the study made no detailed theory of method steps to achieve them.This paper studies the design of the repository and recommendation system is to achieve a repository system and recommendation algorithm will be applied to the design of the repository system. Repository contents to the project as an organizational unit, the main contents of each project, including documents, pictures, video and code. These electronic resources include not only the new achievements made in the project, but also a detailed description of the items used by the knowledge, implementation, implementation code, and some of the results show the video. Electronic Resources project repository is organized as a unit, each project would include the completion of all knowledge and methods of the project, so anyone will be able to obtain these materials themselves to achieve this project. In this paper, personalized recommendation technology into the repository, the repository can recommend resources to users based on their user may be interested in the characteristics of the user’s information resources and score data downloaded. Compared with the traditional recommendation algorithm recommendation algorithm repository overcomes the problems of cold start and sparse matrix, it is possible to achieve a more accurate recommendation result.