Please wait a minute...
Advanced Search
数据分析与知识发现  2022, Vol. 6 Issue (2/3): 33-44     https://doi.org/10.11925/infotech.2096-3467.2021.0962
  专辑 本期目录 | 过刊浏览 | 高级检索 |
基于多特征的技术融合关系预测及其价值评估*
张金柱(),韩永亮
南京理工大学经济管理学院 南京 210094
Predicting Values of Technology Convergence with Multi-Feature Fusion
Zhang Jinzhu(),Han Yongliang
School of Economics and Management, Nanjing University of Science and Technology, Nanjing 210094, China
全文: PDF (994 KB)   HTML ( 17
输出: BibTeX | EndNote (RIS)      
摘要 

【目的】 综合利用专利分类网络结构特征与文本语义特征,基于多种特征形成技术融合关系预测方法和价值评估方法。【方法】 区分专利与专利分类间的关联强度,构建专利分类共现网络,获取专利分类间的网络结构相似性特征,并根据关联强度赋予专利分类以专利文本,利用文本表示学习方法得到其文本语义相似性特征。根据网络结构特征和文本语义特征构建专利分类间多种相似性指标,融合多种指标构成特征向量,利用随机森林模型学习不同指标的权重和贡献,计算技术融合概率,排序得到候选技术融合关系集合。基于专利分类引用网络特征和文献计量特征,从影响力和成长潜力出发,提出领域技术价值、商业价值和战略价值评估指标,利用被引数加以验证,最后用所得方法评估技术融合关系,获取高价值技术融合关系。【结果】 本文方法的TopK预测准确率比单一特征至少提高20%;评测得到的前10对高价值技术融合关系与真实排名相差极小,平均绝对误差仅为3.2。【局限】 选取的数据库存在数据项不统一的问题;只尝试了单一的随机森林方法,未对其他前沿方法进行验证。【结论】 专利分类关联强度能够提高网络分析预测方法的预测效果,同时多特征融合方法相较于单一特征预测方法,能够提高技术融合关系预测效果;另一方面,本文的价值评估方法能够有效实现高价值技术融合关系价值的筛选。

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
张金柱
韩永亮
关键词 多特征技术融合关系预测价值评估    
Abstract

[Objective] This paper proposes a new method to predict technology convergence relationship and their values based on the patent classification network and text semantic features. [Methods] First, we calculated the correlation between patents and their classification to construct the co-occurrence network and obtain their structure similarity features. Then, we connected patent texts and their classification schema with the correlation strength. We also obtained the text semantic similarity features using text representation learning. Second, we constructed similarity indicators with the network structure and text semantic features, which were fused to create a feature vector. Third, we used the random forest model to learn the weights and contributions of different indicators and calculated the technology fusion probability. We also generated the candidate technology fusion relationship set. Fourth, based on the network characteristics and bibliometric characteristics of patent classification and citation, as well as their influence and potential growth, we created the evaluation indices for their technical, commercial and strategic values. Finally, we used the proposed method to evaluate the technology integration relationship. [Results] The accuracy of the proposed method is at least 20% higher than that of single feature prediction. In addition, the top 10 pairs of high-value technology convergence relations that identified by the proposed method have little difference with the real ranking result, in which the MAE is only 3.2. [Limitations] Some data sets are in-consistent, while more machine learning methods need to be utilized. [Conclusions] The feature convergence method has higher prediction accuracy than traditional methods. The proposed method can also effectively evaluate technology convergence relationship value.

Key wordsMultiple Features    Technology Convergence Relationship    Forecast    Value Assessment
收稿日期: 2021-08-31      出版日期: 2022-04-14
ZTFLH:  G350  
基金资助:*国家自然科学基金面上项目(71974095);江苏省研究生科研与实践创新计划项目的研究成果之一(SJCX21_0159)
通讯作者: 张金柱,ORCID:0000-0001-7581-1850     E-mail: zhangjinzhu@njust.edu.cn
引用本文:   
张金柱, 韩永亮. 基于多特征的技术融合关系预测及其价值评估*[J]. 数据分析与知识发现, 2022, 6(2/3): 33-44.
Zhang Jinzhu, Han Yongliang. Predicting Values of Technology Convergence with Multi-Feature Fusion. Data Analysis and Knowledge Discovery, 2022, 6(2/3): 33-44.
链接本文:  
https://manu44.magtech.com.cn/Jwk_infotech_wk3/CN/10.11925/infotech.2096-3467.2021.0962      或      https://manu44.magtech.com.cn/Jwk_infotech_wk3/CN/Y2022/V6/I2/3/33
Fig.1  总体流程
年份 专利数量 专利引用关系数量 涉及IPC数量 IPC共现关系数量
2015 1 673 145 525 1 483
2016 3 012 473 768 3 517
2017 6 848 1 222 1 222 7 866
2018 10 603 1 605 1 644 11 060
Table 1  数据描述
预测方法 Top10 Top50 Top100 Top500
共同邻居指标 0.8 0.70 0.68 0.52
改进共同邻居指标 0.9 0.72 0.71 0.54
资源分配指标 0.9 0.66 0.60 0.51
改进资源分配指标 0.8 0.68 0.68 0.53
Table 2  节点相似性指标改进效果
潜在技术融合关系 语义相似性 基于中心度的结构相似性 基于关联强度的结构相似性
度中心度 中介中心度 接近中心度 改进的共同邻居指标 改进的资源分配指标
['G05B-015', 'G08B-021'] 0.66 0.11 1.65 0.04 7.25 12.75
['G05B-019', 'H02J-013'] 0.65 0.15 1.69 0.10 2.22 7.03
['G07C-001', 'G07F-007'] 0.66 0.02 1.39 0 1.02 1.55
['G08G-001', 'H04N-007'] 0.66 0.08 1.61 0.04 5.17 9.32
['G01D-021', 'G06Q-050'] 0.66 0.11 1.70 0.03 8.74 19.46
['G06Q-020', 'G06Q-030'] 0.66 0.06 1.54 0.02 3.23 4.84
Table 3  技术融合关系预测的输入向量
Fig.2  不同预测方法预测效果比较
指标 文本语义
方法
改进的CN
方法
改进的RA
方法
多特征融合
方法
AUC 0.83 0.66 0.66 0.93
Table 4  不同预测方法AUC比较
Fig.3  多特征融合方法中各特征贡献度
变量 影响力评价 成长潜力评价 截距
商业影响力 技术影响力 融合增长率 ——
专利权利要求数量 引用网络中介中心度 引用网络PageRank 聚合的PageRank
系数 -9.3e+01 -7.6e-02 1.8e+03 7.3e+02 2.2e-01 2.43
Table 5  回归方程的变量与系数
评价指标 得分
MAE 11.73
RMSE 18.17
Table 6  回归方程的各项评价得分
比较项 被引频次排名比较
评估结果 1 2 3 4 5 6 7 8 9 10
真实结果 2 1 3 5 12 17 8 4 6 7
排名误差 1 1 0 1 7 11 1 4 3 3
Table 7  影响力评价方法的实验结果
Fig.4  高价值技术融合关系筛选
技术融合关系 预测价值排名 真实价值排名 排名误差
['F21V-023', 'H04L-029'] 77 120 43
['H01Q-009', 'H04L-029'] 83 122 39
['F21W-131', 'H04L-029'] 86 183 97
['F21S-009', 'H04L-029'] 91 184 93
['H01L-025', 'H04L-029'] 93 105 12
['F21V-021', 'H04L-029'] 97 186 89
['G16H-010', 'H04L-029'] 98 258 160
Table 8  高价值技术融合关系
[1] Danneels E. The Dynamics of Product Innovation and Firm Competences[J]. Strategic Management Journal, 2002, 23(12):1095-1121.
doi: 10.1002/smj.275
[2] Curran C S, Leker J. Patent Indicators for Monitoring Convergence—Examples from NFF and ICT[J]. Technological Forecasting and Social Change, 2011, 78(2):256-273.
doi: 10.1016/j.techfore.2010.06.021
[3] Borés C, Saurina C, Torres R. Technological Convergence: A Strategic Perspective[J]. Technovation, 2003, 23(1):1-13.
doi: 10.1016/S0166-4972(01)00094-3
[4] Islam N, Miyazaki K. Nanotechnology Innovation System: Understanding Hidden Dynamics of Nanoscience Fusion Trajectories[J]. Technological Forecasting and Social Change, 2009, 76(1):128-140.
doi: 10.1016/j.techfore.2008.03.021
[5] Adner R. When are Technologies Disruptive? a Demand-Based View of the Emergence of Competition[J]. Strategic Management Journal, 2002, 23(8):667-688.
doi: 10.1002/smj.246
[6] Kim M, Baek I, Song M. Topic Diffusion Analysis of a Weighted Citation Network in Biomedical Literature[J]. Journal of the Association for Information Science and Technology, 2018, 69(2):329-342.
doi: 10.1002/asi.23960
[7] Liu J S, Kuan C H. A New Approach for Main Path Analysis: Decay in Knowledge Diffusion[J]. Journal of the Association for Information Science and Technology, 2016, 67(2):465-476.
doi: 10.1002/asi.23384
[8] Park I, Yoon B. Technological Opportunity Discovery for Technological Convergence Based on the Prediction of Technology Knowledge Flow in a Citation Network[J]. Journal of Informetrics, 2018, 12(4):1199-1222.
doi: 10.1016/j.joi.2018.09.007
[9] Zhai Y J, Ding Y, Wang F. Measuring the Diffusion of an Innovation: A Citation Analysis[J]. Journal of the Association for Information Science and Technology, 2018, 69(3):368-379.
doi: 10.1002/asi.23898
[10] Yan E J. Research Dynamics, Impact, and Dissemination: A Topic-Level Analysis[J]. Journal of the Association for Information Science and Technology, 2015, 66(11):2357-2372.
doi: 10.1002/asi.23324
[11] Nieminen P, Pölönen I, Sipola T. Research Literature Clustering Using Diffusion Maps[J]. Journal of Informetrics, 2013, 7(4):874-886.
doi: 10.1016/j.joi.2013.08.004
[12] Jee S J, Kwon M, Ha J M, et al. Exploring the Forward Citation Patterns of Patents Based on the Evolution of Technology Fields[J]. Journal of Informetrics, 2019, 13(4):100985.
doi: 10.1016/j.joi.2019.100985
[13] Ji J J, Barnett G A, Chu J X. Global Networks of Genetically Modified Crops Technology: A Patent Citation Network Analysis[J]. Scientometrics, 2019, 118(3):737-762.
doi: 10.1007/s11192-019-03006-1
[14] Jamali H R, Azadi-Ahmadabadi G, Asadi S. Interdisciplinary Relations of Converging Technologies: Nano-Bio-Info-Cogno (NBIC)[J]. Scientometrics, 2018, 116(2):1055-1073.
doi: 10.1007/s11192-018-2776-9
[15] Yan E J, Zhu Y J. Adding the Dimension of Knowledge Trading to Source Impact Assessment: Approaches, Indicators, and Implications[J]. Journal of the Association for Information Science and Technology, 2017, 68(5):1090-1104.
doi: 10.1002/asi.23670
[16] Fukugawa N. Knowledge Creation and Dissemination by Kosetsushi in Sectoral Innovation Systems: Insights from Patent Data[J]. Scientometrics, 2016, 109(3):2303-2327.
doi: 10.1007/s11192-016-2124-x
[17] Luan C J, Liu Z Y, Wang X W. Divergence and Convergence: Technology-Relatedness Evolution in Solar Energy Industry[J]. Scientometrics, 2013, 97(2):461-475.
doi: 10.1007/s11192-013-1057-x
[18] Jeong S, Lee S. What Drives Technology Convergence? Exploring the Influence of Technological and Resource Allocation Contexts[J]. Journal of Engineering and Technology Management, 2015, 36:78-96.
doi: 10.1016/j.jengtecman.2015.05.004
[19] Caviggioli F. Technology Fusion: Identification and Analysis of the Drivers of Technology Convergence Using Patent Data[J]. Technovation, 2016, 55/56:22-32.
doi: 10.1016/j.technovation.2016.04.003
[20] No H J, Park Y. Trajectory Patterns of Technology Fusion: Trend Analysis and Taxonomical Grouping in Nanobiotechnology[J]. Technological Forecasting and Social Change, 2010, 77(1):63-75.
doi: 10.1016/j.techfore.2009.06.006
[21] Kim J, Lee S. Forecasting and Identifying Multi-Technology Convergence Based on Patent Data: The Case of IT and BT Industries in 2020[J]. Scientometrics, 2017, 111(1):47-65.
doi: 10.1007/s11192-017-2275-4
[22] Kim E, Cho Y, Kim W. Dynamic Patterns of Technological Convergence in Printed Electronics Technologies: Patent Citation Network[J]. Scientometrics, 2014, 98(2):975-998.
doi: 10.1007/s11192-013-1104-7
[23] Zhou Y, Dong F, Kong D J, et al. Unfolding the Convergence Process of Scientific Knowledge for the Early Identification of Emerging Technologies[J]. Technological Forecasting and Social Change, 2019, 144:205-220.
doi: 10.1016/j.techfore.2019.03.014
[24] Jeong S, Kim J C, Choi J Y. Technology Convergence: What Developmental Stage are We in?[J]. Scientometrics, 2015, 104(3):841-871.
doi: 10.1007/s11192-015-1606-6
[25] Kim Y J, Lee D H. Technology Convergence Networks for Flexible Display Application: A Comparative Analysis of Latecomers and Leaders[J]. Japan and the World Economy, 2020, 55:101025.
doi: 10.1016/j.japwor.2020.101025
[26] Feng S D, An H Z, Li H J, et al. The Technology Convergence of Electric Vehicles: Exploring Promising and Potential Technology Convergence Relationships and Topics[J]. Journal of Cleaner Production, 2020, 260:120992.
doi: 10.1016/j.jclepro.2020.120992
[27] Cho J H, Lee J, Sohn S Y. Predicting Future Technological Convergence Patterns Based on Machine Learning Using Link Prediction[J]. Scientometrics, 2021, 126(7):5413-5429.
doi: 10.1007/s11192-021-03999-8
[28] Lee C Y, Hong S, Kim J. Anticipating Multi-Technology Convergence: A Machine Learning Approach Using Patent Information[J]. Scientometrics, 2021, 126(3):1867-1896.
doi: 10.1007/s11192-020-03842-6
[29] Preschitschek N, Niemann H, Leker J, et al. Anticipating Industry Convergence: Semantic Analyses vs IPC Co-Classification Analyses of Patents[J]. Foresight, 2013, 15(6):446-464.
doi: 10.1108/FS-10-2012-0075
[30] Kong D J, Yang J Z, Li L F. Early Identification of Technological Convergence in Numerical Control Machine Tool: A Deep Learning Approach[J]. Scientometrics, 2020, 125(3):1983-2009.
doi: 10.1007/s11192-020-03696-y
[31] Eilers K, Frischkorn J, Eppinger E, et al. Patent-Based Semantic Measurement of One-Way and Two-Way Technology Convergence: The Case of Ultraviolet Light Emitting Diodes (UV-LEDs)[J]. Technological Forecasting and Social Change, 2019, 140:341-353.
doi: 10.1016/j.techfore.2018.12.024
[32] Kim T S, Sohn S Y. Machine-Learning-Based Deep Semantic Analysis Approach for Forecasting New Technology Convergence[J]. Technological Forecasting and Social Change, 2020, 157:120095.
doi: 10.1016/j.techfore.2020.120095
[33] 韩卓洋. 专利价值评估体系方法综述及启示[J]. 环渤海经济瞭望, 2020(5):166.
[33] ( Han Zhuoyang. Review Summary and Enlightenment of Patent Value Evaluation System[J]. Economic Outlook the Bohai Sea, 2020(5):166.)
[34] Lanjouw J O, Schankerman M. Characteristics of Patent Litigation: A Window on Competition[J]. The RAND Journal of Economics, 2001, 32(1):129.
doi: 10.2307/2696401
[35] 李清海, 刘洋, 吴泗宗, 等. 专利价值评价指标概述及层次分析[J]. 科学学研究, 2007, 25(2):281-286.
[35] ( Li Qinghai, Liu Yang, Wu Sizong, et al. Patent Value Indicators and Their Structure[J]. Studies in Science of Science, 2007, 25(2):281-286.)
[36] 李春燕, 石荣. 专利质量指标评价探索[J]. 现代情报, 2008, 28(2):146-149.
[36] ( Li Chunyan, Shi Rong. Exploration on Patent Quality Index Evaluation[J]. Modern Information, 2008, 28(2):146-149.)
[37] Lee C Y, Cho Y, Seol H, et al. A Stochastic Patent Citation Analysis Approach to Assessing Future Technological Impacts[J]. Technological Forecasting and Social Change, 2012, 79(1):16-29.
doi: 10.1016/j.techfore.2011.06.009
[38] Fischer T, Leidinger J. Testing Patent Value Indicators on Directly Observed Patent Value—An Empirical Analysis of Ocean Tomo Patent Auctions[J]. Research Policy, 2014, 43(3):519-529.
doi: 10.1016/j.respol.2013.07.013
[39] Blackman M. EPO Patent Information Conference, Stockholm, Sweden, October, 2008[J]. World Patent Information, 2009, 31(2):152-154.
doi: 10.1016/j.wpi.2008.11.003
[40] Grimaldi M, Cricelli L, di Giovanni M, et al. The Patent Portfolio Value Analysis: A New Framework to Leverage Patent Information for Strategic Technology Planning[J]. Technological Forecasting and Social Change, 2015, 94:286-302.
doi: 10.1016/j.techfore.2014.10.013
[41] 赵蕴华, 张静, 李岩, 等. 基于机器学习的专利价值评估方法研究[J]. 情报科学, 2013, 31(12):15-18.
[41] ( Zhao Yunhua, Zhang Jing, Li Yan, et al. Study on Evaluation for Patent Value Based on Machine Learning[J]. Information Science, 2013, 31(12):15-18.)
[42] 邱一卉, 张驰雨, 陈水宣. 基于分类回归树算法的专利价值评估指标体系研究[J]. 厦门大学学报(自然科学版), 2017, 56(2):244-251.
[42] ( Qiu Yihui, Zhang Chiyu, Chen Shuixuan. Research of Patent-Value Assessment Indictor System Based on Classification and Regression Tree Algorithm[J]. Journal of Xiamen University (Natural Science), 2017, 56(2):244-251.)
[43] Ercan S, Kayakutlu G. Patent Value Analysis Using Support Vector Machines[J]. Soft Computing, 2014, 18(2):313-328.
doi: 10.1007/s00500-013-1059-x
[44] Bessen J E. Estimates of Patent Rents from Firm Market Value[J]. SSRN Electronic Journal, 2007: 1604-1616.
[45] Chen Y S, Chang K C. Using Neural Network to Analyze the Influence of the Patent Performance upon the Market Value of the US Pharmaceutical Companies[J]. Scientometrics, 2009, 80(3):637-655.
doi: 10.1007/s11192-009-2095-2
[1] 王若佳, 严承希, 郭凤英, 王继民. 基于用户画像的在线健康社区用户流失预测研究*[J]. 数据分析与知识发现, 2022, 6(2/3): 80-92.
[2] 车宏鑫,王桐,王伟. 前列腺癌预测模型对比研究*[J]. 数据分析与知识发现, 2021, 5(9): 107-114.
[3] 陈东华,赵红梅,尚小溥,张润彤. 数据驱动的大型医院手术室运营预测与优化方法研究*[J]. 数据分析与知识发现, 2021, 5(9): 115-128.
[4] 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.
[5] 刘渊晨, 王昊, 高亚琪. 在线音乐歌单播放量预测及影响因素分析*[J]. 数据分析与知识发现, 2021, 5(8): 100-112.
[6] 苏强, 侯校理, 邹妮. 基于机器学习组合优化方法的术后感染预测模型研究*[J]. 数据分析与知识发现, 2021, 5(8): 65-75.
[7] 徐良辰, 郭崇慧. 基于集成学习的胃癌生存预测模型研究*[J]. 数据分析与知识发现, 2021, 5(8): 86-99.
[8] 张鑫,文奕,许海云. 一种融合表示学习与主题表征的作者合作预测模型*[J]. 数据分析与知识发现, 2021, 5(3): 88-100.
[9] 徐雅斌, 孙秋天. 特定舆情的意见领袖挖掘和关键传播路径预测[J]. 数据分析与知识发现, 2021, 5(2): 32-42.
[10] 吴胜男, 田若楠, 蒲虹君, 梁雯琪, 张亚飞, 于琦, 贺培凤. 基于社交媒体的医药领域关联主题预测方法研究*[J]. 数据分析与知识发现, 2021, 5(12): 98-109.
[11] 吴胜男, 蒲虹君, 田若楠, 梁雯琪, 于琦. 网络结构对链路预测算法的影响研究*——基于元分析视角[J]. 数据分析与知识发现, 2021, 5(11): 102-113.
[12] 余传明, 张贞港, 孔令格. 面向链接预测的知识图谱表示模型对比研究*[J]. 数据分析与知识发现, 2021, 5(11): 29-44.
[13] 丁浩, 艾文华, 胡广伟, 李树青, 索炜. 融合用户兴趣波动时序的个性化推荐模型*[J]. 数据分析与知识发现, 2021, 5(11): 45-58.
[14] 蒋翠清,王香香,王钊. 基于消费者关注度的汽车销量预测方法研究*[J]. 数据分析与知识发现, 2021, 5(1): 128-139.
[15] 柴国荣,王斌,沙勇忠. 基于多机器学习方法联合的公共卫生风险预测研究——以兰州市流感预测为例*[J]. 数据分析与知识发现, 2021, 5(1): 90-98.
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