Search

The ubiquity of the Internet of Things (IoT) across diverse applications underscore its pivotal role in seamlessly integrating physical devices to facilitate efficient data collection, analysis, and automation. Consequently, ensuring the Quality of Service (QoS) emerges as a critical imperative. While numerous studies have proposed methodologies focusing on resource optimization, network management, and data processing techniques, several previous approaches to QoS prediction may face constraints in the Internet of Things (IoT) environment, where the dynamic nature of IoT environments demands predictive models capable of adapting to varying conditions. Integrating user and service location data into QoS prediction models is paramount, as it enables personalized service delivery tailored to the user's specific context, thus enhancing the user experience and overall system performance. This paper presents a novel approach to QoS prediction in IoT, harnessing self-attention and collaborative filtering (CF) techniques while incorporating location-based features such as distance from the user to the service. Experimental evaluations on benchmark datasets reveal that our proposed model improves prediction accuracy by up to 7 % compared to existing methods, underscoring its efficacy in enhancing QoS provisioning in IoT environments. [ABSTRACT FROM AUTHOR]

Choosing appropriate Web services to create new applications plays a significant role for service-based development. The general service recommendation approaches usually prefer to suggest popular and frequently used services. However, nowadays users are becoming increasingly interested in long tail services, which are novel but less concerned. Unfortunately, few of the existing service recommendation methods explicitly takes such long tail services into consideration, leading to the poor diversity and serendipity of recommendation results. Compared with the general service recommendation, the major challenges of recommending long tail services come from: (1) more sparse historical invocation records, and (2) lower quality of textual service description. To address the above challenges, in this paper, we put forward a method for Long tail Service Recommendation based on Cross-view and Contrastive learning (LSRCC). More specifically, we propose a text view that employs recursive neural network, multilayer perceptron and attention mechanism to learn the semantic representations of application and service in a more precise way. In addition, we present an invocation view that employs LightGCN (Light Graph Convolution Network) and local level contrastive learning to capture the implicit connections among applications, services and their tags. Finally, we introduce global level contrastive learning to further integrate the above two views for the long tail service recommendation. The extensive experiments on real-world datasets demonstrate that LSRCC outperforms the state-of-the-art baselines in terms of diversity and tail coverage, with the satisfactory precision and recall, in long tail service recommendation. • A long tail service recommendation method is proposed for application creation. • Contrastive learning is used globally and locally to capture implicit connection. • Double views are used to mitigate the problem of data sparsity and inferiority. • Extensive experiments on the real datasets show its superiority over baselines. [ABSTRACT FROM AUTHOR]

Managers in electric utilities face the disruption of their conventional business model of selling electricity per kilowatt-hour for invariant prices. However, the forthcoming widespread uptake of sustainable energy technologies – such as rooftop solar, batteries, heat pumps and electric vehicles – by residential customers also represents a chance for local utilities to diversify their service portfolio. To appropriately market these technologies to households, utilities need data on consumers. In this paper, we present a novel data-driven service bundle recommendation model incorporating technologies and tariffs for residential customers based on individual household data. We validate the model in a case study and quantify the utility of sharing different levels of household data. We find substantial synergies of flexible sustainable technologies and time-varying tariffs, leading to higher cost reductions for customers than tariff-switching alone that can be recommended based on easy-to-obtain data. This demonstrates a large potential for energy service bundle marketing by local utilities. The presented Machine Learning recommendation models enable more reliable recommendations than a naive benchmark. Our research thus demonstrates the potential of data-driven utility marketing strategies that focus on service bundling and the integration of customers' energy consumption data. [ABSTRACT FROM AUTHOR]

Highlights • Service recommendation for SBS (service-based system) development. • Measuring functional similarity between SBSs using word embeddings. • Two recommendation scenarios are identified for different SBS requirements. • Experiments on a real-world dataset show the effectiveness of our approach. Abstract With the wide adoption of service-oriented computing and cloud computing, service-based systems (SBSs), a kind of software systems that can offer certain functionalities by leveraging one or more Web services, become increasingly popular. A challenging issue in SBS development is to find suitable services from a variety of available (semantics different) services. Towards this issue, we propose a new service recommendation approach that can integrate diverse information of SBSs and their component services. In this research, SBSs, services, their respective attributes (e.g. content and categories) and SBS-service composition relations are modeled as a heterogeneous information network (HIN); and several semantic similarities between SBSs are measured on a set of meta-paths in the HIN. Particularly, a word embedding technique is used to learn word vectors from the content of SBSs and services, which contribute to better functional similarities between SBSs. Afterwards, the combinational weights of different similarities are optimized using a Bayesian personalized ranking algorithm. Services are finally recommended based on collaborative filtering. We identify two recommendation scenarios with different SBS requirements. By conducting a series of experiments on a real-world dataset crawled from the ProgrammableWeb, we validate the effectiveness of our approach and find out the optimal combinations of SBS similarities for those two scenarios. [ABSTRACT FROM AUTHOR]

The integration of coherent services plays a potential role in the field of Service Oriented Applications. Achieving this potential standard crucially depends on the ability to recognize and exploit the available services based on user requirements. In general, the user preferences on Quality of Service (QoS) requirements are fuzzy in nature. In addition to that, the QoS requirements are user dependent even if the functional requirements are the same. With a large number of available services, service selection for dynamic composition at run time is a challenge. Functional and non-functional assumptions made at design time may violate at run-time. These violations require run time reaction, by adopting a run-time process. Therefore, dynamic and fuzzy QoS-aware service discovery for run-time composition and continuous adaptation is a strong requirement in service oriented computing. Considering that different users follow different fuzzy reasoning in various contexts at different times, a fuzzy inference based service selection approach has been proposed in this paper. Continuous adaptation is done, as and when a design time assumption violation is reported by a run-time monitoring system. We have implemented and tested the proposed approach and the results show its effectiveness. [ABSTRACT FROM AUTHOR]

With the rapid progress of Mobile Internet of Things (MIoT), Location-Based Services (LBSs) have become widely popular due to the deployment of diverse devices and facilities. The utilization of spatial–temporal information has led to a considerable improvement in service recommendation in the MIoT-enabled service platform. In practical scenarios, the interaction between the user and the service is often initiated by the joint effect of previous service invocations. Many-to-many and high-order spatial–temporal relations exist among services, which are inadequately reflected by prevailing methods that rely on ordinary graphs and consider spatial–temporal relations among services as pairwise connections. Additionally, most methods indiscriminately conflate the separate influences of spatial and temporal information on user preferences, which can lead to suboptimal service recommendation performance. To this end, we propose an approach powered by spatial–temporal hypergraphs (STH) for personalized service recommendation in the MIoT-enabled service platform, which is inspired by the deep learning paradigm. Specifically, we first construct spatial and temporal hypergraphs based on spatial–temporal information. Then, we develop the spatial–temporal hypergraph neural network-based encoders to learn spatial and temporal representations of services. Finally, we introduce an auxiliary task to differentiate between distinct spatial–temporal effects for improved service recommendations. We conduct extensive experiments on four real-world datasets, and the results demonstrate the effectiveness of STH. Thus, STH not only enhances the user experience but also improves service management in the MIoT-enabled service platform. [ABSTRACT FROM AUTHOR]

Deep Neural Networks (DNN) based collaborative filtering has been successful in recommending services by effectively generalizing graph-structured data. However, most existing approaches focus on first-order interactions. Although recent approaches have utilized high-order connectivity, they still limit themselves to simple interactions and ignore the pattern of structural sub-graphs/motifs. In this study, we first explore the commonly used motifs in the Mashup-API interaction bipartite graph and propose a dedicated algorithm to generate the motif adjacency matrix. We then propose a Motif-based Graph Attention Network for service recommendation (MGSR) that utilizes a motif-based attention mechanism to capture the high-order information of various motifs, and a Collaborative Filtering model to generate the recommendation prediction. We have conducted extensive experiments on ProgrammableWeb dataset and our results demonstrate the superior performance of our proposed framework over some state-of-the-art approaches. [Display omitted] • We report a Motif-based Graph Attention Network for service recommendation (MGSR). • The model aggregates high-order information of motif-based neighbors. • The model identifies all the seven up-to-four-node motifs in bipartite graphs. • The model learns the weights of different nodes from different motifs. [ABSTRACT FROM AUTHOR]

Under the background of the continuous upgrading of customers' personalized needs, enterprises will carry out a series of transformation and upgrading into mass personalization model to adapt to the flexible and changeable market environment, and meet customers' needs. Moreover, the service for customers based on Customer-product Interaction Life Cycle (CILC) is a new innovation service, which is gradually evolved from the pursuit of general functional requirements of products to the smart product service system (SPSS) based on customer participation in product manufacturing and customer-product interaction. Therefore, this paper studies the innovation service system for CILC in SPSS under the new model of mass personalization. Firstly, a comprehension service system for CILC in SPSS is proposed to provide the top-level framework and decision-making of service for enterprises. Secondly, a personalized service recommendation for CILC in SPSS based on improved collaborative filtering algorithm is proposed to realize the recommendation, expansion and value-added of service resources for enterprises. Thirdly, a calculation example of personalized service recommendation for CILC in SPSS is giving. The service system for CILC in SPSS proposed in this paper is an extension and supplement to the theory of SPSS. At the same time, the research of this paper can give a reference for enterprise to plan, set, select, carry out, and maintain service items for CILC in SPSS. • A personalized service recommendation method is proposed. • It is an extension and supplement to the theory of SPSS. • It can give a reference for enterprise to carry out CILC. • It has a certain reference value for industry and government. [ABSTRACT FROM AUTHOR]

Internet of Things (IoT) is the next wave of industry revolution that will initiate many services, such as personal health care and green energy monitoring, which people may subscribe for their convenience. Recommending IoT services to the users based on objects they own will become very crucial for the success of IoT. In this work, we introduce the concept of service recommender systems in IoT by a formal model. As a first attempt in this direction, we have proposed a hyper-graph model for IoT recommender system in which each hyper-edge connects users, objects, and services. Next, we studied the usefulness of traditional recommendation schemes and their hybrid approaches on IoT service recommendation (IoTSRS) based on existing well known metrics. The preliminary results show that existing approaches perform reasonably well but further extension is required for IoTSRS. Several challenges were discussed to point out the direction of future development in IoTSR. [ABSTRACT FROM AUTHOR]

Given the increasing applications of service computing and cloud computing, a large number of Web services are deployed on the Internet, triggering the research of Web service recommendation. Despite of service QoS, the use of user feedback is becoming the current trend in service recommendation. Likewise in traditional recommender systems, sparsity, cold-start and trustworthiness are major issues challenging service recommendation in adopting similarity-based approaches. Meanwhile, with the prevalence of social networks, nowadays people become active in interacting with various computers and users, resulting in a huge volume of data available, such as service information, user-service ratings, interaction logs, and user relationships. Therefore, how to incorporate the trust relationship in social networks with user feedback for service recommendation motivates this work. In this paper, we propose a social network-based service recommendation method with trust enhancement known as RelevantTrustWalker. First, a matrix factorization method is utilized to assess the degree of trust between users in social network. Next, an extended random walk algorithm is proposed to obtain recommendation results. To evaluate the accuracy of the algorithm, experiments on a real-world dataset are conducted and experimental results indicate that the quality of the recommendation and the speed of the method are improved compared with existing algorithms. [ABSTRACT FROM AUTHOR]

Abstract: With the growing interests in semantic web services and context-aware computing, the importance of ontologies, which enable us to perform context-aware reasoning, has been accepted widely. While domain-specific and general-purpose ontologies have been developed, few attempts have been made for a situation ontology that can be employed directly to support activity-oriented context-aware services. In this paper, we propose an approach to automatically constructing a large-scale situation ontology by mining large-scale web resources, eHow and wikiHow, which contain an enormous amount of how-to instructions (e.g., “How to install a car amplifier”). The construction process is guided by a situation model derived from the procedural knowledge available in the web resources. Two major steps involved are: (1) action mining that extracts pairs of a verb and its ingredient (i.e., objects, location, and time) from individual instructional steps (e.g., ) and forms goal-oriented situation cases using the results and (2) normalization and integration of situation cases to form the situation ontology. For validation, we measure accuracy of the action mining method and show how our situation ontology compares in terms of coverage with existing large-scale ontology-like resources constructed manually. Furthermore, we show how it can be utilized for two applications: service recommendation and service composition. [Copyright &y& Elsevier]

Abstract: In a migrating workflow, agents transfer their code (specification) and their execution state to a site, negotiate services to be executed on their behalf, receive the results, and move on. The next place visited by agents, and the next service requested, is determined by both the goals of the workflow and the results of the current requests. However, the primary assumption in such workflow specifications is that the workflow designer has the ability of knowing the perfect knowledge of all the details of the process, and predicting changes of network configuration. In this paper, an efficient service recommendation strategy is suggested for migrating workflows. Services organization in our suggested strategy is based on the differential evolutionary contract net protocol. To evaluate the performance improvement achieved by our strategy, this paper uses F1 that is a well-known performance measure for recommender system. The results show that the new strategy can reduce the communication overhead and substantially improve performance in terms of global optimization and system utilization. [Copyright &y& Elsevier]

As a new computing technology proposed with the development of 5G, IoT technologies and increasing requirement of mobile applications and services, edge computing enables mobile application developers and content providers to serve context-aware mobile services (e.g., mobile app recommendation). Mobile app recommendation is known as an effective solution to overcome the information overload in mobile app markets. Most existing models only consider user-app interaction and feature modeling, and neglect the structural information which actually is a crucial part in the scenario of app recommendation. To fully exploit both structural and feature information for app recommendation, this paper proposes a novel heterogeneous graph neural network framework (HGNRec) including one inner module and one outer module. Specifically, the inner module is able to use a node-level attention to learn the importance between a node and its meta-path based neighbors. The outer module with a path-level attention can learn the importance of different meta-paths. With the learned importance from two modules, the comprehensive embeddings for user and app nodes can be generated by integrating features from meta-path based neighbors. Extensive experiments on the real-world Google Play mobile app dataset demonstrate the effectiveness of HGNRec. • HGNRec is the first attempt to model mobile app recommendation under the heterogeneous graph neural network framework. • Hierarchical attention layers can obtain the significances of different neighbor nodes and meta-paths. • HGNRec performs best when both accuracy and efficiency are considered. • Structural information with more heterogeneous relationships leads a superior performance. [ABSTRACT FROM AUTHOR]