[目的/意义]基于大量专利文献数据的核心技术主题识别有助于识别某技术领域的关键技术、分析关键技术的发展方向,是进行技术创新的基础情报工作,对于研究人员、企业乃至国家层面都具有一定的意义。[方法/过程]提出基于Chunk-LDAvis的核心技术主题识别方法,首先基于经典LDA模型进行主题识别,然后利用名词组块对初始LDA主题识别结果进行标注,构建Chunk-LDA主题识别结果,提高其可解读性;然后基于社会网络分析方法构建主题网络,识别核心技术主题;基于R语言的LDAvis工具包绘制可交互的Chunk-LDAvis核心技术主题关联分析图谱,发现核心技术主题的隐含联系,辅助进行核心技术主题识别。[结果/结论]通过对纳米农业领域进行实证研究,验证了本文提出方法的准确性和可行性。
[Purpose/significance] Core technology topic identification based on a large number of patent documents is helpful to detect key technologies in a technical field and to analyze the direction of the development of key technologies. It is the basic information work for technological innovation and has certain significance for researchers, enterprises and even the national level.[Method/process] This paper proposes a core technology topic identification method based on Chunk-LDAvis. Firstly, it is based on the classic LDA model to identify the topics. Then, the noun chunk is used to mark the results of the initial LDA topic identification, and the result of the Chunk-LDA topic recognition is constructed to improve its interpretability. Then based on the social network analysis method, the topic network is constructed to identify the core technical topics; based on the LDAvis toolkit, the interactive Chunk-LDAvis core technology topic association analysis map is plotted, and the hidden links of the core technical topics are found, and the core technology topic detection is assisted.[Result/conclusion] Through the empirical study on the field of nanoscale agriculture, the accuracy and feasibility of the proposed method are verified.
[1] 王效岳,白如江.海量网络学术文献自动分类技术研究[M].北京:人民出版社,2015:40-42.
[2] SCHANKERMAN M, PAKES A. Estimates of the value of patent rights in European countries during the post-1950 period[J]. Economic journal, 1986, 96(384):1052-1076.
[3] 许海云,岳増慧,雷炳旭,等.基于专利技术功效主题词与专利引文共现的核心专利挖掘[J].图书情报工作,2014,58(4):59-64.
[4] 袁润,钱过.识别核心专利的粗糙集理论模型[J].图书情报工作, 2015,59(2):123-130.
[5] 马永涛,张旭,傅俊英,等.核心专利及其识别方法综述[J].情报杂志,2014,33(5):38-43,70.
[6] KWON O, SEO J, NOH K, et al. Categorizing influential patents using bibliometric analysis of patent citations network[J]. Information-an international interdisciplinary journal, 2007, 10(3):313-326.
[7] CHOI C, PARK Y. Monitoring the organic structure of technology based on the patent development paths[J]. Technological forecasting and social change, 2009, 76(6):754-768.
[8] HSU C W, CHANG P L, HSIUNG CM, et.al. Charting the evolution of biohydrogen production technology through a patent analysis[J]. Biomass & bioenergy, 2015,76(5):1-10.
[9] 张欣,马瑞敏.基于改进PageRank算法的核心专利发现研究[J].图书情报工作,2018,62(10):106-115.
[10] 亢川博,王伟,穆晓敏,等.核心专利识别的综合价值模型[J].情报科学,2018,36(2):67-70.
[11] WANG Y, BAI H J, STANTON M, et al. PLDA:parallel latent Dirichlet allocation for large-scale applications[C]//International conference on algorithmic aspects in information and management. San Francisco:Springer-verlag, 2009:301-314.
[12] NEWMAN M E J,GIRVAN M. Finding and evaluating community structure in networks[J].Physical review, 2004,69(2):108-113.
[13] BLONDEL V D, GUILLAUME J L, LAMBIOTTE R, et al. Fast unfolding of communities in large networks[J]. Journal of statistical mechanics:theory and experiment,2008,30(2):155-168
[14] LEE Y G, SONG Y I. Selecting the key research areas in nano-technology field using technology cluster analysis:a case study based on National R&D Programs in South Korea[J].Technovation, 2007, 27(12):57-64.
[15] 栾春娟,曾国屏.基于SNA核心技术领域测度研究[J].图书情报工作,2011,55(6):33-35.
[16] 范宇,符红光,文奕.基于LDA模型的专利信息聚类技术[J].计算机应用,2013,33(S1):87-89,93.
[17] 李佳佳,马铁驹.基于专利数据的风能核心技术识别及趋势分析[J].科技管理研究,2017(12):129-136.
[18] 伊惠芳,吴红,马永新,等. 基于LDA和战略坐标的专利技术主题分析——以石墨烯领域为例[J].情报杂志,2018,37(5):97-102.
[19] BLEI D M, NG A Y, JORDAN M I. Latent Dirichlet allocation[J].Journal of machine learning research,2003(3):993-1022.
[20] BLEI D M, LAFFERTY J. Dynamic topic models[C]//Proceedings of the 23rd international conference on machine learning. NewYork:ACM,2006:113-120.
[21] WANG X, MCCALLUM A, WEI X. Topical N-Grams:phrase and topic discovery, with an application to information retrieval[C]//IEEE international conference on data mining. Omaha:IEEE Computer Society, 2007:697-702.
[22] ELKISHKY A, SONG Y, VOSS C R, et al. Scalable topical phrase mining from text corpora[J]. Proceedings of the VLDB endowment, 2014, 8(3):305-316.
[23] LI B, WANG B, ZHOU R, et al. CITPM:A cluster-based iterative topical phrase mining framework[C]//International conference on database systems for advanced applications. Dallas:Springer International Publishing, 2016:197-213.
[24] SIEVERT C, SHIRLEY K. LDAvis:a method for visualizing and interpreting topics[C]//Proceedings of the workshop on interactive language learning, visualization, and interfaces. Baltimore:Association for Computational Linguistics, 2014:63-70.
[25] 范云满,马建霞.基于LDA与新兴主题特征分析的新兴主题探测研究[J].情报学报,2014,33(7):698-711.
[26] 张琴,张智雄.基于PhraseLDA模型的主题短语挖掘方法研究[J]. 图书情报工作, 2017, 61(8):120-125.
[27] LANDAUER T K, DUMAIS S T. A solution to Plato's problem:the latent semantic analysis theory of acquisition, induction, and representation of knowledge[J]. Psychological review, 1997, 104(2):211-240.
[28] SHEN C, LI T, DING C H Q. Integrating clustering and multi-document summarization by bi-mixture probabilistic latent semantic analysis (PLSA) with sentence bases[C]//AAAI conference on artificial intelligence. San Francisco:AAAI Press, 2011:914-920.
[29] BONACICH P B. Factoring and weighting approaches to status scores and clique identification[J].Journal of mathematical sociology,1972, 2(1):113-120.
[30] STURROCK K, ROCHA J. A multidimensional scaling stress evaluation table[J]. Field methods, 2016, 12(1):49-60.
