Service discovery method for Internet of Things based on Biterm topic model

doi:10.11772/j.issn.1001-9081.2019091662

Journal of Computer Applications ›› 2020, Vol. 40 ›› Issue (2): 459-464.DOI: 10.11772/j.issn.1001-9081.2019091662

• CCF NDBC 2019 • Previous Articles Next Articles

Service discovery method for Internet of Things based on Biterm topic model

Shuman WANG, Aiping LI(), Liguo DUAN, Jia FU, Yongle CHEN

College of Information and Computer，Taiyuan University of Technology，Taiyuan Shanxi 030024，China

Received:2019-08-12 Revised:2019-09-29 Accepted:2019-10-24 Online:2019-10-31 Published:2020-02-10
Contact: Aiping LI
About author:WANG Shuman， born in 1994， M. S. candidate. Her research interests include service computing， information processing.
DUAN Liguo， born in 1970， Ph. D.， associate professor. His research interests include Chinese information processing， knowledge mapping.
FU Jia， born in 1995， M. S. candidate. Her research interests include service computing， information processing.
CHEN Yongle， born in 1983， Ph. D.， associate professor. His research interests include wireless sensor network， IoT security.
Supported by:
the National Key R&D Programe of China （Rsearch and Development Program — "cyberspace security"(2018YFB0803402)

基于BTM的物联网服务发现方法

王舒漫, 李爱萍(), 段利国, 付佳, 陈永乐

太原理工大学信息与计算机学院，太原 030024

通讯作者: 李爱萍
作者简介:王舒漫（1994—），女，山西长治人，硕士研究生，主要研究方向：服务计算、信息处理
段利国（1970—），男，山西繁峙人，副教授，博士，CCF高级会员，主要研究方向：中文信息处理、知识图谱
付佳（1995—），女，内蒙古通辽人，硕士研究生，主要研究方向：服务计算、信息处理
陈永乐（1983—），男，山东潍坊人，副教授，博士，CCF会员，主要研究方向：无线传感网络、物联网安全。
基金资助:
国家重点研发计划“网络空间安全”专项子课题资助项目(2018YFB0803402)

Abstract

Abstract:

Service description texts for Internet of Things （IoT） are short in length and sparse in text features， and direct modeling the IoT service by using traditional topic model has poor clustering effect， so that the best service cannot be discovered. To solve this problem， an IoT service discovery method based on Biterm Topic Model （BTM） was proposed. Firstly， BTM was employed to mine the latent topic of the existing IoT services， and the service document-topic probability distribution was calculated and deduced through global topic distribution and theme-word distribution. Then， K-means algorithm was used to cluster the services and return the best matching results of service requests. Experimental results show that the proposed method can improve the clustering effect of services for IoT and thus obtain the matched best service. Compared with the methods of HDP （Hierarchical Dirichlet Process） and LDA-K （Latent Dirichlet Allocation based on K-means）， the proposed method achieves better performance in terms of Precision and Normalized Discounted Cumulative Gain （NDCG） for best service discovery.

Key words: service for Internet of Things (IoT), Biterm Topic Model (BTM), short text, topic modeling, service discovery

摘要：

针对物联网（IoT）服务描述文本篇幅较短、特征稀疏，直接采用传统的主题模型对IoT服务建模得到的聚类效果不佳，从而导致无法发现最佳服务的问题，提出了一种基于BTM的IoT服务发现方法。该方法首先利用BTM挖掘现有IoT服务的隐含主题，并通过全局主题分布和主题-词分布计算推理得到服务文档-主题概率分布；其次利用K-means算法对服务进行聚类，并返回服务请求的最佳匹配结果。实验结果分析表明，该方法能够有效提高IoT服务的聚类效果，从而得到匹配的最佳服务。与现有的HDP（Hierarchical Dirichlet Process）、基于K-means的隐狄利克雷分配（LDA-K）等方法相比，该方法进行最佳服务发现的准确度（Precision）和归一化折损累积增益（NDCG）均有一定幅度的提高。

关键词: 物联网服务, BTM, 短文本, 主题建模, 服务发现

CLC Number:

TP391

Shuman WANG, Aiping LI, Liguo DUAN, Jia FU, Yongle CHEN. Service discovery method for Internet of Things based on Biterm topic model[J]. Journal of Computer Applications, 2020, 40(2): 459-464.

王舒漫, 李爱萍, 段利国, 付佳, 陈永乐. 基于BTM的物联网服务发现方法[J]. 《计算机应用》唯一官方网站, 2020, 40(2): 459-464.

Figures/Tables 10

Fig. 1 Service description text for IoT

Fig. 2 Graphical model representation of BTM

Fig. 3 Service discovery framework for IoT based on BTM

Fig. 4 Flowchart of service clustering

Tab. 1 Some parameters information of sensors

设备类型

部分参数

红外

二氧化碳

传感器

功能：监测二氧化碳气体浓度并就地显示

工作温度：0～40 ℃

工作湿度：0～98％RH

供电电压：9～25V DC

输出方式：电流输出型4～20 mA

JCJ800D

压力传感器

功能：实现对液/气体压力的测量与控制

工作温度：-20～75 ℃

工作湿度：0～100％RH

供电电压：24V DC

量程范围：0～1～30 Mpa

输出方式：电压输出型0～10 V

分体式RS485

输出温湿度

变送器

功能：检测温度和湿度

工作温度：-20～55 ℃

工作湿度：5～95%RH

测湿范围：0～100%RH

测温范围：-20～50 ℃

输出方式：电流输出型4～20 mA

SBWR

温度传感器

功能：检测温度

工作温度：25～80 ℃

工作湿度：5～95％RH

测温范围：0～800 ℃

输出方式：电流输出型4～20 mA

Tab. 2 Service document-topic distribution matrix

文档	主题分布概率
文档	T₁	T₂	T₃	T₄	T₅	T₆
D₁	0.000 011 54	0.000 000 04	0.018 148 69	0.000 215 00	0.872 685 27	0.000 099 07
D₂	0.000 002 00	0.000 000 73	0.000 001 41	0.999 960 86	0.000 008 64	0.000 001 01
D₃	0.000 000 06	0.161 681 17	0.774 465 46	0.000 001 87	0.000 000 19	0.024 777 79
D₄	0.000 209 45	0.000 177 40	0.000 000 59	0.000 190 97	0.033 195 96	0.964 057 00
D₅	0.000 000 60	0.719 902 11	0.000 455 16	0.000 000 34	0.279 626 18	0.000 000 25

Fig. 5 Service clustering under different number of topics

Tab. 3 Distance ratio within and between classes under different number of topics with different K

$T$	$K$
$T$	5	6	7	8	9	10	11
10	0.73	0.70	0.67	0.71	0.76	0.83	0.87
15	0.68	0.60	0.53	0.62	0.69	0.75	0.76
20	0.79	0.73	0.71	0.80	0.83	0.82	0.81
30	0.86	0.82	0.79	0.81	0.84	0.87	0.87

Tab. 3 Distance ratio within and between classes under different number of topics with different K

$T$	$K$
$T$	5	6	7	8	9	10	11
10	0.73	0.70	0.67	0.71	0.76	0.83	0.87
15	0.68	0.60	0.53	0.62	0.69	0.75	0.76
20	0.79	0.73	0.71	0.80	0.83	0.82	0.81
30	0.86	0.82	0.79	0.81	0.84	0.87	0.87

Fig. 6 Precision comparison of IoT service query by five methods

Fig. 7 NDCG comparison of five methods under different service queries

References 16

1	中国信息通信研究院.物联网白皮书（2018年）［EB/OL］. ［2019-04-25］.［EB/OL］. ［2019-04-25］. ）
2	GUINARD D， TRIFA V， KARNOUSKOS S， et al. Interacting with the SOA-based Internet of things： discovery， query， selection， and on-demand provisioning of Web services［J］. IEEE Transactions on Services Computing， 2010， 3（3）： 223-235. 10.1109/tsc.2010.3
3	LIU Y， DU F， SUN J， et al. A crowdsourcing-based topic model for service matchmaking in Internet of things［J］. Future Generation Computer Systems， 2018， 87： 186-197. 10.1016/j.future.2018.05.005
4	JIANG B， YE L， WANG J， et al. A semantic-based approach to service clustering from service documents［C］// Proceedings of the 2017 IEEE International Conference on Services Computing. Piscataway： IEEE， 2017： 265-272. 10.1109/scc.2017.41
5	李平. 面向物联网设备的自动语义标注和服务封装的研究与应用［D］. 广州：华南理工大学， 2017： 6-8. 10.1155/2017/9589064
	LI P. Research and application of IoT device-oriented automatic semantic annotations and service encapsulation［D］. Guangzhou： South China University of Technology， 2017： 6-8. 10.1155/2017/9589064
6	王书龙，侯义斌，高放，等.基于本体的物联网设备资源描述模型［J］.北京工业大学学报，2017，43（5）：762-769. 10.11936/bjutxb2016070073
	WANG S L， HOU Y B， GAO F， et al. Ontology-based resource description model for things in IoT［J］. Journal of Beijing University of Technology， 2017， 43（5）：762-769. 10.11936/bjutxb2016070073
7	SANTOS A C， PEDROSA L D， KUIPERS M， et al. Resource description language： a unified description language for network embedded resources［J］. International Journal of Distributed Sensor Networks， 2012， 2012： 860-864. 10.1155/2012/860864
8	TEIXEIRA T， HACHEM S， ISSARNY V. Service oriented middleware for the internet of things： a perspective［C］// Proceedings of the 2011 European Conference on a Service based Internet， LNCS6994. Berlin： Springer， 2011：220-229.
9	CASSAR G， BARNAGHI P， WANG W， et al. A hybrid semantic matchmaker for IoT services［C］// Proceedings of the 2012 IEEE International Conference on Green Computing and Communications. Piscataway： IEEE， 2012：210-216. 10.1109/greencom.2012.40
10	魏强，金芝，许焱. 基于概率主题模型的物联网服务发现［J］. 软件学报， 2014， 25（8）：1640-1658.
	WEI Q， JIN Z， XU Y. Service discovery for Internet of things based on probabilistic topic model［J］. Journal of Software， 2014， 25（8）：1640-1658.
11	肖巧翔，曹步清，张祥平，等. 基于Word2Vec和LDA主题模型的Web服务聚类方法［J］. 中南大学学报（自然科学版）， 2018， 49（12）：2979-2985.
	XIAO Q X， CAO B Q， ZHANG X P， et al. Web services clustering based on Word2Vec and LDA topic model［J］. Journal of Central South University （Science and Technology）， 2018， 49（12）：2979-2985.
12	YAN X， GUO J， LAN Y， et al. A biterm topic model for short texts［C］// Proceedings of the 22nd International Conference on World Wide Web. New York： ACM， 2013：1445-1456. 10.1145/2488388.2488514
13	GRIFFITHS T L， STEYVERS M. Finding scientific topics［J］.PNAS， 2004， 101（S1）：5228-5235. 10.1073/pnas.0307752101
14	ZHANG Y， GONG L， WANG Y. An improved TF-IDF approach for text classification［J］. Journal of Zhejiang University-SCIENCE A， 2005， 6（1）： 49-55. 10.1631/jzus.2005.a0049
15	BLEI D M， NG A Y， JORDAN M I. Latent Dirichlet allocation［J］. Journal of Machine Learning Research， 2003， 3： 993-1022. 10.1145/1015330.1015439
16	CAO B， LIU X， LI B， 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. Piscataway： IEEE， 2016： 212-219. 10.1109/icws.2016.35

[1]	Shigang YANG, Yongguo LIU. Short text classification method by fusing corpus features and graph attention network [J]. Journal of Computer Applications, 2022, 42(5): 1324-1329.
[2]	Yu DENG, Xiaoyu LI, Jian CUI, Qi LIU. Multi-head attention memory network for short text sentiment classification [J]. Journal of Computer Applications, 2021, 41(11): 3132-3138.
[3]	YIN Chunyong, ZHANG Sun. End-to-end adversarial variational Bayes method for short text sentiment classification [J]. Journal of Computer Applications, 2020, 40(9): 2536-2542.
[4]	ZHANG Xiaochuan, DAI Xuyao, LIU Lu, FENG Tianshuo. Chinese short text classification model with multi-head self-attention mechanism [J]. Journal of Computer Applications, 2020, 40(12): 3485-3489.
[5]	CHEN Jie, SHAO Zhiqing, ZHANG Huanhuan, FEI Jiahui. Short text sentiment analysis based on parallel hybrid neural network model [J]. Journal of Computer Applications, 2019, 39(8): 2192-2197.
[6]	HE Yunhua, NIU Tong, LIU Tianyi, XIAO Ke, LU Xiang. Design and optimization of network anonymous scanning system [J]. Journal of Computer Applications, 2019, 39(5): 1385-1388.
[7]	CAO Dawei, HE Chaobo, CHEN Qimai, LIU Hai. Short text clustering algorithm based on weighted kernel nonnegative matrix factorization [J]. Journal of Computer Applications, 2018, 38(8): 2180-2184.
[8]	LIU Zhanghu, CHENG Chunling. Variance reduced stochastic variational inference algorithm for topic modeling of large-scale data [J]. Journal of Computer Applications, 2018, 38(6): 1675-1681.
[9]	DENG Yang, ZHANG Chenxi, LI Jiangfeng. Video shot recommendation model based on emotion analysis using time-sync comments [J]. Journal of Computer Applications, 2017, 37(4): 1065-1070.
[10]	XING Jinbiao, CUI Chaoyuan, SUN Bingyu, SONG Liangtu. Microblog advertisement filtering method based on classification feature extension of latent Dirichlet allocation [J]. Journal of Computer Applications, 2016, 36(8): 2257-2261.
[11]	ZHANG Jiujie, ZHAI Ye, WANG Chunhui, ZHANG Liping, LIU Dongsheng. Evolution pattern recognition and genealogy construction based on clone mapping of versions [J]. Journal of Computer Applications, 2016, 36(7): 2021-2030.
[12]	LI Yaxing, WANG Zhaokai, FENG Xupeng, LIU Lijun, HUANG Qingsong. Micro-blog hot-spot topic discovery based on real-time word co-occurrence network [J]. Journal of Computer Applications, 2016, 36(5): 1302-1306.
[13]	HU Zhikun, SONG Jingye, CHEN Yuan. Random service system model based on UPnP service discovery [J]. Journal of Computer Applications, 2016, 36(3): 591-595.
[14]	LU Ling, YANG Wu, CAO Qiong. Automatic short text summarization method based on multiple mapping [J]. Journal of Computer Applications, 2016, 36(2): 432-436.
[15]	CHEN Qian, GUI Zhiguo, GUO Xin, XIANG Yang. Topic evolution in text stream based on feature ontology [J]. Journal of Computer Applications, 2015, 35(2): 456-460.

Service discovery method for Internet of Things based on Biterm topic model

基于BTM的物联网服务发现方法

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 10

References 16

Related Articles 15

Recommended Articles

Metrics