This paper proposes a new method for calculating similarity between secondary documents of National Science and Technology Library（NSTL）, subsequently develops and implements a subsystem of real-time related documents recommendation. At last, a measurement which can theoretically evaluate the quality of recommendation results is presented. And the experiment results demonstrate that the system can improve the service quality of document retrieval system of NSTL.
张志平 李琳娜. NSTL文献检索系统中相关文献推荐功能的设计及实现[J]. 现代图书情报技术, 2010, 26(7/8): 110-113.
Zhang Zhiping Li Linna. Design and Implementation of Related Document Recommendation in Document Retrieval System of NSTL. New Technology of Library and Information Service, 2010, 26(7/8): 110-113.
［1］ Zhou D, Zhu S, Yu K, et al. Learning Multiple Graphs for Document Recommendations［C］. In: Proceedings of the 17th International Conference on World Wide Web. New York, NY，USA：ACM, 2008: 141-150.
［2］ Chandrasekaran K. Concept Based Author Recommender System for CiteSeer［D］. Kansas，U.S.：University of Kansas, 2007.
［3］Huang Y C. Combining Social Networks and Content for Recommendation in a Literature Digital Library［D］. Taiwan：National Sun Yat-Sen University，2008.
［4］ Smeaton A F, Callan J. Personalisation and Recommender Systems in Digital Libraries［J］. International Journal on Digital Libraries，2005, 5(4): 299-308.
［5］ 陈祖琴, 张惠玲, 葛继科，等. 基于加权关联规则挖掘的相关文献推荐［J］. 现代图书情报技术，2007(10): 57-61.
［6］ 张奇, 黄萱菁, 吴立德. 一种新的句子相似度度量及其在文本自动摘要中的应用［J］. 中文信息学报，2005, 19(2): 93-99.
［7］ 周法国, 杨炳儒. 句子相似度计算新方法及在问答系统中的应用［J］. 计算机工程与应用，2008, 44(1): 165-167，178.
［8］ Thorsten J. A Probabilistic Analysis of the Rocchio Algorithm with TFIDF for Text Categorization［C］. In: Proceedings of the 14th International Conference on Machine Learning. San Francisco, CA, USA: Morgan Kaufmann Publishers Inc.，1997: 143-151.
［9］ Shani G, Gunawardana A. Evaluating Recommendation Systems［EB/OL］.(2009-11-03).［2010-05-07］.http://research.microsoft.com/pubs/115396/EvaluationMetrics.TR.pdf.
［10］ Jrvelin K, Keklinen J. Cumulated Gain-based Evaluation of IR Techniques［J］. ACM Transactions on Information Systems, 2002, 20(4): 422-446.
