Keman Huang

An Empirical Study of Programmable Web: A Network Analysis on a Service-Mashup System

A service ecosystem consists of services and their compositions (i.e., mashups) and evolves as a complex network system. It is driven by continuously emerged new services and the mashups of old services and new ones. Complex network analysis can be a powerful tool to study the static structure as well as the evolution of a service ecosystem. This paper presents a methodology to study such a system and an empirical study of Programmable Web. To the best of our knowledge, Programmable Web is the largest and most active Web APIs and mashups collection and consists of 4337 services and 6092 service compositions by Nov-2011. We conduct a comprehensive network analysis to quantitatively characterize the static structure and dynamic evolution of the ecosystem. The findings of this paper not only can help understand the current usage pattern and the evolution trace of the ecosystem, but also are applicable to other Web service systems.

Recommendation in an Evolving Service Ecosystem Based on Network Prediction

Service computing plays a critical role in business automation and we can observe a rapid increase of web services and their compositions nowadays. Web services, their compositions, providers, consumers, and other entities such as context information, collectively form an evolving service ecosystem. Many service recommendation methods have been proposed to facilitate the use of services. However, existing approaches are mostly based on all-time statistics of usage patterns, and overlook the temporal aspect, i.e., the evolution of the ecosystem. As a result, recommendation may consist of obsolete services and also does not reflect the latest trend in the ecosystem. In order to overcome this limitation, we propose an innovative three-phase network prediction approach (NPA) for evolution-aware recommendation. First, we introduce a network series model to formalize the evolution of the service ecosystem and then develop a network analysis method to study the usage pattern with a special focus on its temporal evolution. Afterward a novel service network prediction method based on rank aggregation is proposed to predict the evolution of the network. Finally, using the network prediction model, we present how to recommend potential compositions, top services and service chains, respectively. Experiments on the real-world ProgrammableWeb data set show that our method achieves a superior performance in service recommendation, compared with those that are agnostic to the evolution of a service ecosystem.

Service Recommendation in an Evolving Ecosystem: A Link Prediction Approach

Services computing is playing a critical role in recent years in many fields and we observe a rapidly increasing number of web accessible services and their compositions nowadays. However, our earlier empirical study reveals that, overall the public available services are under-utilized, and when they are used, they are used mostly in an isolated manner. This phenomenon inspires us to further explore a methodology to help consumers understand the usage pattern of the service ecosystem, including interactions among services, and the evolution of these interactions. Based on the derived usage pattern, this methodology also introduces a service recommendation method that suggests both services and their compositions, in a time-sensitive manner. We firstly construct an evolution network model from the historical usage of the services in the ecosystem. Then a rank-aggregation-based link prediction method is proposed to predict the evolution of the ecosystem. Based on this link prediction method, we can recommend services and compositions of interest to service developers. Through an experiment on the real-world mashup-service ecosystem, i.e., Programmable Web, we demonstrated that our approach can effectively recommend services and compositions with better precision than the methods we compared.

Mirror, Mirror, on the Web, Which Is the Most Reputable Service of Them All?

With the wide adoption of service and cloud computing, nowadays we observe a rapidly increasing number of services and their compositions, resulting in a complex and evolving service ecosystem. Facing a huge number of services with similar functionalities, how to identify the core services in different domains and recommend the trustworthy ones for developers is an important issue for the promotion of the service ecosystem. In this paper, we present a heterogeneous network model, and then a unified reputation propagation URP framework is introduced to calculate the global reputation of entities in the ecosystem. Furthermore, the topic model based on Latent Dirichlet Allocation LDA is used to cluster the services into specific domains. Combining URP with the topic model, we re-rank services reputations to distinguish the core services so as to recommend trustworthy domain-aware services. Experiments on ProgrammableWeb data show that, by fusing the heterogeneous network model and the topic model, we gain a 66.67% improvement on top20 precision and 20%~ 30% improvement on long tail top200~top500 precision. Furthermore, the reputation and domain-aware recommendation method gains a 118.54% improvement on top10 precision.

BSNet: a network-based framework for service-oriented business ecosystem management

As enterprises turning to SOA, services‐oriented business ecosystem (SOBE) has become an important pattern for the organization and management of the massive business services. At the same time, the emergence of Internet of Services (IOS) provides a business model in which service vendors and consumers can interact with each other via the Internet. This paradigm makes it possible that the services in SOBE are managed in an autonomous and coordinated manner. The challenge here is to organize these massive business services, coordinate, and federate them to achieve the benefits of SOA. To address these challenges, this paper presents BSNet, a framework on the basis of the service correlation networks to manage the SOBE. The model consists of a who‐what‐how service correlation network that captures the various relations in SOBE. Finally, a prototype system is developed, and a simulated case study is provided to show the expanded value of our network‐based framework. Copyright

Service Value Broker Patterns: An Empirical Collection and Analysis

The service value broker(SVB) pattern integrates business modeling, knowledge management and economic analysis with relieved complexity, enhanced reusability and efficiency,etc. The study of SVB is an emerging interdisciplinary subject which will help to promote the reuse of knowledge, strategy and experience in service based designs and solutions. In this paper, we focus on enumerating collected SVBs empirically with initial analysis on their composition manners. The results from this paper will play a dominating role in fueling a coming E-service Economics era.

Failure analysis and tolerance strategies in web service ecosystems

Service‐oriented computing and cloud computing are playing critical roles in supporting business collaboration over the Internet. Thanks to the latest development in computing technologies, various large‐scale, evolving, and rapidly growing service ecosystems emerge. However, service failures greatly hamper the usability and reputation of service ecosystems. In the previous work, service failure is not adequately studied from an ecosystems perspective. To address this gap, we propose a service failure analysis framework based on a complex network model of service ecosystem. This framework comprises a feature model of failed services and several service failure impact indicators. By applying the framework, empirical analysis of failed service features and failure impact assessment can be implemented more easily and precisely. Moreover, to provide failure tolerance strategies for service ecosystems, a novel composition‐based service substitution method is designed to replace the failed services with functional similar ones, such that the service systems are more robust when a failure occurs. As the new substitution method requires fewer structural data of services, it is more convenient to be applied in present RESTful Representational State Transfer (REST) service environment. Both the framework and the service substitution method are tested on real‐world data set, and their usability and efficiency are demonstrated. Copyright

Towards Value-Driven Business Modelling Based on Service Brokerage

Service engineering is an emerging interdisciplinary subject which crosscuts business modeling, knowledge management and economic analysis. To satisfy service providers’ profiting goals, the service system modeling needs to take care of both the short and long run customer satisfaction. The ideology of value driven design fits well for this need. We propose to work towards value driven design by introducing a form of service design patterns, we call service value broker(SVB), with the aim to shorten the distance between economical analysis and IT implementation and increase the value added on all sides. SVB allow us to not only study the value added in terms of functional and business aspects, but also reason about the need for brokerage across various domains. In this paper, we model the basis of SVB and its network based organization architecture in the background of Cloud.

Domain-Aware Service Recommendation for Service Composition

Service compositions inherently require multiple services each with its domain-specific functionality. Therefore, how to mine matching patterns between services in relevant domains and compositions becomes crucial to service recommendation for composition. Existing methods usually overlook domain relevance and domain-specific matching patterns, which restrict the quality of recommendations. In this paper, a novel approach is proposed to offer domain-aware service recommendation. First, a K Nearest Neighbor variant (vKNN) based on topic model Latent Dirichlet Allocation (LDA) is introduced to cluster services into semantically coherent domains. On top of service domain clustering results by vKNN, a probabilistic matching model Domain Router (DR) based on Extreme Learning Machine (ELM) is developed for decomposing a requirement to relevant domains. Finally, a comprehensive Domain Topic Matching (DTM) model is built to mine relevant domain-specific matching patterns to facilitate service recommendation. Experiments on a large-scale real-world dataset show that DTM not only gains significant improvement at precision rate but also enhances the diversity of results.

Impacts of Scheduling Algorithms in Services on Collective End-to-End Execution Time Characteristics of Web Service Workflows

Web services usually compose to workflows to satisfy complex demands. End-to-end execution time is widely seen as a key quality metric of web service workflows. That will be affected by many factors. This paper focuses on impacts of an important factor -- scheduling algorithm in services -- on collective end-to-end time characteristics of a set of web service workflows. We develop a novel simulator, in which workflows are simulated to execute. Impacts of different scheduling algorithms are evaluated through comparing simulation results. Simulation results indicate that maximal and average end-toend execution time of most workflows when using earliest deadline first (EDF) scheduling algorithm in services is significant shorter than that when using widely used first in, first out (FIFO) scheduling algorithm.