Please wait a minute...
Data Analysis and Knowledge Discovery  2020, Vol. 4 Issue (12): 55-67    DOI: 10.11925/infotech.2096-3467.2020.0175
Current Issue | Archive | Adv Search |
Recommending Knowledge for Online Health Community Users Based on Fuzzy Cognitive Map
Li He1(),Liu Jiayu1,Shen Wang1,Liu Rui2,Jin Shuaiqi1
1School of Management, Jilin University, Changchun 130022, China
2China-Japan Union Hospital of Jilin University, Changchun 130033, China
Download: PDF (1089 KB)   HTML ( 10
Export: BibTeX | EndNote (RIS)      
Abstract  

[Objective] This paper constructs a fuzzy cognitive map model, aiming to recommend context-driven knowledge for users of online health communities. [Methods] First, we extracted keywords from user comments and used them as concept nodes of the proposed model. Then, we calculated the absolute values of the weight relationship between concept nodes based on the similarity of keyword co-occurrence. Third, we determined the semantic relationship among the keywords through literature reviews and expert collaborations. Finally, we built the fuzzy cognitive map and recommended disease related knowledge using the change of state values among nodes. [Results] Our new model’s precision, recall and F-measure were 0.286, 0.667 and 0.400 respectively. [Limitations] The amount of user comments need to be increased, which will improve the model's performance. [Conclusions] The proposed model optimizes the recommendation mechanism of online health communities and provides better knowledge for patients.

Key wordsFuzzy Cognitive Map      Knowledge Recommendation      Online Health Community      Online Comment     
Received: 09 March 2020      Published: 25 December 2020
ZTFLH:  G354  
Corresponding Authors: Li He     E-mail: lihe200303@163.com

Cite this article:

Li He,Liu Jiayu,Shen Wang,Liu Rui,Jin Shuaiqi. Recommending Knowledge for Online Health Community Users Based on Fuzzy Cognitive Map. Data Analysis and Knowledge Discovery, 2020, 4(12): 55-67.

URL:

https://manu44.magtech.com.cn/Jwk_infotech_wk3/EN/10.11925/infotech.2096-3467.2020.0175     OR     https://manu44.magtech.com.cn/Jwk_infotech_wk3/EN/Y2020/V4/I12/55

比较项 在线健康社区知识推荐 一般信息推荐
信息源 平台专业信息集中 信息异质、异构、分散
推荐对象 用户属性强 用户属性相对弱
推荐结果 查准率要求高 查全率要求高
推荐机制 追求场景化、个性化 推荐领域广、技术易实现
Differences Between Online Health Community Knowledge Recommendation and General Information Recommendation
Framework of FCM Construction and Recommendation for Online Health Community
Representation of Semantic Relationship Between Diseases and Drugs
用户编号(ui) 用户评论中提取与糖尿病相关医学关键词术语(Ci
u1 脓毒(血)症C1、血常规检查C2、感染C3、肝脓肿C5、2型糖尿病C14
u2 甲状腺C4
u3
u4 脓毒(血)症C1、肝脓肿C5
u5 脓毒(血)症C1
u6 感染C3、肝脓肿C5、交感神经C6、病原体C7
u7 胰岛素C8、糖尿病并发症C9、内分泌C10、免疫C11、抗生素C12
u8
u9 肝脓肿C5
u10 2型糖尿病C14
u11 脓毒(血)症C1、肝脓肿C5、1型糖尿病C13、2型糖尿病C14
u12 抗生素C12
u13 脓毒(血)症C1、感染C3
u14 脓毒(血)症C1、感染C3
u15 肝脓肿C5
u16 肝脓肿C5
u17 肝脓肿C5
u18 感染C3、肝脓肿C5
u19
u20
Example of Keyword Extraction
节点 概念 关系 节点 概念 关联强度 节点 概念 关系 节点 概念 关联强度
C1 脓毒(血)症 需要 C2 血常规检查 0.790 C10 内分泌异常 导致 C4 甲状腺功能异常 0.434
C1 脓毒(血)症 说明 C3 感染 0.568 C10 内分泌异常 引发 C9 糖尿病并发症 0.874
C1 脓毒(血)症 改变 C4 甲状腺功能异常 0.490 C10 内分泌异常 正相关 C11 免疫下降 0.775
C1 脓毒(血)症 原因 C5 肝脓肿 0.612 C10 内分泌异常 引发 C13 1型糖尿病 0.807
C1 脓毒(血)症 原因 C7 病原体存在 0.511 C10 内分泌异常 引发 C14 2型糖尿病 0.807
C1 脓毒(血)症 引起 C10 内分泌异常 0.472 C11 免疫下降 正相关 C1 脓毒(血)症 0.471
C1 脓毒(血)症 正相关 C11 免疫下降 0.471 C11 免疫下降 正相关 C3 感染 0.397
C1 脓毒(血)症 药物 C12 需要抗生素 -0.637 C11 免疫下降 正相关 C4 甲状腺功能异常 0.428
C1 脓毒(血)症 正相关 C14 2型糖尿病 0.386 C11 免疫下降 正相关 C5 肝脓肿 0.589
C3 感染 引发 C1 脓毒(血)症 0.568 C11 免疫下降 正相关 C6 交感神经病变 0.439
C3 感染 需要 C2 血常规检查 0.533 C11 免疫下降 正相关 C7 病原体存在 0.439
C3 感染 改变 C4 甲状腺功能异常 0.021 C11 免疫下降 正相关 C9 糖尿病并发症 0.724
C3 感染 引发 C5 肝脓肿 0.655 C11 免疫下降 正相关 C10 内分泌异常 0.775
C3 感染 引发 C9 糖尿病并发症 0.435 C11 免疫下降 正相关 C13 1型糖尿病 0.695
C3 感染 引发 C10 内分泌异常 0.373 C11 免疫下降 正相关 C14 2型糖尿病 0.733
C3 感染 正相关 C11 免疫下降 0.397 C12 需要抗生素 治疗 C1 脓毒(血)症 -0.637
C3 感染 药物 C12 需要抗生素 -0.602 C12 需要抗生素 治疗 C3 感染 -0.602
C4 甲状腺功能异常 正相关 C11 免疫下降 0.428 C12 需要抗生素 治疗 C5 肝脓肿 -0.651
C5 肝脓肿 引发 C1 脓毒(血)症 0.612 C12 需要抗生素 治疗 C7 病原体存在 -0.524
C5 肝脓肿 需要 C2 血常规检查 0.597 C12 需要抗生素 治疗 C9 糖尿病并发症 -0.717
C5 肝脓肿 原因 C3 感染 0.655 C13 1型糖尿病 引发 C3 感染 0.285
C5 肝脓肿 原因 C7 病原体存在 0.386 C13 1型糖尿病 引发 C4 甲状腺功能异常 0.392
C5 肝脓肿 正相关 C11 免疫下降 0.589 C13 1型糖尿病 药物 C8 需要胰岛素 -0.708
C5 肝脓肿 药物 C12 需要抗生素 -0.651 C13 1型糖尿病 引发 C9 糖尿病并发症 0.806
C6 交感神经病变 正相关 C11 免疫下降 0.439 C13 1型糖尿病 原因 C10 内分泌异常 0.807
C7 病原体存在 引发 C1 脓毒(血)症 0.511 C13 1型糖尿病 引起 C11 免疫下降 0.695
C7 病原体存在 需要 C2 血常规检查 0.658 C14 2型糖尿病 引发 C1 脓毒(血)症 0.386
C7 病原体存在 引发 C3 感染 0.385 C14 2型糖尿病 引发 C3 感染 0.336
C7 病原体存在 引发 C4 甲状腺功能异常 0.658 C14 2型糖尿病 改变 C4 甲状腺功能异常 0.412
C7 病原体存在 引发 C5 肝脓肿 0.386 C14 2型糖尿病 并发 C5 肝脓肿 0.421
C7 病原体存在 引发 C11 免疫下降 0.439 C14 2型糖尿病 引发 C6 交感神经病变 0.467
C7 病原体存在 药物 C12 需要抗生素 -0.524 C14 2型糖尿病 药物 C8 需要胰岛素 -0.753
C8 需要胰岛素 影响 C9 糖尿病并发症 -0.853 C14 2型糖尿病 引发 C9 糖尿病并发症 0.809
C8 需要胰岛素 治疗 C13 1型糖尿病 -0.708 C14 2型糖尿病 原因 C10 内分泌异常 0.807
C8 需要胰岛素 治疗 C14 2型糖尿病 -0.753 C14 2型糖尿病 引发 C11 免疫下降 0.733
C9 糖尿病并发症 引发 C4 甲状腺功能异常 0.411
C9 糖尿病并发症 药物 C8 需要胰岛素 -0.853
C9 糖尿病并发症 药物 C12 需要抗生素 -0.717
Diabetic Topic User Comment Keywords Semantic Relation Dictionary
C1
脓毒(血)症
C2
血常规检查
C3
感染
C4
甲状腺功能
异常
C5
肝脓肿
C6
交感神经病变
C7
病原体
存在
C8
需要
胰岛素
C9
糖尿病并发症
C10
内分泌
异常
C11
免疫
下降
C12
需要
抗生素
C13
1型
糖尿病
C14
2型
糖尿病
脓毒(血)症C1 1 0.790 0.568 0.490 0.612 0 0.511 0 0 0.472 0.471 -0.637 0 0.386
血常规检查C2 0 1 0 0 0 0 0 0 0 0 0 0 0 0
感染C3 0.568 0.533 1 0.021 0.655 0 0 0 0.435 0.373 0.397 -0.602 0 0
甲状腺功能异常C4 0 0 0 1 0 0 0 0 0 0 0.428 0 0 0
肝脓肿C5 0.612 0.597 0.655 0 1 0 0.386 0 0 0 0.589 -0.651 0 0
交感神经病变C6 0 0 0 0 0 1 0 0 0 0 0.439 0 0 0
病原体存在C7 0.511 0.658 0.385 0.658 0.386 0 1 0 0 0 0.439 -0.717 0 0
需要胰岛素C8 0 0 0 0 0 0 0 1 -0.853 0 0 0 -0.708 -0.753
糖尿病并发症C9 0 0 0 0.411 0 0 0 -0.853 1 0 0 -0.717 0 0
内分泌异常C10 0 0 0 0.434 0 0 0 0 0.874 1 0.775 0 0.807 0.807
免疫下降C11 0.471 0 0.397 0.428 0.589 0.439 0.439 0 0.724 0.775 1 0 0.695 0.733
需要抗生素C12 -0.637 0 -0.602 0 -0.651 0 -0.524 0 -0.717 0 0 1 0 0
1型糖尿病C13 0 0 0.285 0.392 0 0 0 -0.708 0.806 0.807 0.695 0 1 0
2型糖尿病C14 0.386 0 0.336 0.412 0.421 0.467 0 -0.753 0.809 0.807 0.733 0 0 1
Diabetes FCM Adjacency Matrix W
C1
脓毒(血)症
C2
血常规检查
C3
感染
C4
甲状腺
功能异常
C5
肝脓肿
C6
交感神经病变
C7
病原体存在
C8
需要胰岛素
C9
糖尿病并发症
C10
内分泌异常
C11
免疫
下降
C12
需要
抗生素
C13
1型
糖尿病
C14
2型
糖尿病
A(0) 0 0 0 0 0 0 0 0 0 0 0 0 1 0
A(1) 0.500 0 0.500 0 0.701 6 0.764 2 0.500 0 0.500 0 0.500 0 0.106 8 0.918 2 0.918 4 0.889 4 0.500 0 0.952 6 0.500 0
A(2) 0.994 8 0.996 6 0.997 5 0.999 9 0.996 3 0.966 8 0.962 0 0.005 6 1.000 0 0.999 9 1.000 0 0.008 6 0.998 8 0.997 6
A(3) 1.000 0 1.000 0 1.000 0 1.000 0 1.000 0 0.996 4 0.999 0 0.001 0 1.000 0 1.000 0 1.000 0 0.000 0 0.999 4 0.999 8
A(4) 1.000 0 1.000 0 1.000 0 1.000 0 1.000 0 0.996 7 0.999 1 0.000 9 1.000 0 1.000 0 1.000 0 0.000 0 0.999 4 0.999 8
w(数值变化幅度) 0.500 0 0.500 0 0.298 4 0.235 8 0.500 0 0.496 4 0.499 0 0.105 8 0.081 8 0.081 5 0.110 6 0.500 0 0.046 8 0.499 8
The Status of Each Node at Each Time When Suffering from Type 1 Diabetes
C1
脓毒(血)症
C2
血常规检查
C3
感染
C4
甲状腺
功能异常
C5
肝脓肿
C6
交感神经病变
C7
病原体存在
C8
需要
胰岛素
C9
糖尿病并发症
C10
内分泌异常
C11
免疫
下降
C12
需要
抗生素
C13
1型
糖尿病
C14
2型
糖尿病
A(0) 0 0 0 0 0 0 0 0 0 0 0 0 0 1
A(1) 0.761 0 0.500 0 0.732 6 0.774 9 0.779 5 0.802 3 0.500 0 0.094 6 0.918 9 0.918 4 0.900 2 0.500 0 0.500 0 0.952 6
A(2) 0.999 2 0.999 0 0.999 2 0.999 9 0.999 5 0.992 8 0.981 5 0.005 9 1.000 0 1.000 0 1.000 0 0.002 9 0.995 5 0.999 6
A(3) 1.000 0 1.000 0 1.000 0 1.000 0 1.000 0 0.996 7 0.999 0 0.001 0 1.000 0 1.000 0 1.000 0 0.000 0 0.999 4 0.999 8
A(4) 1.000 0 1.000 0 1.000 0 1.000 0 1.000 0 0.996 7 0.999 1 0.001 0 1.000 0 1.000 0 1.000 0 0.000 0 0.999 4 0.999 8
w(数值变化幅度) 0.239 0 0.500 0 0.267 4 0.225 1 0.220 5 0.194 4 0.499 0 0.093 6 0.081 1 0.081 6 0.099 8 0.500 0 0.499 4 0.047 2
The Status of Each Node at Each Time When Suffering from Type 2 Diabetes
The Performance of Three Recommendation Algorithms
[1] 李月琳, 张秀, 王姗姗 . 社交媒体健康信息质量研究:基于真伪健康信息特征的分析[J]. 情报学报, 2018,37(3):294-304.
[1] ( Li Yuelin, Zhang Xiu, Wang Shanshan . Health Information Quality in Social Media: An Analysis Based on the Features of Real and Fake Health Information[J]. Journal of the China Society for Scientific and Technical Information, 2018,37(3):294-304.)
[2] 张海涛, 崔阳, 王丹 , 等. 基于概念格的在线健康社区用户画像研究[J]. 情报学报, 2018,37(9):912-922.
[2] ( Zhang Haitao, Cui Yang, Wang Dan , et al. Study of Online Healthy Community User Profile Based on Concept Lattice[J]. Journal of the China Society for Scientific and Technical Information, 2018,37(9):912-922.)
[3] 熊回香, 李晓敏, 李跃艳 . 基于图书评论属性挖掘的群组推荐研究[J]. 数据分析与知识发现, 2020,4(2/3):214-222.
[3] ( Xiong Huixiang, Li Xiaomin, Li Yueyan . Group Recommendation Based on Attribute Mining of Book Reviews[J]. Data Analysis and Knowledge Discovery, 2020,4(2/3):214-222.)
[4] 单晓红, 王春稳, 刘晓燕 , 等. 基于在线评论的混合推荐算法[J]. 系统工程, 2019,37(6):130-138.
[4] ( Shan Xiaohong, Wang Chunwen, Liu Xiaoyan , et al. Hybrid Recommendation Algorithms Based on Online Review[J]. Systems Engineering, 2019,37(6):130-138.)
[5] Kosko B . Fuzzy Cognitive Maps[J]. International Journal of Man-Machine Studies, 1986,24(1):65-75.
[6] Jetter A, Schweinfort W . Building Scenarios with Fuzzy Cognitive Maps: An Exploratory Study of Solar Energy[J]. Futures, 2010,43(1):52-66.
[7] Peng Z, Liu W, An S . Haze Pollution Causality Mining and Prediction Based on Multi-Dimensional Time Series with PS-FCM[J]. Information Sciences, 2020,523:307-317.
[8] Amirkhani A, Papageorgiou E I, Mohseni A , et al. A Review of Fuzzy Cognitive Maps in Medicine: Taxonomy, Methods, and Applications[J]. Computer Methods and Programs in Biomedicine, 2017,142:129-145.
pmid: 28325441
[9] Akinnuwesi B A, Adegbite B A, Adelowo F , et al. Decision Support System for Diagnosing Rheumatic-Musculoskeletal Disease Using Fuzzy Cognitive Map Technique[J]. Informatics in Medicine Unlocked, 2020,18:100279.
[10] 房小可, 叶莎莎, 严承希 . 融合情境语义的虚拟学术社区知识推荐模型研究[J]. 情报理论与实践, 2019,42(9):154-159.
[10] ( Fang Xiaoke, Ye Shasha, Yan Chengxi . Research on Knowledge Recommendation Model Integrating Situational Semantics Based on Virtual Academic Community[J]. Information Studies: Theory & Application, 2019,42(9):154-159.)
[11] 李珊, 姚叶慧, 厉浩 , 等. 基于ISA联合聚类的组推荐算法研究[J]. 数据分析与知识发现, 2019,3(8):77-87.
[11] ( Li Shan, Yao Yehui, Li Hao , et al. ISA Biclustering Algorithm for Group Recommendation[J]. Data Analysis and Knowledge Discovery, 2019,3(8):77-87.)
[12] 杨梦晴, 朱庆华 . 在线健康社区用户个人健康信息管理行为特征研究[J]. 图书情报工作, 2020,64(1):105-112.
[12] ( Yang Mengqing, Zhu Qinghua . Research on the Behavior Characteristics of Personal Health Information Management of Online Health Community Users[J]. Library and Information Service, 2020,64(1):105-112.)
[13] 许云红, 李仕林, 许云丽 . 在线健康社区不同级别用户的参与行为研究:基于增长模型视角[J]. 情报杂志, 2020,39(8):137-144.
[13] ( Xu Yunhong, Li Shilin, Xu Yunli . Participation Behavior of Users with Different Level Ranks in Online Health Communities: A Growth Model Perspective[J]. Journal of Intelligence, 2020,39(8):137-144.)
[14] Marchant T . Cognitive Maps and Fuzzy Implications[J]. European Journal of Operational Research, 1999,114(3):626-637.
[15] Fasli M . Interrelations Between the BDI Primitives:Towards Heterogeneous Agents[J]. Cognitive Systems Research, 2003,4(1):1-22.
[16] Harmati I, Koczy L T. Stability of Fuzzy Cognitive Maps with Interval Weights [C]//Proceedings of the 11th Conference of the European Society for Fuzzy Logic and Technology. 2019.
[17] Qiu J, Gu W, Wang G . Causal Reasoning of Emergency Cases Based on Fuzzy Cognitive Map[J]. Procedia Computer Science, 2019,159:1238-1245.
[18] 于丰畅, 陆伟 . 关键词共现网络视角下的学科基础词汇发现[J]. 图书情报工作, 2019,63(9):95-100.
[18] ( Yu Fengchang, Lu Wei . The Discovery of Subject Basic Vocabulary from the Perspective of Keyword Co-occurrence Network[J]. Library and Information Service, 2019,63(9):95-100.)
[19] 李纲, 巴志超 . 共词分析过程中的若干问题研究[J]. 中国图书馆学报, 2017,43(4):93-113.
[19] ( Li Gang, Ba Zhichao . Co-word Analysis: Limitations and Solutions[J]. Journal of Library Science in China, 2017,43(4):93-113.)
[20] 周磊, 杨威, 张玉峰 . 共现矩阵聚类分析的问题与再思考[J]. 情报杂志, 2014,33(6):32-36, 27.
[20] ( Zhou Lei, Yang Wei, Zhang Yufeng . Issues and Re-consideration on Cluster Analysis in Co-occurrence Matrix[J]. Journal of Intelligence, 2014,33(6):32-36, 27.)
[21] 陈果, 何适圆 . 面向网络社区的领域多元概念关联体系融合: 机理与实现[J]. 情报理论与实践, 2019,42(1):147-152.
[21] ( Chen Guo, He Shiyuan . Integration of Multi-component Domain Concept Relation Systems for Network Community: Mechanism and Implementation[J]. Information Studies: Theory & Application, 2019,42(1):147-152.)
[22] 张路 . 基于数据挖掘的选煤生产过程智能诊断研究[D]. 徐州: 中国矿业大学, 2019.
[22] ( Zhang Lu . Research on Intelligent Diagnosis of Coal Preparation Process Based on Data Mining[D]. Xuzhou: China University of Mining and Technology, 2019.)
[23] 张华平, 商建云 . NLPIR-Parser:大数据语义智能分析平台[J]. 语料库语言学, 2019,6(1):87-104.
[23] ( Zhang Huaping, Shang Jianyun . NLPIR-Parser: An Intelligent Semantic Analysis Toolkit for Big Data[J]. Corpus Linguistics, 2019,6(1):87-104.)
[24] 姜春林, 陈玉光 . CSSCI数据导入Bibexcel实现共现矩阵的方法及实证研究[J]. 图书馆杂志, 2010,29(4):58-63, 42.
[24] ( Jiang Chunlin, Chen Yuguang . Transform CSSCI Data to Bibexcel Data to Actualize Co-occurrence Matrix and a Case Study[J]. Library Journal, 2010,29(4):58-63, 42.)
[25] 张亮, 张继春, 张立才 , 等. 负压封闭引流联合前列地尔及依帕司他治疗糖尿病足溃疡[J]. 中国临床研究, 2020,33(3):359-362.
[25] ( Zhang Liang, Zhang Jichun, Zhang Licai , et al. Vacuum Sealing Drainage Combined with Alprostadil and Epalrestat in the Treatment of Diabetic Foot Ulcer[J]. Chinese Journal of Clinical Research, 2020,33(3):359-362.)
[26] Kosko B . Hidden Patterns in Combined and Adaptive Knowledge Networks[J]. International Journal of Approximate Reasoning, 1988,2(4):377-393.
[27] 杨钊斌, 陈惠萍, 赖嘉伟 , 等. 片仔癀对糖尿病并脓毒症肝损伤的治疗作用[J]. 糖尿病新世界, 2019,22(19):18-20.
[27] ( Yang Zhaobin, Chen Huiping, Lai Jiawei , et al. Therapeutic Effect of Pien Tze Huang on Liver Damage Caused by Diabetes and Sepsis[J]. Diabetes New World, 2019,22(19):18-20.)
[28] Ma H, Jia M, Xie M, et al. A Microblog Recommendation Algorithm Based on Multi-Tag Correlation [C]//Proceedings of the 8th International Conference on Knowledge Science, Engineering and Management, Chongqing,China. 2015, 483-488.
[29] 马慧芳, 张迪, 赵卫中 , 等. 基于超图随机游走标签扩充的微博推荐方法[J]. 软件学报, 2019,30(11):3397-3412.
[29] ( Ma Huifang, Zhang Di, Zhao Weizhong , et al. Microblog Recommendation Method Based on Hypergraph Random Walk Tag Extension[J]. Journal of Software, 2019,30(11):3397-3412.)
[1] Quan Lu,Anqi Zhu,Jiyue Zhang,Jing Chen. Research on User Information Requirement in Chinese Network Health Community: Taking Tumor-forum Data of Qiuyi as an Example[J]. 数据分析与知识发现, 2019, 3(4): 22-32.
[2] Lei Yang,Zirun Wang,Guisheng Hou. Discovering Topics of Online Health Community with Q-LDA Model[J]. 数据分析与知识发现, 2019, 3(11): 52-59.
[3] Xu Yingnan. Analysis of Commodity Parameters Browsing Preference in Consumer’s Online Shopping Decision-making——Taking Digital Camera for Example[J]. 现代图书情报技术, 2012, (12): 52-57.
  Copyright © 2016 Data Analysis and Knowledge Discovery   Tel/Fax:(010)82626611-6626,82624938   E-mail:jishu@mail.las.ac.cn