隨著軟體以及雲端平台的流行，軟體服務也被廣泛的開發及使用。而網路服務是使用全球資訊網通訊協定以及服務導向結構的一種軟體服務。在網路人人普及的現代，使用者可以在網路上找到各種符合自己需求的服務，但軟體服務是否可以穩定且良好的提供服務則成為使用上的一項隱憂，因此軟體服務的品質可靠度成為使用者在選擇軟體服務上的一項考量依據，同樣地服務提供者為了能提供更好的使用者體驗也會不停的監測服務狀態並更新服務內容，因此軟體品質成為一項重要的議題。
我們在此論文中，針對目前的軟體服務可靠度衡量方法以及品質預測模型進行討論，可發現目前計算可靠度的方法都是靜態的而對服務品質的預測也有很多不同面向，有預測不同使用者的對服務的使用體驗也有針對服務品質某一特質作時間上的預測，其中許多是關於回應時間的預測研究。然而隨著機器學習被廣泛利用在資料分析以及回歸預測，因此許多研究利用深度學習針對時間序列做預測分。而服務品質訊息也可被視為一個連續的時間序列資料，目前也有一些針對服務品質的序列預測研究，但其方法大多都是針對該研究品質序列作單一輸入的預測模型。然而時間序列資料很有可能有記憶性，因此可以透過在每個資料序列中一次採取多個資料點作為輸入，讓模型可以同時考量過去多個資訊。
另外服務品質包羅萬象，在監控服務時可以同時得到多個資訊，如輸出量和回應時間。而這些不同的品質特質可能有相關性，而這些關聯性可以利用機器學習的能力自我訓練找到進而增加預測服務品質的準確性。因此我們想利用機器學習去作出考量多重條件的模型，也就是讓模型同時輸入多個因素以及多個時間序列資料的方式。預期可以藉由加入影響因子進而提供一個更準確的預測方法，另外一個考量是多個資料變數可以透過增加連續資料而可以讓模型擁有時間性記憶的考量並且可以得知我們的軟體品質序列資料具有規律性。
在我們的結果中，可看出我們選出的最佳模型大多都選擇大於一個變數數量，故在模型在考慮多個變數確實較單一變數輸入更好。我們提出的四個方法中，如過去的方法，單一方法是採取單一個品質序列，結果顯示72% 的實驗裡單一方法表現比其他使用單一時間序列的方法更好。而在其中四個含有多種因素的資料集裡，有三個資料集使用多因素方法預測得更好，也就是使用組合方法平均方法和加總方法。每個模型在不同資料集的平均訓練時間不同，我們使用回歸模型更是花費最少時間，而ARIMA比起其他模型還節省時間卻未有叫好的表現，我們使用CNN和RNN比起其他人使用LSTM和GRU的方法更快速收斂同時也能有不劣於他們的預測表現。

With the growing popularity of software and cloud platforms, services have also been widely developed and used. Web services are software services that use the World Wide Web protocol and service-oriented architecture(SOA). In the popularity of the Internet, users can find a variety of services on the Internet to meet their needs; but whether the software service can be stable and good service is a problem for user; therefore, the reliability of software services has become a consideration for users in selecting software services. Similarly, service providers seek to continuously monitor the service status and update service to provide a better user experience. Therefore, software quality has become an important issue.
In this study, we discuss the current software service reliability measurement method and quality of service (QoS) prediction model. The current methods for calculating reliability are statics and there are many different aspects for predicting service quality. There are QoS prediction approaches relating to the service experience of different users and a time series prediction of future QoS. With machine learning being widely used in data analysis and regression prediction, many studies use it to make prediction points for time series. The service quality message can also be regarded as a continuous time series data. There are also some sequence prediction studies for service quality and most of the methods are models using single data point as input. However, time series data is likely to have a memory characteristic, so it is possible to construct models that can simultaneously consider multiple information at the same time by taking multiple data points at a time in data sequence as input.
In addition, the QoS is varied; thus, we get multiple messages at the same time when monitoring services, such as output and response time. These different quality factors may be relevant, and these associations can be self-trained by using machine learning and increasing the accuracy of predictive service quality. Therefore, we propose methods that use multiple factors based on deep learning considering multiple variables, that is, the way the model inputs multiple factors and multiple time series data at the same time. It is expected that this approach might provide a more accurate prediction method by adding related factors. Another consideration is that multiple variables can increase continuous data to allow the model to have temporal memory considerations and to determine if the QoS series data is regular.
In our results, most of the best models we chose used more than one variable, so it is better to consider multiple variables in the model than single variable inputs. Among the four methods we proposed, the Single method involves using a single quality sequence which likes past approach, and the results demonstrated that in 72% of the experiments, the Single method performeds better than other methods using a single time series. Among the four datasets with multiple factors, three datasets were better predicted using a multi-factor method, that is, using the Composed method or the Average method or the Aggregation method. Each model displayed different average training time in different datasets. Regression model to predict QoS took the least time in all datasets, while autoregressive integrated moving average model (ARIMA) saved more time compared to other models but does not have good performance. We used a convolutional neural network (CNN) and recurrent neural network (RNN) to converge more quickly than other approaches using LSTM and GRU, and to also achieve good prediction performance that was not inferior to other models.

[1] K. Kritikos and D. Plexousakis, "Requirements for QoS-Based Web service Description and Discovery," IEEE Transactions on Services Computing, vol. 2, no. 4, pp. 320-337, Sep 11 Sep. 2009.
[2] D. Menasce, "QoS issues in Web services," IEEE Internet Computing, vol. 6, no. 6, pp. 72-75, Dec. 2002.
[3] P. Killelea, Web Performance Tuning: speeding up the web, 2 ed. O'Reilly Media, Inc, 2002.
[4] C. Schroth and T. Janner, "Web 2.0 and SOA: Converging Concepts Enabling the Internet of Services," IT Professional, vol. 9, no. 3, pp. 36-41, May. 2007.
[5] K. Julisch, C. Suter, T. Woitalla, and O. Zimmermann, "Compliance by design – Bridging the chasm between auditors and IT architects," Computers & Security, vol. 30, no. 6, pp. 410-426, Sep. 2011.
[6] Y. Duan, "A Survey on Service Contract," in Proceeding of 2012 13th ACIS International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing, 2012, pp. 805-810.
[7] M. H. Valipour, B. Amirzafari, K. Maleki, and N. Daneshpour, "A brief survey of software architecture concepts and service oriented architecture," in Proceeding of 2009 2nd IEEE International Conference on Computer Science and Information Technology, 2009, pp. 34-38.
[8] N. Milanovic and M. Malek, "Current solutions for Web service composition," IEEE Internet Computing, vol. 8, no. 6, pp. 51-59, Nov. 2004.
[9] D. Fensel and C. Bussler, "The Web service Modeling Framework WSMF," Electronic Commerce Research and Applications, vol. 1, no. 2, pp. 113-137, Jun. 2002.
[10] R. Hamadi and B. Benatallah, "A Petri net-based model for Web service composition," in Proceedings of the 14th Australasian database conference Adelaide, Australia, 2003, pp. 191-200.
[11] H. Foster, S. Uchitel, J. Magee, and J. Kramer, "Model-based verification of Web service compositions," in Proceeding of 18th IEEE International Conference on Automated Software Engineering, 2003, pp. 152-161.
[12] J. Rao and X. Su, "A Survey of Automated Web service Composition Methods," in SWSWPC 2004: Semantic Web services and Web Process Composition, Berlin, Heidelberg, 2005, pp. 43-54.
[13] Z. Zheng, H. Ma, M. Lyu, and I. King, "QoS-Aware Web service Recommendation by Collaborative Filtering," IEEE Transactions on Services Computing, vol. 4, no. 2, pp. 140-152, Dec. 2011.
[14] Z. Zheng, H. Ma, M. Lyu, and I. King, "WSRec: A Collaborative Filtering Based Web service Recommender System," in Proceeding of 2009 IEEE International Conference on Web services, 2009, pp. 437-444.
[15] A. Charfi and M. Mezini, "Aspect-Oriented Web service Composition with AO4BPEL," in Proceeding of European Conference on Web services, Berlin, Heidelberg, 2004, pp. 168-182.
[16] Y. Liu, A. Ngu, and L. Zeng, "QoS computation and policing in dynamic Web service selection," in Proceedings of the 13th international World Wide Web conference on Alternate track papers & posters, New York, USA, 2004, pp. 66-73.
[17] K. Peng and C. Huang, "Stochastic modelling and simulation approaches to analysing enhanced fault tolerance on service-based software systems," Software Testing Verification and Reliability vol. 26, no. 4, pp. 276-293, Jan. 2015.
[18] K. Peng and C. Huang, "Reliability Evaluation of Service-Oriented Architecture Systems Considering Fault-Tolerance Designs," Journal of Applied Mathematics, vol. 2014, p. 11, Jan. 2014.
[19] K. Peng and C. Huang, "Reliability Analysis of On-Demand Service-Based Software Systems Considering Failure Dependencies," IEEE Transactions on Services Computing, vol. 10, no. 3, pp. 423-435, Aug. 2017.
[20] C. Tsai, "Reliability Prediction and Assessment of Service oriented Software Systems Considering the Effect of Error Propagation," Master's thesis, National Tsing Hua University, Hsinchu, Taiwan, Jul. 2015.
[21] A. Gorbenko, A. Romanovsky, V. Kharchenko, and A. Mikhaylichenko, "Experimenting with exception propagation mechanisms in service-oriented architecture," in Proceedings of the 4th international workshop on Exception handling, Nov. 2008, pp. 1-7.
[22] M. Chiang, "Considering Error Propagation and Fault Tolerance in Reliability Assessment and Sensitivity Analysis of Service Oriented Software Systems,," Master's thesis, National Tsing Hua University, Hsinchu, Taiwan, Jul. 2017.
[23] L. Coppolino, L. Romano, N. Mazzocca, and S. Salvi, "Web services workflow reliability estimation through reliability patterns," in Proceeding of 2007 Third International Conference on Security and Privacy in Communications Networks and the Workshops - SecureComm 2007, 2007, pp. 107-115.
[24] R. Cheung, "A User-Oriented Software Reliability Model," IEEE Transactions on Software Engineering, vol. SE-6, no. 2, pp. 118-125, Mar. 1980.
[25] Y. Dai, Y. Pan, and X. Zou, "A Hierarchical Modeling and Analysis for Grid Service Reliability," IEEE Transactions on Computers, vol. 56, no. 5, pp. 681-691, Apr. 2007.
[26] S. Yacoub, B. Cukic, and H. Ammar, "A scenario-based reliability analysis approach for component-based software," IEEE Transactions on Reliability, vol. 53, no. 4, pp. 465-480, Nov. 2004.
[27] S. Gokhale and K. Trivedi, "Reliability prediction and sensitivity analysis based on software architecture," in Proceeding of 13th International Symposium on Software Reliability Engineering, Annapolis, MD, USA, 2002, pp. 64-75.
[28] K. Go!seva-Popstojanova and M. Hamill, "Architecture-Based Software Reliability: Why Only a Few Parameters Matter?," in Proceeding of 31st Annual International Computer Software and Applications Conference (COMPSAC 2007), 2007, vol. 1, pp. 423-430.
[29] V. Grassi, "Architecture-Based Reliability Prediction for Service-Oriented Computing," in Architecting Dependable Systems III, R. de Lemos, C. Gacek, and A. Romanovsky, Eds. Berlin, Heidelberg, 2005, pp. 279-299.
[30] A. Filieri, C. Ghezzi, V. Grassi, and R. Mirandola, "Reliability Analysis of Component-Based Systems with Multiple Failure Modes," in Proceeding of International Symposium on Component-Based Software Engineering, Berlin, Heidelberg, 2010, pp. 1-20.
[31] K. M. Chan, J. Bishop, J. Steyn, L. Baresi, and S. Guinea, "A Fault Taxonomy for Web service Composition," in Proceeding of International Conference on Service-Oriented Computing, Berlin, Heidelberg, 2009, pp. 363-375.
[32] K. Goseva-Popstojanova et al., "Architectural-level risk analysis using UML," IEEE Transactions on Software Engineering, vol. 29, no. 10, pp. 946-960, Oct. 2003.
[33] T. Pitakrat, D. Okanovic, A. Hoorn, and L. Grunske, "An Architecture-Aware Approach to Hierarchical Online Failure Prediction," in Proceeding of 2016 12th International ACM SIGSOFT Conference on Quality of Software Architectures (QoSA), 2016, pp. 60-69.
[34] Z. Liu, Z. Pu Jia, X. Xue, and J. An, "Reliable Web service composition based on QoS dynamic prediction," Soft Computing, vol. 19, May. 2015.
[35] Z. Zheng, Y. Zhang, and M. Lyu, "Investigating QoS of Real-World Web services," IEEE Transactions on Services Computing, vol. 7, no. 1, pp. 32-39, 2014.
[36] M. Zadeh and M. Seyyedi, "Qos monitoring for Web services by Time Series Forecasting," in Proceeding of 2010 3rd International Conference on Computer Science and Information Technology, 2010, vol. 5, pp. 659-663.
[37] Z. Zheng and M. Lyu, "Collaborative reliability prediction of service-oriented systems," in Proceeding of 2010 ACM/IEEE 32nd International Conference on Software Engineering, 2010, vol. 1, pp. 35-44.
[38] C. Wu, W. Qiu, X. Wang, Z. Zheng, and X. Yang, "Time-Aware and Sparsity-Tolerant QoS Prediction Based on Collaborative Filtering," in Proceeding of 2016 IEEE International Conference on Web services (ICWS), 2016, pp. 637-640.
[39] A. Amin, A. Colman, and L. Grunske, "An Approach to Forecasting QoS Attributes of Web services Based on ARIMA and GARCH Models," in Proceeding of 2012 IEEE 19th International Conference on Web services, 2012, pp. 74-81.
[40] Z. Rahman, O. Hussain, and F. Hussain, "Time Series QoS Forecasting for Management of Cloud Services," in Proceeding of 2014 Ninth International Conference on Broadband and Wireless Computing, Communication and Applications, 2014, pp. 183-190.
[41] R. Calheiros, E. Masoumi, R. Ranjan, and R. Buyya, "Workload Prediction Using ARIMA Model and Its Impact on Cloud Applications’ QoS," IEEE Transactions on Cloud Computing, vol. 3, no. 4, pp. 449-458, Aug. 2015.
[42] L. Li, M. Rong, and G. Zhang, "A Web service QoS prediction approach based on multi-dimension QoS," in Proceeding of 2011 6th International Conference on Computer Science & Education (ICCSE), 2011, pp. 1319-1322.
[43] A. Amin, L. Grunske, and A. Colman, "An automated approach to forecasting QoS attributes based on linear and non-linear time series modeling," in Proceedings of the 27th IEEE/ACM International Conference on Automated Software Engineering, Essen, Germany, 2012, pp. 130-139.
[44] M. Li, J. Huai, and H. Guo, "An Adaptive Web services Selection Method Based on the QoS Prediction Mechanism," in Proceeding of 2009 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology, 2009, vol. 1, pp. 395-402.
[45] G. White, A. Palade, and S. Clarke, "Forecasting qos attributes using lstm networks," in Proceeding of 2018 International Joint Conference on Neural Networks (IJCNN), 2018, pp. 1-8.
[46] P. Zhang, Y. Sun, W. Li, W. Song, and H. Leung, "A Combinational QoS-Prediction Approach Based on RBF Neural Network," in Proceeding of 2016 IEEE International Conference on Services Computing (SCC), 2016, pp. 577-584.
[47] H. Wang, L. Wang, Q. Yu, and Z. Zheng, "Learning the Evolution Regularities for Big Service-Oriented Online Reliability Prediction," IEEE Transactions on Services Computing, vol. PP, pp. 1-1, Nov. 2016.
[48] G. Box, G. M. Jenkins, G. Reinsel, and G. Ljung, Time series analysis: forecasting and control. John Wiley & Sons, Jun. 2015.
[49] S. SAID and D. DICKEY, "Testing for unit roots in autoregressive-moving average models of unknown order," Biometrika, vol. 71, no. 3, pp. 599-607, Dec. 1984.
[50] H. Akaike, "A new look at the statistical model identification," in Selected Papers of Hirotugu Akaike: Springer, 1974, pp. 215-222.
[51] "Postman," [Online] Available: https://www.getpostman.com/. Accessed: Mar. 6, 2019.
[52] Y. Lv, Y. Duan, W. Kang, Z. Li, and F. Wang, "Traffic Flow Prediction With Big Data: A Machine Learning Approach," IEEE Transactions on Intelligent Transportation Systems, vol. 16, no. 2, pp. 865-873, Sep. 2014.
[53] Z. Zhao, W. Chen, X. Wu, P. C. Y. Chen, and J. Liu, "LSTM network: a machine learning approach for short-term traffic forecast," IET Intelligent Transport Systems, vol. 11, no. 2, pp. 68-75, Mar. 2017.
[54] W. Bao, J. Yue, and Y. Rao, "A machine learning framework for financial time series using stacked autoencoders and long-short term memory," PLOS ONE.0180944, 2017.
[55] H. Yang, L. Chan, and I. King, "Support Vector Machine Regression for Volatile Stock Market Prediction," in International Conference on Intelligent Data Engineering and Automated Learning, Berlin, Heidelberg, Aug. 2002, pp. 391-396.
[56] C. Wu, J. Ho, and D. Lee, "Travel-time prediction with support vector regression," IEEE Transactions on Intelligent Transportation Systems, vol. 5, no. 4, pp. 276-281, Dec. 2004.
[57] S. Weisberg, Applied linear regression. John Wiley & Sons, 2005.
[58] D. M. J. J. o. c. i. Hawkins and c. sciences, "The problem of overfitting," Journal of Chemical Information and Computer Sciences, vol. 44, no. 1, pp. 1-12, Dec. 2004.
[59] K. Zhang, W. Zuo, Y. Chen, D. Meng, and L. Zhang, "Beyond a gaussian denoiser: Residual learning of machine cnn for image denoising," IEEE Transactions on Image Processing, vol. 26, no. 7, pp. 3142-3155, Feb. 2017.
[60] "Google Maps Platform," [Online] Available: https://cloud.google.com/maps-platform/. Accessed: Feb. 8, 2019
[61] Y. Zhang, Z. Zheng, and M. Lyu, "WSPred: A Time-Aware Personalized QoS Prediction Framework for Web services," in Proceeding of IEEE 22nd International Symposium on Software Reliability Engineering, Hiroshima, Japan, 2011, pp. 210-219.
[62] J. Malak, M. Mohsenzadeh, and M. Seyyedi, "Web service QoS Prediction Based on Multi Agents," in Proceeding of 2009 International Conference on Computer Technology and Development, 2009, vol. 1, pp. 265-269.
[63] S. Wang, C. Hsu, Z. Liang, Q. Sun, and F. Yang, "Multi-user Web service selection based on multi-QoS prediction," Information Systems Frontiers, vol. 16, no. 1, pp. 143-152, Mar. 2014.
[64] C. Yu and L. Huang, "A Web service QoS prediction approach based on time- and location-aware collaborative filtering," Service Oriented Computing and Applications (SOCA), 10/31 Jun. 2016.
[65] "PayPal," [Online] Available: https://www.paypal.com/tw/webapps/mpp/merchant. Accessed: Mar. 9, 2019.
[66] B. Cavallo, M. Di Penta, and G. Canfora, "An empirical comparison of methods to support QoS-aware service selection," in Proceedings of the 2nd International Workshop on Principles of Engineering Service-Oriented Systems, 2010, pp. 64-70: ACM.
[67] Y. Syu, Y. Fanjiang, J. Kuo, and J. Su, "Quality of Service timeseries forecasting for Web services: A machine learning, Genetic Programming-based approach," in Proceeding of 2016 Annual Conference on Information Science and Systems (CISS), 2016, pp. 343-348.
[68] E. Lehmann and G. Casella, Theory of point estimation, 2 ed. Springer-Verlag New York, 1998.
[69] S. Conte, H. Dunsmore, and V. Shen, Software engineering metrics and models. Benjamin-Cummings Publishing Co., Inc., 1986, p. 396.
[70] J. Cort and M. Kenji, "Advantages of the mean absolute error (MAE) over the root mean square error (RMSE) in assessing average model performance," Climate Research, vol. 30, no. 1, pp. 79-82, Dec. 2005.
[71] C. Willmott and K. Matsuura, "On the use of dimensioned measures of error to evaluate the performance of spatial interpolators," International Journal of Geographical Information Science, vol. 20, no. 1, pp. 89-102, 2006.