新兴技术识别中的不均衡分类研究

doi:10.3772/j.issn.1000-0135.2022.10.006

情报学报

2022, Vol. 41

Issue (10): 1059-1070 DOI: 10.3772/j.issn.1000-0135.2022.10.006

情报技术与应用

本期目录 | 过刊浏览 | 高级检索

新兴技术识别中的不均衡分类研究

卢小宾, 张杨燚, 杨冠灿, 行佳鑫

中国人民大学信息资源管理学院，北京 100872

Imbalanced Classification of Emerging Technologies Identification: Based on Cost-sensitive Random Forest

Lu Xiaobin, Zhang Yangyi, Yang Guancan, Xing Jiaxin

School of Information Resource Management, Renmin University of China, Beijing 100872

摘要
图/表
参考文献
相关文章 (4)

全文: PDF (1648 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要基于大规模专利数据和专利特征指标开展自动化的前瞻性预测已逐渐成为新兴技术识别的研究重点，机器学习方法的引入也让海量技术发明涌现为新兴技术这一小概率事件是一种典型的不均衡分类问题的本质受到关注。本研究目标在于通过优化分类策略改善新兴技术识别中不均衡数据集造成的分类偏向多数类别的识别效果，提出了综合数据、算法和评估三个层面的新兴技术识别不均衡分类优化框架，并以预测癌症药物领域专利是否有成为新兴技术潜质的二分类场景为例开展实证分析。具体改进之处在于：数据层面采纳渐进式重采样思路；算法层面构建代价敏感的随机森林；评估层面引入代价敏感思想，探究在缺乏专家经验时的代价矩阵验证方式。研究结果表明，基于1∶2均衡比例随机欠采样、以ROC-Youden指数阈值代价矩阵构建的代价敏感随机森林在对应的新兴技术识别目标中能正确预测出82.8%的新兴技术和81.6%的普通技术，显著优于本文对照组及现有相关成果，对未来深入挖掘新兴技术识别中不均衡分类问题的本质具有参考价值。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	卢小宾
	张杨燚
	杨冠灿
	行佳鑫

关键词 ：新兴技术识别, 不均衡分类, 代价敏感, 随机森林, 渐进式重采样

收稿日期: 2021-08-16

基金资助:国家社会科学基金重点项目“新时期产业技术情报分析方法体系研究”（21ATQ008）。

作者简介: 卢小宾，男，1963年生，博士，教授，博士生导师，研究方向为信息分析；张杨燚，女，1997年生，硕士，研究方向为专利数据挖掘；杨冠灿，男，1981年生，博士，副教授，研究方向为技术竞争情报，E-mail：yanggc@ruc.edu.cn；行佳鑫，女，1999年生，硕士，研究方向为技术竞争情报；

引用本文:

卢小宾, 张杨燚, 杨冠灿, 行佳鑫. 新兴技术识别中的不均衡分类研究[J]. 情报学报, 2022, 41(10): 1059-1070.
Lu Xiaobin, Zhang Yangyi, Yang Guancan, Xing Jiaxin. Imbalanced Classification of Emerging Technologies Identification: Based on Cost-sensitive Random Forest. 情报学报, 2022, 41(10): 1059-1070.

链接本文:

https://qbxb.istic.ac.cn/CN/10.3772/j.issn.1000-0135.2022.10.006 或 https://qbxb.istic.ac.cn/CN/Y2022/V41/I10/1059

1 乔治·戴, 保罗·休梅克. 沃顿论新兴技术管理[M]. 石莹, 等译. 北京: 华夏出版社, 2002.
2 Lee C Y, Kim J, Kwon O, et al. Stochastic technology life cycle analysis using multiple patent indicators[J]. Technological Forecasting and Social Change, 2016, 106: 53-64.
3 周源, 刘宇飞, 薛澜. 一种基于机器学习的新兴技术识别方法: 以机器人技术为例[J]. 情报学报, 2018, 37(9): 939-955.
4 Porter A L, Garner J, Carley S F, et al. Emergence scoring to identify frontier R&D topics and key players[J]. Technological Forecasting and Social Change, 2019, 146: 628-643.
5 Cobo M J, López-Herrera A G, Herrera-Viedma E, et al. An approach for detecting, quantifying, and visualizing the evolution of a research field: a practical application to the Fuzzy Sets Theory field[J]. Journal of Informetrics, 2011, 5(1): 146-166.
6 Leone Sciabolazza V, Vacca R, Kennelly Okraku T, et al. Detecting and analyzing research communities in longitudinal scientific networks[J]. PLoS One, 2017, 12(8): e0182516.
7 Choi S, Yoon J, Kim K, et al. SAO network analysis of patents for technology trends identification: a case study of polymer electrolyte membrane technology in proton exchange membrane fuel cells[J]. Scientometrics, 2011, 88(3): 863-883.
8 Verhoeven D, Bakker J, Veugelers R. Measuring technological novelty with patent-based indicators[J]. Research Policy, 2016, 45(3): 707-723.
9 卢小宾, 杨冠灿, 徐硕, 等. 计量与演化视角下的新兴技术识别研究进展评述[J]. 情报学报, 2020, 39(6): 651-661.
10 Longadge R, Dongre S S. Class imbalance problem in data mining: review[J]. International Journal of Computer Science and Network, 2013, 2(1): 83-87.
11 周志华. 机器学习[M]. 北京: 清华大学出版社, 2016.
12 叶志飞, 文益民, 吕宝粮. 不平衡分类问题研究综述[J]. 智能系统学报, 2009, 4(2): 148-156.
13 Japkowicz N, Stephen S. The class imbalance problem: a systematic study[J]. Intelligent Data Analysis, 2002, 6(5): 429-449.
14 Weiss G M, Hirsh H. A quantitative study of small disjuncts[C/OL]// AAAI-00 Proceedings. Palo Alto: AAAI Press, 2000. https://www.aaai.org/Papers/AAAI/2000/AAAI00-102.pdf.
15 向鸿鑫, 杨云. 不平衡数据挖掘方法综述[J]. 计算机工程与应用, 2019, 55(4): 1-16.
16 Menardi G, Torelli N. Training and assessing classification rules with imbalanced data[J]. Data Mining and Knowledge Discovery, 2014, 28(1): 92-122.
17 Chawla N V, Bowyer K W, Hall L O, et al. SMOTE: synthetic minority over-sampling technique[J]. Journal of Artificial Intelligence Research, 2002, 16: 321-357.
18 Han H, Wang W Y, Mao B H. Borderline-SMOTE: a new over-sampling method in imbalanced data sets learning[C]// Proceedings of the International Conference on Intelligent Computing. Heidelberg: Springer, 2005: 878-887.
19 He H B, Bai Y, Garcia E A, et al. ADASYN: adaptive synthetic sampling approach for imbalanced learning[C]// Proceedings of the 2008 IEEE International Joint Conference on Neural Networks. IEEE, 2008: 1322-1328.
20 Prusa J, Khoshgoftaar T M, Dittman D J, et al. Using random undersampling to alleviate class imbalance on tweet sentiment data[C]// Proceedings of the 2015 IEEE International Conference on Information Reuse and Integration. IEEE, 2015: 197-202.
21 Yuan J H, Li J M, Zhang B. Learning concepts from large scale imbalanced data sets using support cluster machines[C]// Proceedings of the 14th ACM International Conference on Multimedia. New York: ACM Press, 2006: 441-450.
22 Zhang J P, Mani I. KNN approach to unbalanced data distributions: a case study involving information extraction[C]// Proceedings of the ICML'2003 Workshop on Learning from Imbalanced Datasets. ICML, 2003: 126.
23 Tomek I. Two modifications of CNN[J]. IEEE Transactions on Systems, Man, and Cybernetics, 1976, SMC-6(11): 769-772.
24 Ali-Gombe A, Elyan E. MFC-GAN: class-imbalanced dataset classification using Multiple Fake Class Generative Adversarial Network[J]. Neurocomputing, 2019, 361: 212-221.
25 Douzas G, Bacao F. Effective data generation for imbalanced learning using conditional generative adversarial networks[J]. Expert Systems With Applications, 2018, 91: 464-471.
26 Zhou Z H, Liu X Y. Training cost-sensitive neural networks with methods addressing the class imbalance problem[J]. IEEE Transactions on Knowledge and Data Engineering, 2006, 18(1): 63-77.
27 李勇, 刘战东, 张海军. 不平衡数据的集成分类算法综述[J]. 计算机应用研究, 2014, 31(5): 1287-1291.
28 刘定祥, 乔少杰, 张永清, 等. 不平衡分类的数据采样方法综述[J]. 重庆理工大学学报(自然科学), 2019, 33(7): 102-112.
29 Lin E L, Chen Q, Qi X M. Deep reinforcement learning for imbalanced classification[J]. Applied Intelligence, 2020, 50(8): 2488-2502.
30 Shi M, Tang Y F, Zhu X Q, et al. Multi-class imbalanced graph convolutional network learning[C]// Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence. California: International Joint Conferences on Artificial Intelligence Organization, 2020: 2879-2885.
31 Ghorbani M, Kazi A, Soleymani Baghshah M, et al. RA-GCN: graph convolutional network for disease prediction problems with imbalanced data[J]. Medical Image Analysis, 2022, 75: 102272.
32 Liu W K, Zhang H, Ding Z Y, et al. A comprehensive active learning method for multiclass imbalanced data streams with concept drift[J]. Knowledge-Based Systems, 2021, 215: 106778.
33 Kang X, Shi X F, Wu Y N, et al. Active learning with complementary sampling for instructing class-biased multi-label text emotion classification[J/OL]. IEEE Transactions on Affective Computing, (2020-11-16). http://dx.doi.org/10.1109/TAFFC.2020.3038401.
34 Wu X J, Meng S F. E-commerce customer churn prediction based on improved SMOTE and AdaBoost[C]// Proceedings of the 2016 13th International Conference on Service Systems and Service Management. IEEE, 2016: 1-5.
35 游子莹. 不均衡样本的分类优化方法[D]. 武汉: 华中科技大学, 2018.
36 孙炜. 基于代价敏感的改进AdaBoost算法在不平衡数据中的应用[D]. 广州: 暨南大学, 2018.
37 翟夕阳, 王晓丹, 李睿, 等. 采用多类代价指数损失函数的代价敏感AdaBoost算法[J]. 西安交通大学学报, 2017, 51(8): 33-39.
38 王学玲, 王建林. 基于代价敏感的AdaBoost算法改进[J]. 计算机应用与软件, 2013, 30(10): 123-125, 138.
39 Zhou Z H. Cost-sensitive learning[C]// Proceedings of the International Conference on Modeling Decisions for Artificial Intelligence. Heidelberg: Springer, 2011: 17-18.
40 平瑞, 周水生, 李冬. 高度不平衡数据的代价敏感随机森林分类算法[J]. 模式识别与人工智能, 2020, 33(3): 249-257.
41 Fan W, Stolfo S J, Zhang J X, et al. AdaCost: misclassification cost-sensitive boosting[C]// Proceedings of the Sixteenth International Conference on Machine Learning. San Francisco: Morgan Kaufmann Publishers, 1999: 97-105.
42 Domingos P. MetaCost: a general method for making classifiers cost-sensitive[C]// Proceedings of the Fifth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: ACM Press, 1999: 155-164.
43 Knoll U, Nakhaeizadeh G, Tausend B. Cost-sensitive pruning of decision trees[C]// Proceedings of the European Conference on Machine Learning. Heidelberg: Springer, 1994: 383-386.
44 Rotolo D, Hicks D, Martin B R. What is an emerging technology?[J]. Research Policy, 2015, 44(10): 1827-1843.
45 Trajtenberg M. A penny for your quotes: patent citations and the value of innovations[J]. The RAND Journal of Economics, 1990, 21(1): 172-187.
46 Tong X S, Frame J D. Measuring national technological performance with patent claims data[J]. Research Policy, 1994, 23(2): 133-141.
47 Breitzman A, Thomas P. Inventor team size as a predictor of the future citation impact of patents[J]. Scientometrics, 2015, 103(2): 631-647.
48 Lanjouw J O, Schankerman M. Characteristics of patent litigation: a window on competition[J]. The RAND Journal of Economics, 2001, 32(1): 129-151.
49 Reitzig M. Improving patent valuations for management purposes—validating new indicators by analyzing application rationales[J]. Research Policy, 2004, 33(6/7): 939-957.
50 Park H, Yoon J. Assessing coreness and intermediarity of technology sectors using patent co-classification analysis: the case of Korean national R&D[J]. Scientometrics, 2014, 98(2): 853-890.
51 Sternitzke C. The international preliminary examination of patent applications filed under the Patent Cooperation Treaty—a proxy for patent value?[J]. Scientometrics, 2009, 78(2): 189-202.
52 Trajtenberg M, Henderson R, Jaffe A. University versus corporate patents: a window on the basicness of invention[J]. Economics of Innovation and New Technology, 1997, 5(1): 19-50.
53 Narin F, Noma E, Perry R. Patents as indicators of corporate technological strength[J]. Research Policy, 1987, 16(2-4): 143-155.
54 Arts S, Appio F P, Looy B. Inventions shaping technological trajectories: do existing patent indicators provide a comprehensive picture?[J]. Scientometrics, 2013, 97(2): 397-419.
55 Breiman L. Random forests[J]. Machine Learning, 2001, 45: 5-32.
56 尹华, 胡玉平. 一种代价敏感随机森林算法[J]. 武汉大学学报(工学版), 2014, 47(5): 707-711.
57 Chen C, Liaw A, Breiman L. Using random forest to learn imbalanced data[R]. Berkeley: University of California, 2004: Report No.666.
58 Fawcett T. An introduction to ROC analysis[J]. Pattern Recognition Letters, 2006, 27(8): 861-874.
59 赵永彬, 陈硕, 刘明, 等. 基于置信度代价敏感的支持向量机不均衡数据学习[J]. 计算机工程, 2015, 41(10): 177-180, 185.
60 Korkmaz S, ?ahman M A, Cinar A C, et al. Boosting the oversampling methods based on differential evolution strategies for imbalanced learning[J]. Applied Soft Computing, 2021, 112: 107787.
61 Akbani R, Kwek S, Japkowicz N. Applying support vector machines to imbalanced datasets[C]// Proceedings of the 15th European Conference on Machine Learning. Heidelberg: Springer, 2004: 39-50.
62 Mullick S S, Datta S, Dhekane S G, et al. Appropriateness of performance indices for imbalanced data classification: an analysis[J]. Pattern Recognition, 2020, 102: 107197.
63 Kim J, Kim J. The impact of imbalanced training data on machine learning for author name disambiguation[J]. Scientometrics, 2018, 117(1): 511-526.
64 Peng Y C, Li C Y, Wang K, et al. Examining imbalanced classification algorithms in predicting real-time traffic crash risk[J]. Accident Analysis & Prevention, 2020, 144: 105610.
65 Liu X. Classification accuracy and cut point selection[J]. Statistics in Medicine, 2012, 31(23): 2676-2686.
66 Maloof M A. Learning when data sets are imbalanced and when costs are unequal and unknown[C]// Proceedings of the ICML’03 Workshop on Learning from Imbalanced Data Sets II. Washington DC: ICML, 2003: 328-334.
67 Abdel-Aty M, Uddin N, Pande A. Split models for predicting multivehicle crashes during high-speed and low-speed operating conditions on freeways[J]. Transportation Research Record: Journal of the Transportation Research Board, 2005, 1908(1): 51-58.