[目的/意义] 改善虚拟健康社区知识聚合质量，为虚拟健康社区服务提供技术方法支持。[方法/过程] 运用谱聚类方法对虚拟健康社区中的知识进行抽取，利用概念相似度计算得到知识主题相似度矩阵，根据该相似度矩阵进行谱聚类。[结果/结论] 利用好大夫在线健康咨询平台发布的信息作为数据来源进行方法验证。结果表明，当聚类个数为5时，本文提出的方法得分值最高。通过谱聚类的方法充分挖掘虚拟健康社区潜在信息，改善了知识聚合质量，为知识聚合和知识服务提供了一条新途径。
[Purpose/significance] To improve the quality of knowledge aggregation in healthy virtual communities and provide technical method support for virtual healthy community services. [Method/process] The method of spectral clustering was applied to knowledge in the virtual healthy community was extracted, and the semantic similarity matrix of the text was obtained by using the keyword co-occurrence. The spectral clustering was performed according to the text semantic similarity matrix, and the text was aggregated into text clusters. [Result/conclusion] The information published by the doctor’s online health consultation platform was used as a data source for method validation. The results show that when the number of clusters is 5, the proposed method has the highest score. This method of spectral clustering considers the semantic relationship between words, fully exploits the potential information of virtual healthy community, improves the quality of knowledge aggregation, and provides a new way for knowledge aggregation and knowledge service.
[1] 新华社.我国社会主要矛盾转化的背后[EB/OL].[2019-06-26]. http://cpc.people.com.cn/19th/n1/2017/1021/c414305-29600806.html.
[2] HAJLI M N. Developing online health communities through digital media[J]. International journal of information management,2014, 34(2):311-314.
[3] 毕强.数字资源:从整合到聚合的转变[J].数字图书馆论坛,2014(6):1.
[4] 李亚婷.知识聚合研究述评[J].图书情报工作,2016,60(21):128-136.
[5] 王敬东.基于知识聚合的数字图书馆信息智能检索模型[J].图书馆学研究,2014(21):72-76,71.
[6] 贯君,毕强,赵夷平.基于关联数据的知识聚合与发现研究进展[J].情报资料工作,2015(3):15-21.
[7] 李洁.基于SNA的馆藏数字资源知识聚合可视化研究[D].长春:吉林大学,2016.
[8] 陈果,朱茜凌,肖璐.面向网络社区的知识聚合:发展、研究基础与展望[J].情报杂志,2017,36(12):193-197,192.
[9] 张连峰,李慧,遆云鹤. 基于虚拟学术社区的知识聚合模型构建研究[J].情报科学,2019,37(6):55-60,74.
[10] 胡媛,刁首琪,朱益平,等. 基于知识聚合的数字图书馆社区服务推送系统设计与实现[J].情报科学,2017,35(11):72-77.
[11] 商宪丽,王学东,张煜轩. 基于标签共现的学术博客知识资源聚合研究[J].情报科学,2016,34(5):125-129.
[12] LIANG K, WANG C,ZHAN Y Y.Knowledge aggregation and intelligent guidance for fragmented learning[J]. Procedia computer science, 2018,131(4):656-664.
[13] TARKO V,DRAGOSALIGICA P.From "Broad Studies" to Internet-based "Expert Knowledge Aggregation".Notes on the methodology and technology of knowledge integration[J]. Futures, 2011,43(9):986-995.
[14] RITOU M, BELKADI F, YAHOUNI Z,et al. Knowledge-based multi-level aggregation for decision aid in the machining industry[J]. CIRP annals, 2019,68(1):475-478.
[15] OOSTERMANA J, YANG J,ALESSANDRO B, et al.On the impact of knowledge extraction and aggregation on crowdsourced annotation of visual artworks[J]. Computer networks,2015,90(29):133-149.
[16] JANANI J,VIJAYARANI S. Text document clustering using spectral clustering algorithm with particle swarm optimization[J]. Expert systems with applications, 2019,134(15):192-200.
[17] LI X, WANG Z J, HU R L,et al. Recommendation algorithm based on improved spectral clustering and transfer learning[J]. Pattern analysis and applications,2019,22(2):633-647.
[18] 李航. 统计学习方法[M].2版.北京:清华大学出版社,2019.
[19] 王春柳,杨永辉,邓霏,等.文本相似度计算方法研究综述[J].情报科学,2019,37(3):158-167.
[20] 李枫林,柯佳. 基于深度学习的文本表示方法[J].情报科学,2019,37(1):156-164.
[21] LIU K H, HOGAN W R,REBECCA S. Crowley.Natural language processing methods and systems for biomedical ontology learning[J].Journal of biomedical informatics 2011, 44(1):163-179.
[22] ANDRÉSPAREDES-VALVERDE M, ÁNGELRODRÍGUEZ-GARCÍA M, RUIZ-MARTÍNEZ A, et al.ONLI:an ontology-based system for querying DBpedia using natural language paradigm[J].Expert systems with applications, 2015, 42(12):5163-5176.
[23] FRANTZI K T,ANANINADOU S. The C-Value/NC-Value domain independent method for multi-word term extraction[J]. Journal of natural language processing, 2008,6(3):145-179.
[24] 廖福燕.本体构建中概念和关系获取方法研究[D].西安:西安建筑科技大学,2011.
