Neighborhood-aware web service quality prediction using deep learning

Web services QoS (Quality of Service) prediction in dynamic environment is a critical method to support service selection and service composition. The existing QoS prediction methods focus on the user-side QoS but not the server-side. However, the prediction accuracy of QoS prediction methods for user-side QoS are restricted by the subjectivity of user's feedback, while the server-side QoS prediction methods could get higher prediction accuracy with relying on objective QoS measured value. To this end, we purposed a server-side QoS prediction method for Web service dubbed QoSPAC. Based on Kalman Filter, QoSPAC generates the QoS predicted value and uses the QoS measured value to calibrate it. Furthermore, QoSPAC measures the prediction accuracy, which is also used to calibrate and optimize the prediction period. Experimental results demonstrate that QoSPAC can adapt to the noteworthy measurement error, and can significantly improve the prediction accuracy compared with the conventional technology.

With the great development of mobile services, the Quality of Services (QoS) becomes an essential factor to meet end users’ personalized requirement on the nonfunctional performance of mobile services. However, most of the QoS values in real cases are unattainable because a service user would only invoke some specific mobile services. Therefore, how to predict the missing QoS values and recommend high-quality services to end users becomes a significant challenge in mobile service recommendation research. Previous QoS prediction researches demonstrate that the nonfunctional performance of mobile services is closely related to users’ location information. However, most location-aware QoS prediction methods ignore the premise that the obtainable QoS values observed by different users in same location region would probably be untrustworthy, which will lead to inaccurate and unreliable prediction results. To make credible location-aware QoS prediction, we propose a hybrid matrix factorization method integrated location and reputation information (LRMF) to predict the unattainable QoS values. Our approach firstly cluster users into different locational region based on their geographical distribution, and then we compute users’ reputation to identify untrustworthy users in every locational region. Finally, the unknown QoS values can be predicted by integrating locational cluster information and users’ reputation into a hybrid matrix factorization model. Comprehensive experiments are conducted on a public QoS dataset which contains sufficient real-world service invocation records. The evaluation results indicate that our LRMF method can effectively reduce the impact of unreliable users on QoS prediction and make credible mobile service recommendation.

Predicting quality of services (QoS) is an important need when ranking cloud services for selection. QoS values of cloud services usually depend heavily on the user’s and service’s environments. Therefore, personalized QoS value prediction for cloud services is more desirable to users. Collaborative Filtering (CF) has recently been applied to personalized QoS prediction for services on the Web. However, seldom did they take the context information of service users and services into consideration. The following paper presents a CF-based method for predicting QoS values of cloud services. The method exploits not only the QoS information of users and services, but also one type of the most important context information of users and services, i.e., their geographic locations. Experiments conducted on a real dataset show that geographic location information is indeed helpful for improving the QoS prediction performance. The experimental results also demonstrate that the proposed method is significantly better than previous methods in prediction accuracy.

QoS prediction for web services in cloud environments based on swarm intelligence search

With the fast-developing domain of cyber-physical systems (CPS), constructing the CPS with high-quality services becomes an imperative task. As one of the effective solutions for information overload in CPS construction, quality of service (QoS)-aware service recommendation has drawn much attention in academia and industry. However, the lack of most QoS values limits the recommendation performance and it is time-consuming for users to get the QoS values by invoking all the services. Therefore, a powerful prediction model is required to predict the unobserved QoS values. Considering the fact that most existing QoS prediction models are unable to effectively address the data-sparsity problem, a novel two-stage framework called AgQ is proposed for QoS prediction. Specifically, a data augmentation strategy is designed in the first stage to enlarge the training set by drawing additional virtual instances. In the second stage, a prediction model is applied that considers both virtual and factual instances during the training procedure. We conduct extensive experiments on the WSDream dataset to demonstrate the effectiveness of the our QoS prediction framework and verify that the data augmentation strategy can indeed alleviate the data-sparsity problem. In terms of mean absolute error, taking the Multilayer Perceptron model as an example, the maximum improvement achieves 5% under 5% sparsity.

Quality of Service (QoS) has been widely used to support dynamic Web Service (WS) selection and composition. Due to the volatile nature of QoS parameters, QoS prediction has been put forward to understand the trend of QoS data volatility and estimate QoS values in dynamic environments. In order to provide adaptive and effective QoS prediction, we propose a WS QoS prediction approach, named WS-QoSP, based on the technique of forecast combination. Different from the existing QoS prediction approaches that choose a most feasible forecasting model and predict relying on this "best" model, WS-QoSP selects multiple potential forecasting models, combines the results of the selected models to optimize the overall forecast accuracy. Results of real data experiments demonstrate the diversified forecast accuracy gains by using WS-QoSP.

With the rapid growth of Web services, recommending suitable services to users has become a big challenge. The existing service recommendation works by Quality of Service (QoS) prediction fail to fully consider the influence of situation information, such as time, location, and user relations thoroughly. Two issues must be resolved to consider situation information: issue one, rating scarcity, is that there are less data to learn when importing more situations; issue two is that an effective approach is needed to adapt many situational factors. Our solution is a two-phase method: first, to overcome rating scarcity, data is augmented with estimations of unknown QoS values by learning from observable factors. The augmented data is then used to learn the important latent factors associated with the situational factors for QoS prediction. Experiments on data of real service invocations in different situations show improvement of our method in terms of QoS prediction accuracy over several existing methods, especially in the severe rating scarcity condition. In addition, analysis on parameter selection of proposed method can further assist in obtaining better QoS prediction in practical use.

Quality of Service (QoS) prediction has played an important role in service computing. However, in the real-world scenario of Web service, many user-observed QoS values are unknown and vary over time. In order to provide high accurate and efficient QoS prediction performance for Web services, we propose a personalized and accurate QoS prediction approach namely PAOMF. Our prediction model is built by employing matrix factorization and online stochastic gradient descent algorithm. Extensive experiments are conducted on real world public datasets, which demonstrate the effectiveness and efficiency of our proposed approach.

As a nonfunctional attribute of Web services, Quality of Service (QoS) is a crucial criterion that researchers utilize to develop personalized service recommendation. The QoS history is thin and contains noisy data because of the network environment and other inevitable factors. In order to increase the effectiveness of Web service recommendation, it is essential to accurately predict the missing QoS values. Therefore, a model for QoS prediction based on improved Conditional Variational Autoencoder is proposed in order to improve the prediction accuracy of missing QoS. The Bias based Singular Value Decomposition (BiasSVD) is employed by the model to pre-populate the initial sparse QoS matrix with missing values. By improved Conditional Variational Autoencoder, the pre-filled matrix is learned and the QoS is reconstructed. Then, the users' plausible similar neighbors are filtered using the Pearson Correlation Coefficient and Independent Sample T-test. The missing QoS is predicted by combining the reconstruction matrix. Last but not least, the experimental results on the actual dataset WSDream demonstrate that the model improves the QoS prediction accuracy to considerable degree.

Web Services Composition (WSC) is one of the main benefits of Service Oriented Architecture (SOA), however, its optimization is still a highly discussed research topic. In order to have an optimal composition, and therefore the best composite service, the composition mechanism must meet the requirements and issues of this process. Among the most important needs, we focus on the quality of web services involved, since the functional properties are not sufficient to determine the most appropriate service to meet the business goal. Researchers in this area propose different approaches and architectures to integrate Quality Of Service (QoS) values as a selection criterion, but they don't mention how to automatically generate those information and they don't consider the QoS values variations. They evaluate the QoS once and use calculated values in a static way. In our research work, we propose to automate the web service database filling. So, we don't have to manually read the WSDL file and extract necessary information for each operation. Moreover, we propose a QoS monitor component to calculate and update stored QoS values.

A personalized clustering-based and reliable trust-aware QoS prediction approach for cloud service recommendation in cloud manufacturing

Web service composition (WSC) is the task of combining a chain of connected single services together to create a more complex and value-added composite service. Quality of service (QoS) has been mostly applied to represent nonfunctional properties of web services and differentiate those with the same functionality. Many research has been done on QoS-aware service composition, as it significantly affects the quality of a composite service. However, existing methods are restricted to predefined workflows, which can incur a couple of limitations, including the lack of guarantee for the optimality on overall QoS and for the completeness of finding a composite service solution. In this paper, instead of predefining a workflow model for service composition, we propose a novel planning-based approach that can automatically convert a QoS-aware composition task to a planning problem with temporal and numerical features. Furthermore, we use state-of-the-art planners, including an existing one and a self-developed one, to handle complex temporal planning problems with logical reasoning and numerical optimization. Our approach can find a composite service graph with the optimal overall QoS value while satisfying multiple global QoS constraints. We implement a prototype system and conduct extensive experiments on large web service repositories. The experimental results show that our proposed approach largely outperforms existing ones in terms of solution quality and is efficient enough for practical deployment.

More recently, with the increasing demand of web services on the World Wide Web used in the Internet of Things (IoTs), there has been a growing interest in the study of efficient web service quality evaluation approaches through the use of prediction strategies to obtain accurate quality-of-service (QoS) values. The unpredictable network environment imposes very challenging obstacles to web service QoS prediction. Most of the traditional web service QoS prediction approaches are implemented only using a set of static model parameters with the help of designer's priori knowledge. Unlike the traditional QoS prediction approaches, our algorithm in this paper is realized by incorporating approximate dynamic programming (ADP)-based online parameter tuning strategy into the QoS prediction approach. Through online learning and optimization, the proposed approach implements the QoS prediction with automatic parameter tuning capability, and the prior knowledge or identification of the prediction model is not required. The near-optimal performance of QoS prediction can therefore be achieved. Experiment studies are carried out to demonstrate the effectiveness of the proposed ADP-based prediction approach.

Web service discovery (WSD) is the task of matchmaking a set of relevant web services. Quality of service (QoS) has recently been applied to represent non-functional properties of web services. Therefore, when those services provide the same functionality but have different QoS values, how to effectively filter out the services that cannot satisfy the QoS constraints and rank the remaining services is still an open research issue. In this paper, we propose an integrated approach that allows a service requester to specify a functionality request, multiple QoS constraints and their preferences, and our method discovers a set of the most appropriate ranked services with QoS utility aggregation. By conducting empirical experiments on simulated web services, we validate the feasibility of our service discovery approach. The running example shows that, our approach can find more appropriate services by the satisfiability of multiple QoS constraints and the ranking of aggregated QoS.

Constraint-based Quality of Service (QoS) prediction is a method for predicting violations of Service Level Agreements (SLAs) in an executing instance of a service orchestration. It uses assumptions about the ranges of QoS values for component services in the orchestration. Experiments suggest that the method, when given correct component QoS assumptions, produces highly accurate predictions according to a series of quality-of-prediction metrics, and that it does so well ahead of the time when the prediction is to happen. We study the behavior of this method when the component QoS assumptions become incorrect or too vague. We conclude that the effect is a graceful deterioration in prediction quality, unless gross (order-of-magnitude) imprecisions are introduced. However, the method is very sensitive to the loss of information on the lower bounds for component QoS values, since the knowledge of the upper bounds is not sufficient for failure prediction.