融合多维信息的Web服务表征方法

doi:10.3778/j.issn.1673-9418.2101024

计算机科学与探索 ›› 2022, Vol. 16 ›› Issue (7): 1561-1569.DOI: 10.3778/j.issn.1673-9418.2101024

融合多维信息的Web服务表征方法

张祥平¹^,², 刘建勋¹^,²^,⁺(), 肖巧翔², 曹步清¹^,²

1.湖南科技大学服务计算与软件服务新技术湖南省重点实验室,湖南湘潭 411201
2.湖南科技大学计算机科学与工程学院,湖南湘潭 411201

收稿日期:2021-01-07 修回日期:2021-03-08 出版日期:2022-07-01 发布日期:2021-03-25
作者简介:张祥平（1993—）,男,福建三明人,博士研究生,主要研究方向为服务计算、服务推荐。
ZHANG Xiangping, born in 1993, Ph.D. candidate. His research interests include service computing and service recommendation.
刘建勋（1970—）,男,湖南衡阳人,博士,教授,主要研究方向为服务计算、云计算。
LIU Jianxun, born in 1970, Ph.D., professor. His research interests include service computing and cloud computing.
肖巧翔（1993—）,女,湖南衡阳人,硕士,主要研究方向为服务聚类。
XIAO Qiaoxiang, born in 1993, M.S. Her research interest is service clustering.
曹步清（1979—）,男,湖南湘潭人,博士,教授,主要研究方向为面向服务的软件工程、服务计算、云计算。
CAO Buqing, born in 1979, Ph.D., professor. His research interests include service-oriented soft-ware engineering, service computing and cloud computing.
基金资助:
国家重点研发计划(2020YFB1707602)

Multidimensional Information-Based Web Service Representation Method

ZHANG Xiangping¹^,², LIU Jianxun¹^,²^,⁺(), XIAO Qiaoxiang², CAO Buqing¹^,²

1. Hunan Key Lab for Services Computing and Novel Software Technology, Hunan University of Science and Tech-nology, Xiangtan, Hunan 411201, China
2. School of Computer Science and Engineering, Hunan University of Science and Technology, Xiangtan, Hunan 411201, China

Received:2021-01-07 Revised:2021-03-08 Online:2022-07-01 Published:2021-03-25
Supported by:
the National Key Research and Development Program of China(2020YFB1707602)

摘要/Abstract

摘要：

随着面向服务体系结构（SOA）技术的发展,Web服务的数量增长迅速。正确高效地对Web服务进行聚类或分类,能够有效地提高服务发现质量以及促进服务组合效率。然而,现有的Web服务建模方法（如LDA主题模型）难以从稀疏的Web服务数据中获得精确有效的信息用于Web服务聚类。针对这个问题,提出了一种融合多维信息的Web服务表征方法（MISR）。首先,将高斯混合模型和Word2Vec算法相结合生成包含Web服务功能主题信息和语义信息的词向量表征。然后,抽取出Web服务中包含的标签-词汇信息、流行度以及Web服务共现信息,结合前一步生成的向量生成包含多维信息的Web服务表征向量。最后,在Web服务聚类和Web服务分类两个任务上对MISR方法的有效性进行验证。在真实数据集上进行Web API服务聚类实验,实验结果表明,相比于LDA、Word2Vec、Doc2Vec、WT-LDA、HDP-SOM、GWSC,提出的方法在Micro-F1值上有38.8%、54.5%、15.3%、33.3%、44.7%、9.7%的提升。

关键词: Web服务表征, 高斯混合模型, Word2Vec, Web服务聚类, Web服务分类

Abstract:

With the development of service-oriented architecture (SOA) technology, the amount of Web service is increasing. Clustering or classifying Web services correctly are an effective way to improve the quality of Web service discovery and the efficiency of Web service composition. However, the existing Web service modeling methods (such as latent Dirichlet allocation topic model) are difficult to obtain accurate and effective Web service representation from a sparse Web service dataset for Web service clustering. To solve this problem, this paper proposes a multi-dimensional information-based Web service representation method (MISR). First, it generates word vectors which contain topic and semantic information implicit in Web service description with Gaussian mixture model and Word2Vec. Then, the MISR algorithm combines tag-word relationship, popularity, and co-occurrence information together for generating multi-dimensional information Web service representation. Web service clustering and Web service classification are used for evaluating the effectiveness of MISR. Based on a real-world dataset of API services, the experiment results show that compared with LDA, Word2Vec, Doc2Vec, WT-LDA, HDP-SOM, GWSC, the proposed method has 38.8%, 54.5%, 15.3%, 33.3%, 44.7%, 9.7% improvement in Micro-F1 value.

Key words: Web service representation, Gaussian mixture model, Word2Vec, Web service clustering, Web service classification

中图分类号:

TP301

张祥平, 刘建勋, 肖巧翔, 曹步清. 融合多维信息的Web服务表征方法[J]. 计算机科学与探索, 2022, 16(7): 1561-1569.

ZHANG Xiangping, LIU Jianxun, XIAO Qiaoxiang, CAO Buqing. Multidimensional Information-Based Web Service Representation Method[J]. Journal of Frontiers of Computer Science and Technology, 2022, 16(7): 1561-1569.

图/表 8

表1 符号说明

Table 1 Description of notations

符号	对应含义
$w i$	第 $i$ 个单词
$w v i$	第 $i$ 个单词对应的词向量
$c k$	高斯混合模型划分出的第 $k$ 个类簇（主题类）
$w v c i$	使用主题信息增强的第 $i$ 个单词对应的词向量
$t j$	第 $j$ 个标签
$w v t i$	使用标签信息增强的第 $i$ 个单词对应的词向量
$a w v i$	融合主题信息和标签信息的第 $i$ 个单词对应的词向量
$p o p (a i)$	第 $i$ 个API的流行度
$C o$	API共现矩阵
$s v d$	第 $d$ 个Web服务的表征向量（不含附加信息）
$s d$	第 $d$ 个Web服务的表征向量（包含附加信息）

表1 符号说明

Table 1 Description of notations

符号	对应含义
$w i$	第 $i$ 个单词
$w v i$	第 $i$ 个单词对应的词向量
$c k$	高斯混合模型划分出的第 $k$ 个类簇（主题类）
$w v c i$	使用主题信息增强的第 $i$ 个单词对应的词向量
$t j$	第 $j$ 个标签
$w v t i$	使用标签信息增强的第 $i$ 个单词对应的词向量
$a w v i$	融合主题信息和标签信息的第 $i$ 个单词对应的词向量
$p o p (a i)$	第 $i$ 个API的流行度
$C o$	API共现矩阵
$s v d$	第 $d$ 个Web服务的表征向量（不含附加信息）
$s d$	第 $d$ 个Web服务的表征向量（包含附加信息）

图1 主题信息增强的词向量生成过程

Fig.1 Generation process of topic-augmented word embedding

图2 标签信息增强的词向量生成过程

Fig.2 Generation process of tag-augmented word embedding

表2 API服务聚类结果对比

Table 2 Results of API services clustering

对比方法	准确率	召回率	Micro-F1
LDA	0.48	0.52	0.49
Word2Vec	0.44	0.44	0.44
Doc2Vec	0.60	0.59	0.59
WT-LDA	0.49	0.54	0.51
HDP-SOM	0.48	0.47	0.47
GWSC	0.64	0.61	0.62
MISR	0.68	0.68	0.68

表3 Mashup服务聚类结果对比

Table 3 Results of Mashup services clustering

对比方法	准确率	召回率	Micro-F1
LDA	0.43	0.42	0.41
Word2Vec	0.44	0.41	0.42
Doc2Vec	0.57	0.63	0.60
WT-LDA	0.49	0.54	0.52
HDP-SOM	0.46	0.45	0.45
GWSC	0.64	0.62	0.63
MISR	0.69	0.73	0.71

表4 API服务分类结果对比

Table 4 Results of API services classification

对比方法	准确率	召回率	Micro-F1
LDA	0.61	0.61	0.61
Word2Vec	0.71	0.72	0.71
Doc2Vec	0.64	0.61	0.62
WT-LDA	0.60	0.56	0.58
HDP-SOM	0.58	0.56	0.57
GWSC	0.75	0.73	0.74
MISR	0.85	0.86	0.85

表5 Mashup服务分类结果对比

Table 5 Results of Mashup services classification

对比方法	准确率	召回率	Micro-F1
LDA	0.50	0.43	0.46
Word2Vec	0.53	0.49	0.50
Doc2Vec	0.56	0.48	0.52
WT-LDA	0.53	0.60	0.56
HDP-SOM	0.55	0.49	0.52
GWSC	0.74	0.72	0.73
MISR	0.87	0.87	0.87

图3 Mashup服务分类结果对比

Fig.3 Results of Mashup services classification

参考文献 19

[1]	曹步清, 肖巧翔, 张祥平, 等. 融合SOM功能聚类与Deep-FM质量预测的API服务推荐方法[J]. 计算机学报, 2019, 42(6): 1367-1383.
	CAO B Q, XIAO Q X, ZHANG X P, et al. An API service recommendation method via combining self-organization map-based functionality clustering and deep factorization machine-based quality prediction[J]. Chinese Journal of Computers, 2019, 42(6): 1367-1383.
[2]	LIU X M, AGARWAL S, DING C, et al. An LDA-SVM ac-tive learning framework for Web service classification[C]// Proceedings of the 2016 IEEE International Conference on Web Services, San Francisco, Jun 27-Jul 2, 2016. Washing-ton: IEEE Computer Society, 2016: 49-56.
[3]	CHEN J J, CAO B Q, CAO Y C, et al. A mobile application classification method with enhanced topic attention me-chanism[C]// LNCS 1042: Proceedings of the 14th CCF Conference on Computer Supported Cooperative Work and Social Compting, Kunming, Aug 16-18, 2019. Cham: Springer, 2019: 683-695.
[4]	CHEN L, WANG Y L, YU Q, et al. WT-LDA: user tagging augmented LDA for Web service clustering[C]// LNCS 8274: Proceedings of the 11th International Conference on Service-Oriented Computing, Berlin, Dec 2-5, 2013. Berlin, Heidel-berg: Springer, 2013: 162-176.
[5]	SHI M, LIU J X, CAO B Q, et al. A prior knowledge based approach to improving accuracy of Web services clustering[C]// Proceedings of the 15th International Conference on Service Computing, San Francisco, Jul 2-7, 2018. Piscata-way: IEEE, 2018: 1-8.
[6]	XIAO Q X, CAO B Q, ZHANG X P, et al. Web services clus-tering based on HDP and SOM neural network[C]// Pro-ceedings of the 15th IEEE International Conference on Ubiquitous Intelligence & Computing, Guangzhou, Oct 8-12, 2018. Piscataway: IEEE, 2018: 397-404.
[7]	CAO Y C, LIU J X, SHI M, et al. Relationship network augmented Web services clustering[C]// Proceedings of the 2019 IEEE International Conference on Web Services, Mi-lan, Jul 8-13, 2019. Piscataway: IEEE, 2019: 247-254.
[8]	XIA B F, FAN Y S, TAN W, et al. Category-aware API clus-tering and distributed recommendation for automatic mashup creation[J]. IEEE Transactions on Services Computing, 2015, 8(5): 674-687. DOI URL
[9]	MICHAEL R T F, SIVA P, BALAKRISHNAN R, et al. QoS based classification using K-nearest neighbor algorithm for effective Web service selection[C]// Proceedings of the 2015 IEEE International Conference on Electrical, Com-puter and Communication Technologies, Coimbatore, Mar 5-7, 2015. Piscataway: IEEE, 2015: 1-4.
[10]	李征, 王健, 张能. 一种面向主题的领域服务聚类方法[J]. 计算机研究与发展, 2014, 51(2): 408-419.
	LI Z, WANG J, ZHAGN N. A topic-oriented clustering app-roach for domain services[J]. Journal of Computer Research and Development, 2014, 51(2): 408-419.
[11]	CAO B Q, LIU X Q, LI J, et al. Mashup service clustering based on an integration of service content and network via exploiting a two-level topic model[C]// Proceedings of the 2016 IEEE International Conference on Web Services, San Francisco, Jun 27-Jul 2, 2016. Washington: IEEE Compu-ter Society, 2016: 212-219.
[12]	SHI M, LIU J X, ZHOU D, et al. WE-LDA: a word em-beddings augmented LDA model for Web services clus-tering[C]// Proceedings of the 2017 IEEE International Con-ference on Web Services, Honolulu, Jun 25-30, 2017. Pisca-taway: IEEE, 2017: 9-16.
[13]	CAO Y C, LIU J X, CAO B Q, et al. Web services classi-fication with topical attention based Bi-LSTM[C]// Procee-dings of the 15th EAI International Conference on Collabo-rative Computing:Networking, Applications and Work-sharing, London, Aug 19-22, 2019. Cham: Springer, 2019: 394-407.
[14]	肖勇, 刘建勋, 胡蓉, 等. 基于GAT2VEC的Web服务分类方法[J]. 软件学报, 2021, 32(12): 3751-3767.
	XIAO Y, LIU J X, HU R, et al. GAT2VEC-based Web service classification method[J]. Journal of Software, 2021, 32(12): 3751-3767.
[15]	MIKOLOV T, CHEN K, CORRADO G, et al. Efficient esti-mation of word representations in vector space[C]// Procee-dings of the 1st International Conference on Learning Rep-resentations, Scottsdale, May 2-4, 2013: 1-15.
[16]	REYNOLDS D. Gaussian mixture models[M]// Encyclopedia of Biometrics. Berlin: Springer, 2009: 659-663.
[17]	DASGUPTA S. Experiments with random projection[C]// Pro-ceedings of the 16th Conference on Uncertainty in A.pngi-cial Intelligence, Stanford, Jun 30-Jul 3, 2000. San Fran-cisco: Morgan Kaufmann Publishers Inc., 2000: 143-151.
[18]	BLEI D, NG A, JORDAN M. Latent Dirichlet allocation[J]. Journal of Machine Learning Research, 2003, 3(4/5): 993-1022.
[19]	LE Q V, MIKOLOV T. Distributed representations of sen-tences and documents[C]// Proceedings of the 31st Interna-tional Conference on Machine Learning, Beijing, Jun 21-26, 2014: 1188-1196.

融合多维信息的Web服务表征方法

Multidimensional Information-Based Web Service Representation Method

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 8

参考文献 19

相关文章 5

编辑推荐

Metrics

[1]	朱壮壮, 周治平. 高斯混合生成模型检测健康数据异常[J]. 计算机科学与探索, 2022, 16(5): 1128-1135.
[2]	程琪芩，万良. BiLSTM在跨站脚本检测中的应用研究[J]. 计算机科学与探索, 2020, 14(8): 1338-1347.
[3]	张墨华，彭建华. 面向图像复原的分层贝叶斯局部高斯混合模型[J]. 计算机科学与探索, 2020, 14(2): 325-335.
[4]	李聪，葛洪伟. 非线性幂变换Gammachirp滤波器的鲁棒语音特征提取[J]. 计算机科学与探索, 2019, 13(8): 1351-1359.
[5]	李峰，侯加英，曾荣仁，凌晨. 融合词向量的多特征句子相似度计算方法研究[J]. 计算机科学与探索, 2017, 11(4): 608-618.