改进的聚类分析算法在科研立项管理中的应用研究

The Research and Application of the Scientific Research Project Management Based on the Improve of Clustering Algorithm

针对目前的科研项目管理信息系统仅对科研项目进行低水平管理,无法区分、甄别科研内容等问题,对k-means聚类分析技术进行改进,并进一步将该项技术应用于科研立项管理中,通过对科研立项申报书进行聚类分析,得出立项申请中相似的项目和创新的项目,为科研立项提供智能型的决策支持,避免了重复立项和重复研究,使得计算机应用技术更好地服务于科研项目管理。

For current low-level management of scientific research project information and incapability to distinguish or identify the contents of research projects,the paper improves the k-means clustering analysis technique,and further applies this technique into research project initialization management.Through clustering analysis of research project initialization declaration documents,decision-makers can find out the repetitive studies and the innovative projects.It intelligently supports the decision-making in project initialization by avoiding repetitive projects and studies,and makes it possible for computer application technology to better serve scientific research project management.