科技项目及其成果文献的相关性评估研究  

Analysis of the Relevance Evaluation of Scientific-Technological Projects and Achievements

作　　者：梁继文 杨建林[1,2] 王伟 王飞[3] Liang Jiwen;Yang Jianlin;Wang Wei;Wang Fei(School of Information Management,Nanjing University,Nanjing 210023;Jiangsu Key Laboratory of Data Engineering and Knowledge Service,Nanjing 210023;Jiangsu Institute of Science and Technology Information,Nanjing 210042)

机构地区：[1]南京大学信息管理学院,南京210023 [2]江苏省数据工程与知识服务重点实验室,南京210023 [3]江苏省科学技术情报研究所,南京210042

出　　处：《情报学报》2022年第2期155-166,共12页Journal of the China Society for Scientific and Technical Information

摘　　要：科技项目后评估对项目的完成质量与效益进行评判,是科技管理全流程中的核心环节。现有科技项目后评估研究更多的是关注如何构建评估体系和量化评价指标,鲜有基于内容来探讨项目与成果的相关性。本文使用科技报告作为科技项目的补充信息,构建基于BERT(bidirectional encoder representations from transformers)架构的相似度计算模型,通过文本语义匹配程度对科技项目-成果文献的相关性评估进行探索。研究结果表明,构建的融合模型可以有效地评估科技项目-成果文献的相关性。此外,本文对项目-成果相关性较低的情况以及科技报告监管过程中存在的问题进行了分析,旨在建设优质科技资源,完善科技项目后评估体系,综合提升科技情报服务效率。The post evaluation of scientific-technological(S&T)projects is the core link in the entire process of S&T man-agement,which judges the completion quality and benefits of the project.Existing research more deeply considers building an evaluation system and quantitative evaluation indicators,while the research on determining the relevance of projects and results at the content level is scarce.This paper uses S&T reports as supplementary information for S&T projects,building a similarity calculation model based on the Bidirectional Encoder Representations from Transformers(BERT)ar-chitecture,and it explores S&T projects’correlation evaluation and achievement.The results show that the fusion model can effectively evaluate the relevance of S&T projects and achievement.This study also analyzes the low relevance of the projects and achievements and the existing problems in the S&T reports.We aimed to comprehensively improve the effi-ciency of S&T information services.

关 键 词：科技项目 科技报告 项目后评估 成果评估 成果相关性 

分 类 号：G350[文化科学—情报学] TP311.13[自动化与计算机技术—计算机软件与理论]