Please wait a minute...
Advanced Search
数据分析与知识发现  2021, Vol. 5 Issue (11): 89-101     https://doi.org/10.11925/infotech.2096-3467.2021.0544
  研究论文 本期目录 | 过刊浏览 | 高级检索 |
基于图注意力网络的开放式创新社区用户创意潜在价值发现研究*
王松1,杨洋1(),刘新民1,2
1山东科技大学经济管理学院 青岛 266590
2青岛农业大学 青岛 266109
Discovering Potentialities of User Ideas from Open Innovation Communities with Graph Attention Network
Wang Song1,Yang Yang1(),Liu Xinmin1,2
1College of Economics and Management, Shandong University of Science and Technology, Qingdao 266590, China
2Qingdao Agricultural University, Qingdao 266109, China
全文: PDF (1800 KB)   HTML ( 28
输出: BibTeX | EndNote (RIS)      
摘要 

【目的】 为解决开放式创新社区中因信息过载导致用户创意价值未能及时发现的问题,探索用户创意潜在价值早期发现方法,提高社区创新资源的利用效果。【方法】 设计用户创意的双重网络结构,构建基于图注意力网络的用户创意潜在价值发现模型,学习表达双重网络的节点特征及网络间映射关系,实现用户创意潜在价值早期发现。【结果】 应用典型开放式创新社区数据进行实证研究,结果表明,基于图注意力网络、使用双重网络结构特征的用户创意潜在价值发现模型的准确率为90.49%,高于其他相关基线模型。【局限】 仅在魅族社区数据集上验证模型,未来可拓展到其他领域开放式创新社区。【结论】 双重网络结构与图注意力网络相结合,有效提升了社区用户创意潜在价值发现模型的准确性,为社区针对性引导用户参与,充分挖掘社区创新资源提供技术支持。

服务
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章
王松
杨洋
刘新民
关键词 开放式创新社区双重网络结构图注意力网络潜在价值发现    
Abstract

[Objective] This paper proposes a method to discover the potentialities of user ideas, aiming to effectively identify creative ones from open innovation communities. [Methods] First, we analyzed the formation process of creative value and constructed the dual network structure for user ideas. Then, we developed a model based on graph attention networks to discover their potential values. Third, we trained the model to learn the node characteristics of this dual network and mapped the relationships between networks. [Results] The model was empirically examined with data from a typical open innovation community. The results show that the proposed model achieved an accuracy rate of 90.49%, higher than other relevant baseline models. [Limitations] The model was only validated on the Meizu community dataset, which needs to be expanded to other open innovation communities in future studies. [Conclusions] The combination of the dual network structure and the graph attention network can effectively identify the potential value of user ideas in the open innovation community, which provides technical support for increasing user participation and fully utilizes the community innovation resources.

Key wordsOpen Innovation Community    Dual Network Structure Model    Graph Attention Network    Potential Value Discovery
收稿日期: 2021-06-01      出版日期: 2021-12-23
ZTFLH:  G206  
基金资助:*国家自然科学基金项目(71471105);山东省社会科学规划项目(18CGLJ38)
通讯作者: 杨洋,ORCID:0000-0003-1967-9781     E-mail: 17863907712@163.com
引用本文:   
王松, 杨洋, 刘新民. 基于图注意力网络的开放式创新社区用户创意潜在价值发现研究*[J]. 数据分析与知识发现, 2021, 5(11): 89-101.
Wang Song, Yang Yang, Liu Xinmin. Discovering Potentialities of User Ideas from Open Innovation Communities with Graph Attention Network. Data Analysis and Knowledge Discovery, 2021, 5(11): 89-101.
链接本文:  
https://manu44.magtech.com.cn/Jwk_infotech_wk3/CN/10.11925/infotech.2096-3467.2021.0544      或      https://manu44.magtech.com.cn/Jwk_infotech_wk3/CN/Y2021/V5/I11/89
Fig.1  用户创意双重网络结构模型
所属网络 名称 符号 含义
用户社交网络 用户专业性 professionalism 用户节点的专业程度
度中心性 degree 与其他用户节点的连接数量
中间中心性 betweenness 用户节点位置的重要性
接近中心性 closeness 用户节点到其他节点用户的距离
用户领袖性 pagerank 用户节点的影响力
内容知识网络 全息性 holographic 创意内容可拓展的信息量,全面程度
丰富性 richness 呈现创意的多种形式
情感极性 emotionality 创意带有的情感倾向
语义关联性 adjacency 创意核心语义关联的复杂程度
Table 1  用户创意双重网络结构特征
Fig.2  GAT用户创意潜在价值发现模型结构
符号 含义 符号 含义
post_id 创意编号 post_if_has_pic 创意是否有图片
post_theme 创意主题 post_content 创意内容
post_author_id 发表创意的用户编号 post_author_name 发表创意的用户昵称
post_reputation 发表创意的用户声望 review_author_id 评论用户编号
review_author_name 评论用户昵称 review_content 评论内容
Table 2  基础数据特征
Fig.3  魅族社区用户创意的双重网络结构模型
Fig.4  创意价值的主题数-困惑度曲线图
Fig.5  魅族社区用户创意潜在价值发现模型训练结果对比
输入 准确率 误差
用户创意双重网络特征 90.49% 0.882 8
内容知识网络特征 79.31% 0.909 0
用户社交网络特征 60.00% 1.194 7
Table 3  不同输入的结果比较表
模型 准确率
图注意力模型 90.49%
图神经网络 78.64%
BP神经网络 35.25%
CNN神经网络 45.68%
支持向量机 34.58%
随机森林 34.24%
Table 4  实验结果对比表
[1] Von Briel F, Recker J. Lessons from a Failed Implementation of an Online Open Innovation Community in an Innovative Organization[J]. MIS Quarterly Executive, 2017, 16(1):35-46.
[2] 胡媛, 韦肖莹, 王灿. 微博信息质量评价指标体系构建研究[J]. 情报科学, 2017, 35(6):44-50.
[2] (Hu Yuan, Wei Xiaoying, Wang Can. Research on the Construction of Micro-blog Information Quality Evaluation Indicator System[J]. Information Science, 2017, 35(6):44-50.)
[3] 易明, 张婷婷, 李梓奇. 多维特征下社会化问答社区答案排序研究[J]. 图书情报工作, 2020, 64(17):103-113.
[3] (Yi Ming, Zhang Tingting, Li Ziqi. Research on the Ranking of Social Q&A Community Answers Based on Multidimensional Features[J]. Library and Information Service, 2020, 64(17):103-113.)
[4] 易明, 张婷婷. 大众性问答社区答案质量排序方法研究[J]. 数据分析与知识发现, 2019, 3(6):12-20.
[4] (Yi Ming, Zhang Tingting. Ranking Answer Quality of Popular Q&A Community[J]. Data Analysis and Knowledge Discovery, 2019, 3(6):12-20.)
[5] 吴雅威, 张向先, 陶兴, 等. 基于用户感知的学术问答社区答案质量评价指标构建[J]. 情报科学, 2020, 38(10):141-147.
[5] (Wu Yawei, Zhang Xiangxian, Tao Xing, et al. Construction of Answer Quality Evaluation Index Based on User Perception of Academic Question and Answer Community[J]. Information Science, 2020, 38(10):141-147.)
[6] 祝琳琳, 李贺, 刘金承, 等. 在线评论信息质量感知评价指标体系构建研究[J]. 情报理论与实践, 2021, 44(4):1-15.
[6] (Zhu Linlin, Li He, Liu Jincheng, et al. Research on the Construction of Evaluating Indicator System of Information Quality Perception for Online Reviews[J]. Information Studies: Theory & Application, 2021, 44(4):1-15.)
[7] 单晓红, 王春稳, 刘晓燕, 等. 基于知识网络的开放式创新社区知识发现研究[J]. 复杂系统与复杂性科学, 2020, 17(1):62-70,94.
[7] (Shan Xiaohong, Wang Chunwen, Liu Xiaoyan, et al. Research on Knowledge Discovery of Open Innovation Community Based on Knowledge Network[J]. Complex Systems and Complexity Science, 2020, 17(1):62-70, 94.)
[8] 沈旺, 李世钰, 刘嘉宇, 等. 问答社区回答质量评价体系优化方法研究[J]. 数据分析与知识发现, 2021, 5(2):83-93.
[8] (Shen Wang, Li Shiyu, Liu Jiayu, et al. Research on the Optimization Method of Q&A Community Response Quality Evaluation[J]. Data Analysis and Knowledge Discovery, 2021, 5(2):83-93.)
[9] 郭顺利, 张向先, 陶兴, 等. 社会化问答社区用户生成答案质量自动化评价研究——以“知乎”为例[J]. 图书情报工作, 2019, 63(11):118-130.
[9] (Guo Shunli, Zhang Xiangxian, Tao Xing, et al. Research on Automated Evaluation of User Generated Answer Quality in Social Question and Answer Community - Taking “Zhihu” as an Example[J]. Library and Information Service, 2019, 63(11):118-130.)
[10] Thomas N K, Welling M. Semi-Supervised Classification with Graph Convolutional Networks[OL]. arXiv Preprint, arXiv: 1609.02907.
[11] 洪闯, 李贺, 毛太田. 开放式创新社区用户知识贡献的采纳机理研究[J]. 现代情报, 2020, 40(5):33-40.
[11] (Hong Chuang, Li He, Mao Taitian. Study on the Adoption Mechanism of Knowledge Contribution from Open Innovation Community Users[J]. Journal of Modern Information, 2020, 40(5):33-40.)
[12] 张晓娟, 周学春. 社区治理策略、用户就绪和知识贡献研究:以百度百科虚拟社区为例[J]. 管理评论, 2016, 28(9):72-82.
[12] (Zhang Xiaojuan, Zhou Xuechun. Research on Community Governing Strategy, User Readiness and Knowledge Contribution: Taking Baidu Encyclopedia as an Example[J]. Management Review, 2016, 28(9):72-82.)
[13] 高贝伦, 韦铁. 创意标题对创意实施可能性的影响:基于首因效应与认知心理学视角[J]. 科技管理研究, 2019, 39(11):275-282.
[13] (Gao Beilun, Wei Tie. Impact of Idea Titles on Idea Implementation: Based on Primacy Effect and Cognitive Psychology[J]. Science and Technology Management Research, 2019, 39(11):275-282.)
[14] 李贺, 祝琳琳, 闫敏, 等. 开放式创新社区用户信息有用性识别研究[J]. 数据分析与知识发现, 2018, 2(12):12-22.
[14] (Li He, Zhu Linlin, Yan Min, et al. Identifying Useful Information from Open Innovation Community[J]. Data Analysis and Knowledge Discovery, 2018, 2(12):12-22.)
[15] 沈洪洲, 史俊鹏, 马巧慧. 社会化问答社区回答内容质量影响特征研究——以“知乎”为例[J]. 情报杂志, 2020, 39(10):169-175,202.
[15] (Shen Hongzhou, Shi Junpeng, Ma Qiaohui. Research on Influential Features of the Answer Quality in Social Q&A Community: Taking “Zhihu” as an Example[J]. Journal of Intelligence, 2020, 39(10):169-175, 202.)
[16] 张海涛, 张鑫蕊, 周红磊, 等. 融合用户偏好与内容特征的短视频传播效果评价研究[J]. 图书情报工作, 2020, 64(16):81-91.
[16] (Zhang Haitao, Zhang Xinrui, Zhou Honglei, et al. Research on the Evaluation of Short Video Communication Effect Based on User Preference and Content Characteristics[J]. Library and Information Service, 2020, 64(16):81-91.)
[17] Klein M, Garcia A C B. High-speed Idea Filtering with the Bag of Lemons[J/OL]. SSRN Electronic Journal, 2014. http://dx.doi.org/10.2139/ssrn.2501787.
[18] 金燕, 孙佳佳. 基于用户画像的UGC质量预判模型[J]. 情报理论与实践, 2019, 42(10):77-83.
[18] (Jin Yan, Sun Jiajia. UGC Quality Prediction Model Based on Persona[J]. Information Studies: Theory & Application, 2019, 42(10):77-83.)
[19] 尹鹏博, 潘伟民, 彭成, 等. 基于用户特征分析的微博谣言早期检测研究[J]. 情报杂志, 2020, 39(7):81-86.
[19] (Yin Pengbo, Pan Weimin, Peng Cheng, et al. Research on Early Detection of Weibo Rumors Based on User Characteristics Analysis[J]. Journal of Intelligence, 2020, 39(7):81-86.)
[20] 姜雯, 许鑫, 武高峰. 附加情感特征的在线问答社区信息质量自动化评价[J]. 图书情报工作, 2015, 59(4):100-105.
[20] (Jiang Wen, Xu Xin, Wu Gaofeng. Online Q&A Community Automatically Information Quality Evaluation with Sentiment Feature[J]. Library and Information Service, 2015, 59(4):100-105.)
[21] Sahu T P, Nagwani N K, Verma S. Topical Authoritative Answerer Identification on Q&A Posts Using Supervised Learning in CQA Sites[C]// Proceedings of the 9th Annual ACM India Conference. 2016: 129-132.
[22] 朱益平, 杜海娇, 张佳, 等. 基于RS-BP神经网络的政务微信公众号信息质量评价模型研究[J]. 情报科学, 2021, 39(2):54-61,69.
[22] (Zhu Yiping, Du Haijiao, Zhang Jia, et al. Research on Information Quality Evaluation Model of Government WeChat Public Account Based on RS-BP Neural Networks[J]. Information Science, 2021, 39(2):54-61, 69.)
[23] Yue T, Wang H, Cheng S, et al. Deep Learning Based QoE Evaluation for Internet Video[J]. Neurocomputing, 2020, 386:179-190.
doi: 10.1016/j.neucom.2019.12.082
[24] Li L, He D Q, Jeng W, et al. Answer Quality Characteristics and Prediction on an Academic Q&A Site: A Case Study on ResearchGate[C]// Proceedings of the 24th International Conference on World Wide Web. 2015: 1453-1458.
[25] 王伟, 冀宇强, 王洪伟, 等. 中文问答社区答案质量的评价研究:以知乎为例[J]. 图书情报工作, 2017, 61(22):36-44.
[25] (Wang Wei, Ji Yuqiang, Wang Hongwei, et al. Evaluating Chinese Answers’ Quality in the Community QA System: A Case Study of Zhihu[J]. Library and Information Service, 2017, 61(22):36-44.)
[26] Petar V K, Guillem C, Arantxa C, et al. Graph Attention Networks[OL]. arXiv Preprint, arXiv: 1710.10903v1.
[27] 周欢, 刘嘉, 王欢芳. 基于图神经网络和标签可重叠社区的社会化影视推荐模型[J]. 情报理论与实践, 2021, 44(6):164-170.
[27] (Zhou Huan, Liu Jia, Wang Huanfang. A Social Movie Recommendation Model Based on Graph Neural Network and Tag Overlapping Community[J]. Information Studies: Theory & Application, 2021, 44(6):164-170.)
[28] 张晓丹. 改进的图神经网络文本分类模型应用研究——以NSTL科技期刊文献分类为例[J]. 情报杂志, 2021, 40(1):184-188.
[28] (Zhang Xiaodan. The Application of Improved Graph Neural Network in Big Data Classification of Scientific and Technological Documents[J]. Journal of Intelligence, 2021, 40(1):184-188.)
[29] 秦成磊, 章成志. 基于层次注意力网络模型的学术文本结构功能识别[J]. 数据分析与知识发现, 2020, 4(11):26-42.
[29] (Qin Chenglei, Zhang Chengzhi. Recognizing Structure Functions of Academic Articles with Hierarchical Attention Network[J]. Data Analysis and Knowledge Discovery, 2020, 4(11):26-42.)
[30] Wu Z, Pi D, Chen J, et al. Rumor Detection Based on Propagation Graph Neural Network with Attention Mechanism[J]. Expert Systems with Applications, 2020, 158:113595.
doi: 10.1016/j.eswa.2020.113595
[31] 范涛, 王昊, 吴鹏. 基于图卷积神经网络和依存句法分析的网民负面情感分析研究[J]. 数据分析与知识发现, 2021, 5(9):97-103.
[31] (Fan Tao, Wu Hao, Wu Peng. Sentiment Analysis of Online Users’ Negative Emotions Based on Graph Convolutional Network and Dependency Parsing[J]. Data Analysis and Knowledge Discovery, 2021, 5(9):97-103.)
[32] 顾耀文, 张博文, 郑思, 等. 基于图注意力网络的药物ADMET分类预测模型构建方法[J]. 数据分析与知识发现, 2021, 5(8):76-85.
[32] (Gu Yaowen, Zhang Bowen, Zheng Si, et al. Building a Drug ADMET Classification Prediction Model Based on Graph Attention Network[J]. Data Analysis and Knowledge Discovery, 2021, 5(8):76-85.)
[33] Freeman L C. Centrality in Social Networks Conceptual Clarification[J]. Social Networks, 1978, 1(3):215-239.
doi: 10.1016/0378-8733(78)90021-7
[34] 陈思菁, 李纲, 毛进, 等. 突发事件信息传播网络中的关键节点动态识别研究[J]. 情报学报, 2019, 38(2):178-190.
[34] (Chen Sijing, Li Gang, Mao Jin, et al. Dynamic Identification of Key Nodes in Information Propagation Networks During Emergencies[J]. Journal of the China Society for Scientific and Technical Information, 2019, 38(2):178-190.)
[35] Liu W Y, Wen Y D, Yu Z D, et al. Large-Margin Softmax Loss for Convolutional Neural Networks[C]// Proceedings of the 33rd International Conference on Machine Learning. 2016.
[36] Aur D, Vila-Rodriguez F. Dynamic Cross-Entropy[J]. Journal of Neuroscience Methods, 2017, 275:10-18.
doi: 10.1016/j.jneumeth.2016.10.015
[37] Ng A Y. Feature Selection, L1 vs. L2 Regularization, and Rotational Invariance[C]// Proceedings of the 21st International Conference on Machine Learning. 2004: 78-85.
[38] Chujo K, Utiyama M. Understanding the Role of Text Length, Sample Size and Vocabulary Size in Determining Text Coverage[J]. Reading in a Foreign Language, 2005, 17(1):1-22.
[39] Leebron E J. Visual Persuasion: The Role of Images in Advertising[J]. Journal of Broadcasting and Electronic Media, 1997, 41(4):589-592.
[40] Kakol M, Nielek R, Wierzbicki A. Understanding and Predicting Web Content Credibility Using the Content Credibility Corpus[J]. Information Processing and Management, 2017, 53(5):1043-1061.
doi: 10.1016/j.ipm.2017.04.003
[41] Rokova V, George E I. EMVS: The EM Approach to Bayesian Variable Selection[J]. Journal of the American Statistical Association, 2014, 109(506):828-846.
doi: 10.1080/01621459.2013.869223
[42] Blei D M, Ng A Y, Jordan M I. Latent Dirichlet Allocation[J]. Journal of Machine Learning Research, 2003, 3(4-5):993-1022.
[43] Kingma D, Ba J. A Method for Stochastic Optimization[OL]. arXiv Preprint, arXiv: 1412.6980v9.
[44] Gonzalez R C. Deep Convolutional Neural Networks[J]. IEEE Signal Processing Magazine, 2018, 35(6):79-87.
doi: 10.1109/MSP.2018.2842646
[45] Zhou S Q, Ding L X, Zhang J, et al. Linearization Learning Method of BP Neural Networks[J]. Wuhan University Journal of Natural Sciences, 1997, 2(1):35-39.
doi: 10.1007/BF02834910
[46] Nick G, Matthias S. Support Vector Machines[J]. Stata Journal, 2016, 16(4):917-937.
doi: 10.1177/1536867X1601600407
[47] Breiman L. Random Forests[J]. Machine Learning, 2001, 45(1):5-32.
doi: 10.1023/A:1010933404324
[1] 单晓红,王春稳,刘晓燕,韩晟熙,杨娟. 开放式创新社区领先用户识别——知识基础观视角*[J]. 数据分析与知识发现, 2021, 5(9): 85-96.
[2] 顾耀文, 张博文, 郑思, 杨丰春, 李姣. 基于图注意力网络的药物ADMET分类预测模型构建方法*[J]. 数据分析与知识发现, 2021, 5(8): 76-85.
[3] 李贺, 祝琳琳, 闫敏, 刘金承, 洪闯. 开放式创新社区用户信息有用性识别研究*[J]. 数据分析与知识发现, 2018, 2(12): 12-22.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
版权所有 © 2015 《数据分析与知识发现》编辑部
地址:北京市海淀区中关村北四环西路33号 邮编:100190
电话/传真:(010)82626611-6626,82624938
E-mail:jishu@mail.las.ac.cn