Kwei-Jay Lin

University of California, Irvine | UCI · Department of Electrical Engineering and Computer Science

In visual-based localization, prior research falls short in addressing challenges for the Internet of Things with limited computational resources. The dominant state-of-the-art models are based on separate feature extractors and descriptors without consideration of the constraints of small hardware, the issue of inconsistent image scale, or the pre...

Building accurate human behavior models is necessary for ambient intelligence. However, human activity recognition (HAR) in continuously monitored physical space faces challenges to achieve a good performance yet using only simple computing resources. In this work, we model HAR as an edge classification problem for a collaborative event graph of co...

Casting sequential recommendation (SR) as a reinforcement learning (RL) problem is promising and some RL-based methods have been proposed for SR. However, these models are sub-optimal due to the following limitations: a) they fail to leverage the supervision signals in the RL training to capture users’ explicit preferences, leading to slow converge...

Explicit and implicit knowledge about users and items have been used to describe complex and heterogeneous side information for recommender systems (RSs). Many existing methods use knowledge graph embedding (KGE) to learn the representation of a user-item knowledge graph (KG) in low-dimensional space. In this article, we propose a lightweight end-t...

Nowadays, AI has many applications in everyday human activities such as exercise, eating, sleeping, and automobile driving. Tech companies can apply AI to identify individual behaviors (e.g., walking, eating, driving), analyze them, and offer personalized feedback to help individuals make improvements accordingly. While offering personalized feedba...

Localization is an important technology for smart services like autonomous surveillance, disinfection or delivery robots in future distributed indoor IoT applications. Visual-based localization (VBL) is a promising self-localization approach that identifies a robot’s location in an indoor or underground 3D space by using its camera to scan and matc...

Real-time/online activity recognition (AR) is an important technology in smart Internet of Things (IoT) systems where users are assisted by smart devices in their daily activities. How to generate appropriate feature representation from sensor event streaming is a challenging issue for accurate and efficient real-time AR. Previous AR models that re...

Many advances have been introduced recently for service-oriented computing and applications (SOCA). The Internet of Things (IoT) has been pervasive in various application domains. Fog/Edge computing models have shown techniques that move computational and analytics capabilities from centralized data centers where most enterprise business services h...

This paper presents the edge intelligence support for smart Internet of Things (IoT) using the service-oriented architecture. We propose an edge intelligence framework for building smart IoT applications. The proposed edge intelligence framework pushes the streaming processing capability from cloud core to edge devices, in order to better support t...

Gait velocity has become a valid and important metric for senior populations. However, existing approaches to measure gait velocity are either limited to specific location or too expensive to be applied. IoT middleware allows the systems to collect data in a large space in a collaborative manner. However, it is a great challenge of integrating larg...

Social recommender system, using social relation networks as additional input to improve the accuracy of traditional recommender systems, has become an important research topic. However, most existing methods utilize the entire user relationship network with no consideration to its huge size, sparsity, imbalance and noise issues. This may degrade t...

Ubiquitous sensing and actuating devices are now everywhere in our living environment as part of the global cyber–physical ecosystem. Sensing and actuating capabilities can be modeled as services to compose intelligent Internet of Things (IoT) applications. An issue for perpetually running and managing these IoT devices is the energy cost. One ener...

Internet of Things (IoT) enables many physical devices to connect to Internet, including wearable devices, home appliances, ambient devices, etc. IoT application developers have created new services and applications to control actuators to make our lives easier. Since many applications may co-exist in a given environment, some of them may want to u...

Self-Management of services and devices in Internet of Things (IoT) systems and Machine-to-Machine (M2M) systems is a fundamental challenge due to the large number of devices and context-aware services. In this paper, we present the mechanism to automatically manage the services and devices according to context of the system and location of the dev...

The world is seeing more sensing and actuating devices deployed in our environment as part of the global digital ecosystem. One issue for perpetually running Internet of Things (IoT) devices is the energy efficiency. Many new IoT devices are running on powerful platforms that have ample computing and memory capacities to support multiple services....

This paper presents an on-line admission control methodology for periodic and aperiodic service processes with end-to-end real-time constraints. Both types of service process requests dynamically join and leave a system at run time. During the admission test, the schedulability of a periodic task is determined by using its fixed task capacity. Aper...

This paper presents an energy sentient methodology for mapping and deploying IoT applications on sensor devices installed in a target environment. The deployment decision support is part of the WuKong intelligent middleware which is designed to automatically discover and manage smart sensor devices. Given a pre-defined flow-based service process, W...

With the considerable growth of user-generated content, online reviews are becoming extremely valuable sources for mining customers’ opinions on products and services. However, most of the traditional opinion mining methods are coarse-grained and cannot understand natural languages. Thus, aspect-based opinion mining and summarization are of great i...

For advanced features of next generation vehicles, the real-time programs in automotive embedded systems are dramatically increasing. For such large volume program codes, this paper proposes a novel framework to use high-density and low-cost nonvolatile memory, i.e., NAND flash memory, as a low-cost means of storing and executing hard real-time pro...

Many SOA applications are now migrating to cloud servers, due to cloud's low cost provision, elasticity for growth, and better availability. Although cloud service providers have implemented reliable management system for their infrastructure, hybrid cloud users often lack easy monitoring and diagnosis facilities to discover the root cause of end-t...

To facilitate computing resources commercialization and to encourage more users to participate in the trading, we propose a computing resource market model CR-Market in this paper. Compared with existing Grid and Cloud markets, CR-Market has the following distinctive features: (1) Its architecture is based on the combination of Grid and Cloud. Mark...

Service-Oriented Architecture (SOA) is a software paradigm with a uniform means to offer, discover, compose services' capabilities to meet desired effects with measurable performance expectations. While most of today's SOA tools support service process compositions by considering only functional attributes, additional support for composing service...

This paper presents a project on building intelligent middleware on wireless sensor devices for machine-to-machine systems (M2M). Most current M2M and IoT applications are built for fixed sensor platforms with a specific wireless network support. Such software supports a unique set of sensors, and results in applications that are fragile and cannot...

This paper addresses the problem of resource reservation for applications using the real-time service-oriented architecture paradigm. Real-time services must be completed by their deadlines. They can be scheduled anywhere within an execution interval. Some services have a large execution interval which gives them more flexibility during admission c...

Most current sensor network applications are built for fixed sensor platforms with a specific wireless network support, building on top of a lightweight OS or even directly on the hardware, and writing applications in an imperative way and from each local node's perspective. This results in software that supports only a specific set of sensors, and...

This paper presents a robust scheduling framework for service processes with end-to-end deadlines. Service processes are admitted using capacity-based admission control, which assigns intermediate deadlines for individual services. At run-time, the system can respond to intermediate deadline faults by adjusting the deadlines assigned and reevaluate...

Service-Oriented Architecture (SOA) provides a framework for service composition and reconfiguration. In addition to making composition decisions by considering functional attributes, technology for composition by meeting non-functional requirements such as QoS is also needed. In this paper, we present an approach to compose business process by con...

With the growth of mobile applications, user context becomes a dynamic but important factor for quality of service (QoS) in the cyber world. Many digital services have distinctly different performance under different service contexts. At runtime, an unexpected transition of user context may cause an application to fail to deliver the required QoS....

Service-oriented architectures (SOA) are being adopted in a variety of industries. Some of them must support real-time activities. In this article, the authors present RT-Llama, a novel architecture for real-time SOA to support predictability in business processes. Upon receiving a user-requested process and deadline, our proposed architecture can...

Service flow in SOA systems need to detect quality of service (QoS) problems and to guarantee end-to-end performance. In previous work, we have proposed two faulty service identification methods: a dependency matrix based diagnosis and a Bayesian network based diagnosis. In this paper, we present a hybrid diagnosis to achieve high diagnosis accurac...

The automotive design paradigm is shifting from the "one-function on one ECU (Electronic Control Unit)" paradigm toward the "multi-function on one ECU" paradigm to reduce the ever increasing number of ECUs in a vehicle. In order to support such paradigm shift, this paper proposes a HW (hardware) resource componentizing technique that provides the i...

SService-oriented architectures (SOA) are being adopted in a variety of industries. Some of them must support real-time activities. In this paper, the authors present RT-Llama, a novel architecture for real-time SOA to support predictability in business processes. Upon receiving a user-requested process and deadline, our proposed architecture can r...

This paper presents an on-line admission control algorithm for one-time aperiodic service requests and periodically generated service processes with real time constraints. Both types of service requests dynamically come and leave the system at run time, making any off-line analysis infeasible. During admission test, periodic task schedulability is...

This paper presents a study on building intelligent middleware on wireless sensor network (WSN) for large-scale cyber-physical systems (LCPS). A large portion of WSN projects has focused on low-level algorithms such as routing, MAC layers, and data aggregation. At a higher level, various WSN applications with specific target environments have been...

When running a business process in SOA, systems need an efficient mechanism to detect performance issues and identify root causes. In this paper, we study the Bayesian network diagnosis model to identify faulty services in a business process by monitoring a subset of services selected as evidence channels. Both local and global optimal evidence cha...

This paper addresses the problem of resource reservation for applications built using the real-time service-oriented architecture (RT-SOA) paradigm. Real-time services can choose some time inside of a large interval for execution reservation. This property gives services more chance to be accepted by the schedulability admission control. However, t...

Service-oriented architectures (SOA) provide application systems the flexibility and cost-savings of dynamically composing workflows from reusable services. However, current SOA frameworks do not provide support for real-time workflow planning and execution. The goal of the RT-Llama SOA middleware framework is to address these new requirements. It...

Programs in hard real-time systems have stringent timing requirements. To ensure that the deadlines will always be met, real time systems must be designed to have flexible performances so that they can handle variable system loads. We describe a lan guage called Flex which makes it possible to implement real-time systems which may respond to dynami...

This paper studies a flexible scheduling data structure that is used to manage out-of-order service requests in real-time service-oriented architecture (RT-SOA) applications. We propose a hierarchical TBTree (hTBTree) to record server availability across a hierarchy of time intervals with different degrees of time granularity. Real-time services ne...

In this paper, we study the algorithms to estimate the response times of service executions during real-time service composition. By using reservation schedules to derive server utilization, we can compute a more realistic response time for each service than its static execution time. Reservation schedules are kept in a system data structure called...

Given the distributed and dynamic nature of SOC, and the autonomy of the agents involved, it is inevitable that failures will sometimes occur. Such failures might be due to simple errors or bugs, or may be a result of explicit malicious behaviour on behalf of an agent. In this chapter we introduce a number of future directions that will support the...

Service processes in SOA are composed dynamically by services from different service providers. At run-time, some services may become faulty and cause a service process to violate its end-to-end quality of service (QoS) constraints. We propose an effective approach for replacing only faulty services and some of their neighboring services to maintai...

Sustainability is an important issue with a growing interest. Two ICT technologies provide useful support for the sustainability of industrial systems: service-oriented architecture (SOA) and cyber-physical systems (CPS). SOA has been adopted in a variety of industrial systems due to its integration flexibility and process composability. CPS is a n...

For the last decade, the area of e-business and e-service has expanded rapidly. Service Oriented Architecture (SOA) is introduced to create and execute a business process on the Web. In order to create and execute the business process, relevant web services need to be composed or reconfigured based on the business process logic. Traditionally, web...

Enabled by Service-Oriented Architecture (SOA), recently Software as a Service (SaaS) and Cloud computing are gaining momentum in the industry. An open issue is how to ensure accountability in business services offered through Internet. Traditionally a contract is an effective legal means to uphold accountability in business transactions. In this p...

Service-oriented architectures SOA are being adopted in a variety of industries. Some of them must support real-time activities. In this article, the authors present RT-Llama, a novel architecture for real-time SOA to support predictability in business processes. Upon receiving a user-requested process and deadline, our proposed architecture can re...

In the business context, accountability has become a major concern for businesses around the world in aftermath of corporate scandals and fallouts. However, accountability has not been rigorously considered in IT system technologies and solutions. The goal of this study is to provide a clear understanding of accountability concept in service-orient...

Service science, management and engineering (SSME) research is to study the methodology and technology for service innovation, design, development and delivery. Since service industry is very quality-sensitive and trust-dependent, we propose a service accountability management framework to detect, diagnose, defuse and disclose the root cause for an...

One of the challenges in SOA is to provide an efficient and effective way to identify compatible Web service(s) and compose them into a service process that satisfies a userpsilas functional and non-functional needs. Quality of service (QoS), like response time or throughput, might need to be optimized for the selected service process. This paper d...

Abstract—Service-oriented architectures (SOA), though widely accepted in a variety of industries, must be en- hanced,to support,real-time activities in order to gain even greater adoption. We present RT-Llama, a novel architecture for real-time SOA to support,predictability in business processes. Based on a user-specified process and deadline, our...

Using SOA, service processes can be composed statically or dynamically using services provided by different service providers. Some services may become faulty at runtime and cause the service process to violate the end-to-end quality of service (QoS) constraint. We propose an efficient approach for replacing faulty services to ensure that the recon...

In SOA, services may become volatile and fail to deliver the quality of service as requested by users. In this paper, we present an approach for repairing failed services by replacing them with new services and ensuring the new service process still meets the user specified end-to-end QoS constraints. An iterative structural inspection algorithm is...

In service-oriented computing (SOC) environments, service clients interact with service providers for services or transactions. From the point view of service clients, the trust status of a service provider is a critical issue to consider, particularly when the service provider is unknown to them. Typically, the trust evaluation is based on the fee...

The Intelligent Accountability Middleware Architecture (Llama) project supports dependable service-oriented architecture (SOA) monitoring, runtime diagnosis, and reconfiguration. At its core, Llama implements an accountability service bus that users can install on existing service-deployment infrastructures. It collects and monitors service executi...

When a business process fails to deliver the quality of service (QoS), it is necessary to identify the faulty services that cause the problem. In this paper, we present a framework for diagnosing QoS problems in SOA. A set-covering algorithm is used to select the locations of run-time service data collection, or probes. The framework then creates a...

Service-oriented architecture (SOA) provides the flexibility of dynamically composing business processes in enterprise computing. However, they must be enhanced to support real-time activities in future SOA applications such as media streaming, control applications, cyber-physical, and intelligent vehicle systems. In this paper, we present the RT-L...

In this chapter, the authors present a novel perspective by using the Creative Commons (CC) licensing model to compare 10 commonly used OSS licenses. The authors also propose a license compatibility table to show that whether it is possible to combine OSS with CC-licensed open content in a creative work. By using the CC licensing concept to interpr...

In this chapter, the authors present a novel perspective by using the Creative Commons (CC) licensing model to compare 10 commonly used OSS licenses. The authors also propose a license compatibility table to show that whether it is possible to combine OSS with CC-licensed open content in a creative work. By using the CC licensing concept to interpr...

Enterprises are turning to SOA for the flexible deployment of business processes. While current monitoring tools can detect service execution problems in an enterprise’s servers and report such problems to human decision makers, they may not closely monitor the external services they use, diagnose the root cause of process problems, and automatical...

For the last decade, the areas of e-business and e-service have expanded rapidly. Service oriented architecture (SOA) is introduced to create and execute business process on the Web. In SOA, the importance of managing the reputation ofservices has emerged. Thus, there has been much research to evaluate and manage the reputation of services. This pa...

The flexibility and dynamism of service-oriented architecture (SOA) makes it critical to monitor and manage services behaviors at runtime for performance assurance. In this paper, two efficient evidence channel selection (ECS) algorithms are designed to select service run-time data collection locations for business process management. The design of...

One of the most well-studied scheduling algorithms for real-time systems is the Rate Monotonic (RM) scheduling for periodic tasks. In this paper we derive a generalized RM schedulability bound by considering relative period ratios among tasks in a system. We show that schedulability bounds published earlier are special cases of our generalized boun...

EDZL (Earliest Deadline first until Zero Laxity) is an efficient and practical scheduling algorithm on multiprocessor systems. It has a comparable number of context switch to EDF (Earliest Deadline First) and its schedulable utilization seems to be higher than that of EDF. Previously, there was a conjecture that the utilization bound of EDZL is 3m/...

The reputation-oriented trust issue is critical to e-commerce applications and has drawn much attention from both industry and the research community. Existing e-commerce systems have introduced trust management mechanisms that provide some rating information to customers. However, more comprehensive mechanisms should be provided to more precisely...

Service-oriented architecture (SOA) provides a flexible paradigm to dynamically compose service processes from individual services. The flexibility, on the other hand, makes it necessary to monitor and manage service behaviors at runtime for performance assurance. One solution is to deploy software monitoring agents. In this paper, we present an ap...

The development of mobile services becomes more challenging when context-awareness is taken into account. There are many problems in designing a successful context-aware mobile SOA. In this paper, we introduce the model of "dynamic capabilities" to study how systems can dynamically integrate, compose, and reconfigure internal and external competenc...

One of the challenges in SOA is to provide an efficient and effective way to select compatible Web service(s) and compose them into a functional service process that satisfies a user's needs. Moreover, the Web services(s) selected may be integrated with structures such as branches and merges to fulfill a task. This paper describes our system archit...

Real-time systems using rate-monotonic fixed priority scheduling can be checked for schedulability either by sufficient but pessimistic schedulability conditions or by exact testing. Exact testing provides a more precise result but may not be performed in polynomial time. Audsley et al. proposed one of the earliest methods by iteratively deriving t...

The goal of our service science and engineering research is to advance IT-based services technology for service innovation, composition and delivery. In this project, we study the service accountability framework to detect, diagnose, and defuse the real problem in a service process. This is important when a service process has external service prov...

E-commerce is a big business with a growing market size and has been a major driving force in the IT industry for the past decade. Companies now need to provide online shopping or marketing Web presence to allow for direct customer connections. In this article, the author reviews some primary e-commerce technologies, including auctions, negotiation...

As it lacks central management in Peer-to-Peer (P2P) e-commerce environments, prior to new transactions with an unknown peer, the trust evaluation is critical, which relies on the transaction history data. Tradi- tionally the evaluation process is based on other peers' recommendations neglecting transaction amounts. This may lead to the bias in the...

Accountability is a composite measure for different but related quality aspects. To be able to ensure accountability in practice, it is required to define specific quality attributes of accountability, and metrics for each quality attribute. In this paper, we propose a quality detection and diagnosis framework for the service accountability. We fir...

As more enterprises turn to SOA for the flexible deployment of business processes, adaptable and highly configurable infrastructures can be used to ensure that end-to-end quality of service requirements are met via automatic and dynamic process monitoring, diagnosis, and routing. We present the intelligent accountability middleware architecture (LL...

In both e-commerce (EC) and service-oriented computing (SOC) environments, sellers or service providers interact with customers or service clients for services or transactions. From the point view of customers or service clients, the trust status of the seller or service provider is a critical issue to consider, particularly when the seller or serv...

Many optimal uniprocessor schedulers, such as earliest deadline first (EDF) and rate monotonic (RM), do not have a good schedulability bound on multiprocessor systems. In this paper, we study an on-line algorithm earliest deadline first until Zero laxity (EDZL) for multiprocessor systems. A set of tasks scheduled by EDZL is scheduled using EDF unti...

Feasibility and schedulability problems have received considerable attention from the real-time systems research community in recent decades. Since the publication of the Liu and Layland bound, many researchers have tried to improve the schedulability bound of the RM scheduling. The LL bound does not make any assumption on the relationship between...