面向TRIZ理论使用者的多标签专利分类

摘要/Abstract

摘要： 专利是创新的结果,更是再创造的知识源泉,对专利技术知识依据创新需求的分类可有效帮助设计者进行创新设计。依据TRIZ理论对产品专利进行自动分类,以辅助利用专利蕴含的技术冲突进行产品创新设计。TRIZ原始的发明原理过于抽象以及有些原理之间有重叠,文中对40个原始的发明原理进行重组,形成20个新的类别。专利自动分类是一类典型的多标签分类问题,文中从Pro_Techniques和CREAX两个软件中收集了针对发明原理进行具体解释的专利数据,并依据此数据集对问题转换和自适应算法两类多标签分类算法进行对比分析。采用海明损失、测度等评估特性评估了上述算法的性能和质量。结果表明,在使用TRIZ专利数据集时,问题转换方法分类性能要明显优于自适应算法。

关键词: 专利分类,发明原理,TRIZ,多标签

Abstract: Patent is not only the results but also the resources of products innovation．It can help the designer make innovation effectively,if we classify technique knowledge of patent based on innovative demand．The classification of products patent based on the TRIZ can assistant in using technique contradiction addressed in patent to make innovative design．The original Inventive Principles is so abstract that some principles are overlapped.Paper analyzed the 40IPs and grouped them into new 20classes．Patent classification problem is known as multi-label classification problem．Pro-Techniques,CREAX,these two softwares supply patents which explain Inventive Principles in detail．The dataset is used to compare the performance of multi-label classification algorithm:problem transformation and algorithm adaptation．Several measures such as hamming loss,F-measure have been proposed in the literature for the evaluation of multilable classifiers．The result shows Problem Transformation performs more excellent than Algorithm Adaptation using TRIZ patent datasets.

Key words: Patent classification,Inventive principle,TRIZ,Multi-label

袁力,陈阳,赵勇. 面向TRIZ理论使用者的多标签专利分类[J]. 计算机科学, 2013, 40(Z11): 255-258. https://doi.org/

YUAN Li,CHEN Yang and ZHAO Yong. Multi-label Patent Classification Oriented to TRIZ Users[J]. Computer Science, 2013, 40(Z11): 255-258. https://doi.org/

参考文献

[1] 左晶．IPC和USC分类体系下专利检索的对比分析[J]．现代情报,2007,1:130-132
[2] Meyer D．Support Vector Machines[J]．The Interface to libsvm in package e1071．e1071Vignette,2012
[3] Chou K C,Shen H B．Predicting eukaryotic protein subcellular location by fusing optimized evidence-theoretic K-nearest neighbor classifiers[J]．Journal of Proteome Research,2006,5(8):1888-1897
[4] 卢长林．手扶电动整枝机[D].1992
[5] Elisseeff A,Weston J．A kernel method for multi-labelled classification[C]∥Advances in neural information processing systems．2001:681-687
[6] Godbole S,Sarawagi S．Discriminative methods for multi-labeled classification[M]∥Advances in Knowledge Discovery and Data Mining．Berlin Heidelberg:Springer,2004:22-30
[7] Crammer K,Singer Y．A family of additive online algorithms for category ranking[J]．The Journal of Machine Learning Research,2003,3:1025-1058
[8] Zhang M L,Zhou Z H．Multilabel neural networks with applications to functional genomics and text categorization[J]．IEEE Transactions on Knowledge and Data Engineering,2006,18(10):1338-1351
[9] Zhang M L,Zhou Z H．A k-nearest neighbor based algorithm for multi-label classification[C]∥Granular Computing,2005IEEE International Conference on．IEEE,2005,2:718-721
[10] Tsoumakas G,Dimou A,Spyromitros E,et al．Correlation-based pruning of stacked binary relevance models for multi-label lear-ning[C]∥Proceeding of ECML/PKDD 2009Workshop on Learning from Multi-Label Data．Bled,Slovenia,2009:101-116
[11] Tsoumakas G,Katakis I,Vlahavas I．Mining multi-label data[M]．Data mining and knowledge discovery handbook．US:Springer,2010:667-685
[12] Read J．A pruned problem transformation method for multi-label classification[C]∥Proc．2008New Zealand Computer Science Research Student Conference (NZCSRS 2008)．2008:143-150
[13] Zhang M L,Zhou Z H．A k-nearest neighbor based algorithm for multi-label classification[C]∥Granular Computing,2005IEEE International Conference on．IEEE,2005,2:718-721
[14] Gao S,Wu W,Lee C H,et al．A MFoM learning approach to robust multiclass multi-label text categorization[C]∥Proceedings of the twenty-first international conference on Machine lear-ning．ACM,2004:42
[15] Williams T,Domb E．Reversability of the 40Principles of Problem Solving[J]．The TRIZ Journal,May 1998
[16] Cong H,Tong L H．Grouping of TRIZ Inventive Principles tofacilitate automatic patent classification[J]．Expert Systems with Applications,2008,34(1):788-795
[17] Pro_Techniques．http://www.iwint.com.cn/
[18] CREAX http://www.creax.com
[19] 费洪晓,康松林,朱小娟,等．基于词频统计的中文分词的研究[J]．计算机工程与应用,2005,41(7):67-68
[20] 王素格,魏英杰．停用词表对中文文本情感分类的影响[J]．情报学报,2008,27(2):175-179
[21] Erk K,Padó S．A structured vector space model for word meaning in context[C]∥Proceedings of the Conference on Empirical Methods in Natural Language Processing．Association for Computational Linguistics,2008:897-906
[22] 任永功,杨荣杰,尹明飞,等．基于信息增益的文本特征选择方法[J]．计算机科学,2012,39(11):127-130
[23] Tsoumakas G,Vlahavas I．Random k-labelsets:An ensemblemethod for multilabel classification[C]∥Machine Learning:ECML 2007．Berlin Heidelberg:Springer,2007:406-417
[24] Yang Y,Liu X．A re-examination of text categorization methods[C]∥Proceedings of the 22nd annual international ACM SIGIR conference on Research and development in information retrie-val．ACM,1999:42-49
[25] Tsoumakas G,Xioufis E S,Vilcek J,et al．MULAN:A Java Library for Multi-Label Learning[J]．Journal of Machine Learning Research,2011,12(7):2411-2414
[26] Hall M,Frank E,Holmes G,et al．The WEKA data mining soft-ware:an update[J]．ACM SIGKDD Explorations Newsletter,2009,11(1):10-18

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed