科技资源元数据的关联与推荐方法

Association and Recommendation Method for Metadata of Scientific and Technical Resources

大数据背景下,科技资源发现和推荐的关键是建立海量、多类型科技资源间的关联,并对其进行相关度排序。在深入研究科技基础性工作专项科技资源核心元数据的基础上,选择科技资源的内容特征、资源地点和资源时间为关联要素。然后结合专家打分和层次分析法,提出了科技资源元数据语义相关度算法,建立了科技资源间的关联。进一步按照相关度计算结果对科技资源进行排序,并将相关度高的科技资源优先推荐给用户。最后以科技基础性工作专项项目汇交的科技资源元数据为例,开展了科技资源元数据关联与推荐的实践。本研究提出的方法为促进海量科技资源的精准发现、智能推荐与共享应用提供了借鉴。

In the context of big data,efficient discovery and recommendation for scientific and technical data resources is to build the association between these data resources and then sort them by relevancy.Based on the investigation of core metadata of National Special Program on Basic Works for Science and Technology of China,this study chooses the content,location and temporal information of the data resources as association factors.Then,a semantic relevance algorithm is proposed based on the method of expert scoring and analytic hierarchy process,and the semantic association between these data resources is achieved in this study.These data resources are able to be sorted in terms of the semantic relevance,and the data resources with high relevance value can be recommended to the users.The proposed method is validated in the application case of data archiving and sharing for the projects of National Special Program on Basic Works for Science and Technology of China,and it has great significance in promoting the accurate discovery,intelligent recommendation and sharing for scientific data.