[1]于志刚,成思源,杨雪荣,等.融合可拓学与在线评论挖掘的产品改进需求识别研究[J].智能系统学报,2023,18(5):1047-1059.[doi:10.11992/tis.202208052]
YU Zhigang,CHENG Siyuan,YANG Xuerong,et al.Product improvement demand identification incorporating topology and online comment mining[J].CAAI Transactions on Intelligent Systems,2023,18(5):1047-1059.[doi:10.11992/tis.202208052]
点击复制
YU Zhigang,CHENG Siyuan,YANG Xuerong,et al.Product improvement demand identification incorporating topology and online comment mining[J].CAAI Transactions on Intelligent Systems,2023,18(5):1047-1059.[doi:10.11992/tis.202208052]
融合可拓学与在线评论挖掘的产品改进需求识别研究
《智能系统学报》[ISSN 1673-4785/CN 23-1538/TP] 卷:
18
期数:
2023年第5期
页码:
1047-1059
栏目:
学术论文—机器学习
出版日期:
2023-09-05
- Title:
- Product improvement demand identification incorporating topology and online comment mining
- Keywords:
- topology; online comment; matter-element model; affair-element model; emotion inclination pointwise mutual information algorithm; correlative network; implication system; demand identification
- 分类号:
- TP18
- 摘要:
- 利用形式化、条理化的手段从在线评论当中挖掘并理清不同类别的产品改进需求是产品改进中亟待解决的问题，考虑到产品设计信息的多层次、多特征性，提出一种基于基元模型的在线评论产品设计信息分层表示及识别产品改进需求的可拓分析方法。利用PYLDA-vis交互式可视化库从在线评论中提取产品要素，将产品要素划分为实体?功能?属性3个层次，并采用物元模型进行分层表达；结合情感倾向点互信息算法计算各个层次用户观点的评价值，借助事元模型对评价值较低的对象特征进行用户需求表达；利用相关网分析方法确定缺陷物元之间的关联关系，蕴含系分析方法找出产品改进需求的实现途径。以一款养生壶的在线评论为例，通过对在线评论的挖掘找到用户对该产品的改进需求，并识别出实现需求的途径。融合可拓学与在线评论挖掘的产品改进需求识别方法可实现用户需求的分层表达，形式化、条理化地理清产品改进设计信息，为识别产品改进途径提供科学依据和有效方法。
- Abstract:
- Formal and organized mining and clarifying of different categories of product improvement demands from online comments are necessary for the product improvement process. On considering the multilevel and multifeature nature of product design information, a topological analysis method based on a matter-element model used for hierarchical representation of online review product design information and identification of product improvement demands was proposed. PYLDAvis interactive visualization library was used to extract product elements from online comments, and the product elements were then divided into three levels, namely entity, function, and attribute. In addition, the matter-element model was used for hierarchical expression. The emotion inclination pointwise mutual information algorithm was combined to calculate the evaluation value of user opinions at all levels, and an affair–element model was used to express the user’s demand for the object features with low evaluation value. The correlation network analysis method was used to determine the relationship between defective matter elements, and the implication system analysis method was used to determine the realization of the product improvement demand. The case study took the online reviews of a health pot as an example, the user’s improvement demands for the product were found by mining online reviews, and means to achieve the demands were identified. A product improvement demand identification method incorporating topology and online comment mining can realize the hierarchical expression of user demands and clarify product improvement design information in a formalized and organized manner, providing a scientific basis and effective methods for identifying the improvement paths of products.
- 参考文献/References:
-
[1] LAI Xinjun, ZHANG Qixiang, CHEN Qingxin, et al. The analytics of product-design requirements using dynamic Internet data: application to Chinese smartphone market[J]. International journal of production research, 2019, 57(18): 5660-5684.
[2] 冯立杰, 关柯楠, 王金凤. 基于在线评论考虑潜在用户需求的产品创新方案识别研究[J]. 情报理论与实践, 2022, 45(6): 129-137
FENG Lijie, GUAN Kenan, WANG Jinfeng. Research on product innovation scheme identification considering potential user needs based on online comments[J]. Information studies:theory & application, 2022, 45(6): 129-137
[3] 王慧. 基于特征提取和情感分析的用户需求挖掘研究[D]. 杭州: 浙江理工大学, 2019: 5?17.
WANG Hui. Research on user demand mining based on feature extraction and emotion analysis[D]. Hangzhou: Zhejiang Sci-Tech University, 2019: 5?17.
[4] 徐凯. 基于产品特征的用户评论情感倾向分析研究[D]. 合肥: 合肥工业大学, 2015: 8?11.
XU Kai. Research on sentiment orientation analysis of user reviews based on product features[D]. Hefei: Hefei University of Technology, 2015: 8?11.
[5] 李贺, 曹阳, 沈旺, 等. 基于LDA主题识别与Kano模型分析的用户需求研究[J]. 情报科学, 2021, 39(8): 3-11,36
LI He, CAO Yang, SHEN Wang, et al. User demand based on LDA subject identification and Kano model analysis[J]. Information science, 2021, 39(8): 3-11,36
[6] DAY M Y, TENG H C. A study of deep learning to sentiment analysis on word of mouth of smart bracelet[C]//Proceedings of the 2017 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining. New York: ACM, 2017: 763?770.
[7] YEH C T, CHEN Muchen. Applying Kansei Engineering and data mining to design door-to-door delivery service[J]. Computers & industrial engineering, 2018, 120: 401-417.
[8] 李江泳, 张伟. 在线评论数据挖掘视角下的产品改良设计研究[J]. 包装工程, 2021, 42(6): 135-141
LI Jiangyong, ZHANG Wei. Product improvement design from the perspective of online comment data mining[J]. Packaging engineering, 2021, 42(6): 135-141
[9] 萧筝. 客户需求信息处理理论和方法研究[D]. 武汉: 武汉理工大学, 2013: 5?20.
XIAO Zheng. Research on theories and methods of customer requirement information processing[D]. Wuhan: Wuhan University of Technology, 2013: 5?20.
[10] 赵云彦, 韩冬楠, 尤立思, 等. 基于用户需求层次分析的女性安全防护产品设计研究[J]. 包装工程, 2021, 42(16): 120-126
ZHAO Yunyan, HAN Dongnan, YOU Lisi, et al. The design of women’s safety protection products based on the level of users’ demand[J]. Packaging engineering, 2021, 42(16): 120-126
[11] 纪雪, 高琦, 李先飞, 等. 考虑产品属性层次性的评论挖掘及需求获取方法[J]. 计算机集成制造系统, 2020, 26(3): 747-759
JI Xue, GAO Qi, LI Xianfei, et al. Reviews mining and requirements elicitation methodology considering product attributes’ hierarchy[J]. Computer integrated manufacturing systems, 2020, 26(3): 747-759
[12] 肖寒琼, 张馨遇, 肖宇晗, 等. 基于方面词的用户消费心理画像方法[J]. 数据分析与知识发现, 2022, 6(6): 22-31
XIAO Hanqiong, ZHANG Xinyu, XIAO Yuhan, et al. Creating consumer psychology portrait with aspect words[J]. Data analysis and knowledge discovery, 2022, 6(6): 22-31
[13] 张文旭, 肖人彬, 林文广. 基于产品性能词典的评论数据驱动客户需求模型研究[J]. 中国机械工程, 2020, 31(15): 1866-1876
ZHANG Wenxu, XIAO Renbin, LIN Wenguang. Review data driven customer need model research based on product performance lexicon[J]. China mechanical engineering, 2020, 31(15): 1866-1876
[14] 刘晓敏, 李娇蓉, 陈亮, 等. 融合可拓学与多层次知识建模的仿生原型获取[J]. 机械工程学报, 2019, 55(21): 150-160
LIU Xiaomin, LI Jiaorong, CHEN Liang, et al. Bionic prototype acquisition incorporating extension and multi-level knowledge modeling[J]. Journal of mechanical engineering, 2019, 55(21): 150-160
[15] BLEI D, NG A, JORDAN M. Latent dirichlet allocation[J]. Journal of machine learning research, 2003(3): 933-1022.
[16] 杨凡. 基于LDA主题模型的在线评论聚类分析与推荐[D]. 大连: 大连理工大学, 2018.
YANG Fan. Online reviews clustering analysis and recommendation based on LDA topic model[D]. Dalian: Dalian University of Technology, 2018.
[17] SIEVERT C, SHIRLEY K. LDAvis: a method for visualizing and interpreting topics[C]//Proceedings of the Workshop on Interactive Language Learning, Visualization, and Interfaces. Stroudsburg: Association for Computational Linguistics, 2014: 63?70.
[18] 陈晓菁, 成思源, 杨雪荣. 面向专利产品创新设计的可拓变换方法研究[J]. 包装工程, 2020, 41(24): 137-142
CHEN Xiaojing, CHENG Siyuan, YANG Xuerong. Extension transform method for patent product innovation design[J]. Packaging engineering, 2020, 41(24): 137-142
[19] GRIFFIN A, HAUSER J R. The voice of the customer[J]. Marketing science, 1993, 12(1): 1-27.
[20] 高新勤, 金雨昊, 王雪萍, 等. 基于在线评论挖掘的产品感性评价方法研究[J]. 现代制造工程, 2021(12): 13-20
GAO Xinqin, JIN Yuhao, WANG Xueping, et al. Research on product perceptual evaluation method based on online review mining[J]. Modern manufacturing engineering, 2021(12): 13-20
[21] 杨春燕. 可拓创新方法[M]. 北京: 科学出版社, 2017.
[22] TURNEY P D, LITTMAN M L. Measuring praise and criticism[J]. ACM transactions on information systems, 2003, 21(4): 315-346.
[23] 彭德中, 章毅, 吕建成, 等. 一种基于SO-PMI商品评价信息的情感分析方法: CN105022805B[P]. 2018-05-04.
PENG Dezhong, ZHANG Yi, LYU Jiancheng, et al. Emotional analysis method based on SO-PMI (semantic orientation-pointwise mutual information) commodity evaluation information: CN105022805B[P]. 2018-05-04.
[24] 李文军, 杨春燕, 汤龙, 等. 可拓学中相关关系的变换方法研究[J]. 智能系统学报, 2019, 14(4): 619-626
LI Wenjun, YANG Chunyan, TANG Long, et al. Research on the transformation method for the correlation relation in extenics[J]. CAAI transactions on intelligent systems, 2019, 14(4): 619-626
[25] 花黄伟, 杨春燕. 基于因果分析的事元蕴含系及其应用[J]. 智能系统学报, 2017, 12(1): 60-66
HUA Huangwei, YANG Chunyan. A ffair-elements implication system andits application based on causal analysis[J]. CAAI transactions on intelligent systems, 2017, 12(1): 60-66
- 相似文献/References:
-
[1]张海涛,董春游.基于Web的可拓专家系统研究[J].智能系统学报,2009,4(2):175.
ZHANG Hai-tao,DONG Chun-you.An extendable expert system based on the Web[J].CAAI Transactions on Intelligent Systems,2009,4():175.
[2]杨春燕,蔡 文.挖掘同对象信息元的传导知识[J].智能系统学报,2008,3(4):305.
YANG Chun-yan,CA IWen.M in ing conductive knowledge in information elements possessed by an identical object[J].CAAI Transactions on Intelligent Systems,2008,3():305.
[3]叶广仔,李卫华,李淑飞.可拓策略生成系统的构件化设计与实现[J].智能系统学报,2010,5(4):366.
YE Guang-zai,LI Wei-hua,LI Shu-fei.The Design and Implementation of Extension Strategy Generation System Based on Components[J].CAAI Transactions on Intelligent Systems,2010,5():366.
[4]李承晓,李卫华.租房可拓策略生成系统[J].智能系统学报,2011,6(3):272.
LI Chengxiao,LI Weihua.Research on a tenement extension strategy generation system[J].CAAI Transactions on Intelligent Systems,2011,6():272.
[5]曹礼园,李卫华.基于基元和知网的问题相关度计算[J].智能系统学报,2015,10(2):234.[doi:10.3969/j.issn.1673-4785]
CAO Liyuan,LI Weihua.Calculation of correlation problem based on basic element and HowNet[J].CAAI Transactions on Intelligent Systems,2015,10():234.[doi:10.3969/j.issn.1673-4785]
[6]文天柱,许爱强,邓露.可拓否定选择算法及其在故障诊断中的应用[J].智能系统学报,2015,10(3):488.[doi:10.3969/j.issn.1673-4785.201402020]
WEN Tianzhu,XU Aiqiang,DENG Lu.A new negative selection algorithm based on Extenics and its application in fault diagnosis[J].CAAI Transactions on Intelligent Systems,2015,10():488.[doi:10.3969/j.issn.1673-4785.201402020]
[7]陶星,李卫华,汪中飞.基于知网的可拓领域信息元库的构建方法[J].智能系统学报,2015,10(5):790.[doi:10.11992/tis.201412006]
TAO Xing,LI Weihua,WANG Zhongfei.Construction of HowNet-based extendable domain information element base[J].CAAI Transactions on Intelligent Systems,2015,10():790.[doi:10.11992/tis.201412006]
[8]杨春燕,李卫华,汤龙,等.基于可拓学和HowNet的策略生成系统研究进展[J].智能系统学报,2015,10(6):823.[doi:10.11992/tis.201507057]
YANG Chunyan,LI Weihua,TANG Long,et al.Strategy-generating system based on Extenics and HowNet[J].CAAI Transactions on Intelligent Systems,2015,10():823.[doi:10.11992/tis.201507057]
[9]王定桥,李卫华,杨春燕.从用户需求语句建立问题可拓模型的研究[J].智能系统学报,2015,10(6):865.[doi:10.11992/tis.201507038]
WANG Dingqiao,LI Weihua,YANG Chunyan.Research on building an extension model from user requirements[J].CAAI Transactions on Intelligent Systems,2015,10():865.[doi:10.11992/tis.201507038]
[10]范锐,彭银桥,陈月峰,等.基于可拓学的自适应软件形式化方法[J].智能系统学报,2015,10(6):901.[doi:10.11992/tis.201507043]
FAN Rui,PENG Yinqiao,CHEN Yuefeng,et al.A method for self-adaptive software formal modeling by Extenics[J].CAAI Transactions on Intelligent Systems,2015,10():901.[doi:10.11992/tis.201507043]
备注/Memo
收稿日期:2022-8-31。
基金项目:广东省研究生教育创新计划项目（2021JGXM043）；广东省科技计划项目（2014A040402004）.
作者简介:于志刚，硕士研究生,主要研究方向为可拓创新方法、数据挖掘;成思源，教授，博士，主要研究方向为技术创新方法、专利挖掘、逆向工程技术。主持并完成国家自然科学基金项目1项、广东省自然科学基金项目2项、广东省科技计划项目4项。获国家级教学成果二等奖1项、广东省教学成果一等奖3项，获得发明专利授权20余项。出版著作8部，发表学术论文70余篇;杨雪荣，副教授，博士，主要研究方向为机器视觉、逆向工程技术、TRIZ理论及应用。主持并完成国家自然科学基金项目1项、广东省自然科学基金项目1项、广东省科技计划项目2项。获广东省教学成果一等奖1项，获发明专利授权10余项。出版 著作5部，发表学术论文40余篇
通讯作者:成思源.E-mail：imdesign@gdut.edu.cn
更新日期/Last Update:
1900-01-01