基于机器学习的服务组合技术研究进展

doi:10.16180/j.cnki.issn1007-7820.2022.11.009

摘要/Abstract

摘要：

服务组合是服务计算领域内的经典研究问题,在工业及学术领域内受到广泛关注。随着云原生和微服务技术不断普及,服务组合领域涌现出一系列富有创新性的研究。随着计算机技术和人工智能的快速发展,深度学习、强化学习等机器学习算法被越来越多地应用于传统服务组合问题中。文中介绍了服务组合问题常见的分类及面临的挑战,并对近年来涌现的机器学习算法在服务组合问题中的应用进行了归纳介绍。此外,文中还总结了基于机器学习算法解决服务组合问题时所面临的问题,并对今后的发展方向进行了展望。

关键词: 机器学习, 深度学习, 强化学习, 人工智能, 服务组合, 组合优化, 服务计算, 服务质量

Abstract:

Service composition is a classic research problem in the field of service computing, which has received extensive attention in both industrial and academic fields. With the increasing popularity of cloud-native and micro-service technologies, a series of innovative researches have emerged in the field of service composition. With the rapid development of computer technology and artificial intelligence, machine learning algorithms such as deep learning and reinforcement learning are increasingly applied to traditional service composition problems. This study introduces the common classification and challenges of service composition problems, and summarizes the application of machine learning algorithms emerging in service composition problems in recent years. Additionally, the proposed study also summarizes the problems faced in the direction of solving service composition problems based on machine learning algorithms, and looks forward to the future development direction.

Key words: machine learning, deep learning, reinforcement learning, artificial intelligence, service composition, portfolio optimization, services computing, quality of service

中图分类号:

TP389.1

张晔,鲍亮. 基于机器学习的服务组合技术研究进展[J]. 电子科技, 2022, 35(11): 58-63.

ZHANG Ye,BAO Liang. Research Progress of Service Composition Based on Machine Learning[J]. Electronic Science and Technology, 2022, 35(11): 58-63.

图/表 2

表1

表2

参考文献 34

[1]	Wang H, Hu X, Yu Q, et al. Integrating reinforcement learning and skyline computing for adaptive service composition[J]. Information Sciences, 2020, 51(9):141-160.
[2]	Khakhkhar S, Kumar V, Chaudhary S. Dynamic service composition[J]. International Journal of Computer Science & Artificial Intellige, 2012, 2(3):32-42.
[3]	Silva A S, Ma H, Mei Y, et al. A survey of evolutionary computation for web service composition: A technical perspective[J]. IEEE Transactions on Emerging Topics in Computational Intelligence, 2020, 4(4):538-554. doi: 10.1109/TETCI.2020.2969213
[4]	Peng S, Wang H, Yu Q. Multi-clusters adaptive brain storm optimization algorithm for QoS-aware service composition[J]. IEEE Access, 2020(8):48822-48835.
[5]	Li C, Li J, Chen H. A meta-heuristic-based approach for Qos-aware service composition[J]. IEEE Access, 2020(8):69579-69592.
[6]	Alayed H, Dahan F, Alfakih T, et al. Enhancement of ant colony optimization for QoS-aware web service selection[J]. IEEE Access, 2019(7):97041-97051.
[7]	Gabrel V, Manouvrier M, Moreau K, et al. QoS-aware automatic syntactic service composition problem: complexity and resolution[J]. Future Generation Computer Systems, 2018, 80(1):311-321. doi: 10.1016/j.future.2017.04.009
[8]	Li J, Fan G, Zhu M, et al. Pre-joined semantic indexing graph for QoS-aware service composition[C]. Milan: Proceedings of the IEEE International Conference on Web Services, 2019.
[9]	Fan S L, Peng K Y, Yang Y B. Large-scale QoS-aware service composition integrating chained dynamic programming and hybrid pruning[C]. Seattle: Proceedings of the International Conference on Web Services, 2018.
[10]	Xu X, Sheng Q Z, Wang Z, et al. Novel artificial bee colony algorithms for QoS-aware service selection[J]. IEEE Transactions on Services Computing, 2016, 12(2):247-261. doi: 10.1109/TSC.2016.2612663
[11]	Fan S L, Ding F, Guo C H, et al. Supervised web service composition integrating multi-objective QoS optimization and service quantity minimization[C]. Seattle: Proceedings of the International Conference on Web Services, 2018.
[12]	Xie Y, Guo Y, Mi Z, et al. Loosely coupled cloud robotic framework for QoS-driven resource allocation-based Web service composition[J]. IEEE Systems Journal, 2019, 14(1):1245-1256. doi: 10.1109/JSYST.2019.2904098
[13]	Min X, Xu X, Liu Z, et al. An approach to resource and QoS-aware services optimal composition in the big service and internet of things[J]. IEEE Access, 2018(6):39895-39906.
[14]	Chen Y, Huang J, Lin C, et al. Multi-objective service composition with QoS dependencies[J]. IEEE Transactions on Cloud Computing, 2016, 7(2):537-552. doi: 10.1109/TCC.2016.2607750
[15]	Gavvala S K, Jatoth C, Gangadharan G R, et al. QoS-aware cloud service composition using eagle strategy[J]. Future Generation Computer Systems, 2019, 90(2):273-290. doi: 10.1016/j.future.2018.07.062
[16]	Moustafa A, Ito T. A deep reinforcement learning approach for large-scale service composition[C]. Tokyo: Proceedings of the International Conference on Principles and Practice of Multi-Agent Systems, 2018.
[17]	Philip S Y. Multi-agent reinforcement learning for service composition[C]. San Francisco: Proceedings of the IEEE International Conference on Services Computing, 2016.
[18]	Wang H B, Wang X J, Zhang X Z, et al. Effective service composition using multi-agent reinforcement learning[J]. Knowledge-Based Systems, 2016, 92(3):151-168. doi: 10.1016/j.knosys.2015.10.022
[19]	Wang H, Wang X, Hu X, et al. A multi-agent reinforcement learning approach to dynamic service composition[J]. Info-rmation Sciences, 2016, 36(3):96-119.
[20]	Ren L, Wang W, Xu H. A reinforcement learning method for constraint-satisfied services composition[J]. IEEE Transactions on Services Computing, 2017(3):1-12.
[21]	杨波, 胡国兵. 利用记忆单元改进DQN的Web服务组合优化方法[J]. 计算机应用与软件, 2020, 37(11):11-17.
	Yang Bo, Hu Guobing. A web service composition optimization approach based on memory units and improved DQN[J]. Computer Applications and Software, 2020, 37(11):11-17.
[22]	Liu J W, Hu L Q, Cai Z Q, et al. Large-scale and adaptive service composition based on deep reinforcement learning[J]. Journal of Visual Communication and Image Representation, 2019, 6(5):1-5. doi: 10.1006/jvci.1995.1001
[23]	Wang H B, Gu M Z, Yu Q, et al. Large-scale and adaptive service composition using deep einforcement learning[C]. Malaga: Proceedings of the International Conference on Service-Oriented Computing, 2017.
[24]	Wang H, Gu M, Yu Q, et al. Adaptive and large-scale service composition based on deep reinforcement learning[J]. Knowledge-Based Systems, 2019, 18(7):75-90.
[25]	Zhao Y, Wang S, Zou Y, et al. Automatically learning user preferences for personalized service composition[C]. Honolulu: Proceedings of the IEEE International Conference on Web Services, 2017.
[26]	Zanbouri K, Jafari Navimipour N. A cloud service composition method using a trust-based clustering algorithm and honeybee mating optimization algorithm[J]. International Journal of Communication Systems, 2020, 33(5):4259-4269.
[27]	Li F, Zhang L, Liu Y, et al. A clustering network-based approach to service composition in cloud manufacturing[J]. International Journal of Computer Integrated Manufacturing, 2017, 30(12):1331-1342. doi: 10.1080/0951192X.2017.1314015
[28]	Khanouche M E, Attal F, Amirat Y, et al. Clustering-based and QoS-aware services composition algorithm for ambient intelligence[J]. Information Sciences, 2019, 48(2):419-439.
[29]	Silva A S, Mei Y, Ma H, et al. Evolutionary computation for automatic web service composition:An indirect representation approach[J]. Journal of Heuristics, 2018, 24(3):425-456. doi: 10.1007/s10732-017-9330-4
[30]	Sadeghiram S, Ma H, Chen G. Cluster-guided genetic algorithm for distributed data-intensive web service composition[C]. Rio de Janeiro: Proceedings of the IEEE Congress on Evolutionary Computation, 2018.
[31]	Haytamy S, Omara F. A deep learning based framework for optimizing cloud consumer QoS-based service composition[J]. Computing, 2020, 10(2):1117-1137.
[32]	Labbaci H, Medjahed B, Aklouf Y. A deep learning approach for long term QoS-compliant service composition[C]. Malaga: Proceedings of the International Conference on Service-Oriented Computing, 2017.
[33]	Wang H, Li J, Yu Q, et al. Integrating recurrent neural networks and reinforcement learning for dynamic service composition[J]. Future Generation Computer Systems, 2020, 10(7):551-563.
[34]	Elsayed D H, Nasr E S, Alaa El Din M, et al. A new hybrid approach using genetic algorithm and Q-learning for QoS-aware web service composition[C]. Cairo: Proceedings of the International Conference on Advanced Intelligent Systems and Informatics, 2017.

算法	机器学习框架	备注
文献[16]	Double Q-Learning	面向大规模动态服务组合环境
文献[17]	Multi-Q-Learning	面向动态服务组合环境
文献[18]	Multi-Q-Learning	面向大规模动态服务组合环境
文献[19]	分布式Q-Learning	面向大规模动态服务组合环境
文献[20]	Q-Learning	面向动态服务组合环境,同时考虑用户对QoS的约束
文献[21]	LSTM改进的Deep Q-Learning	面向大规模服务组合环境
文献[22]	基于RNN的Deep Q-Learning	将神经网络用于预测目标函数,改善目标函数表达和泛化能力
文献[23]	基于RNN的Deep Q-Learning	面向自适应和大规模服务组合
文献[24]	基于RNN的Deep Q-Learning	面向自适应性和大规模服务组合
文献[25]	Q-Learning	自动提取用户偏好及优先级,根据优先级赋予优化目标几组不同权重,进而使用强化学习算法得出多组推荐组合
文献[1]	Q-Learning	面向大规模动态服务组合环境

算法	机器学习框架	备注
文献[26]	k-means	减少搜索空间
文献[27]	基于概念相似度的聚类算法	将功能相近的服务聚簇为抽象服务
文献[28]	k-means	使用聚类方法,解决全局QoS约束问题
文献[30]	k-means	解决服务分布式性质
文献[4]	TWSVM	减少搜索空间
文献[31]	LSTM	预测服务未来QoS变化趋势
文献[32]	LSTM	预测服务未来QoS变化趋势
文献[33]	LSTM	预测服务未来QoS变化趋势
文献[34]	Q-Learning	生成GA的初始种群