A Discriminant Framework for Detecting Similar Scientific Research Projects Based on Big Data Mining

摘要: Scientific research projects play an important role in promoting the science and technology competitiveness of a country. Due to lack information open sharing, it is possible approve similar or duplicated by different government departments. In some way, these are waste scientific resources. To avoid such problem, this paper proposes discriminant framework for detecting based on big data mining technologies, providing evidence-based decision making departments during project approval process. Firstly, we construct file associated with officially approved projects, including titles, principal investigators, organizations, keywords, bibliographies published scholar papers. Secondly, proposed detect from above file. Finally, adopt Hadoop architecture speed up algorithm. demonstrate effectiveness feasibility framework, implement prototype system detection.