Knowledge base question answering (KBQA) aims to answer natural language questions using large-scale knowledge bases (KBs). Among various KBQA approaches, semantic parsing-based (SP-based) methods have demonstrated strong effectiveness by generating concise logical forms (LFs) that capture complex subgraph structures and semantic information. Recent research suggests that integrating large language models (LLMs) with SP can achieve significant improvements in the performance and efficiency of KBQA by facilitating the direct generation of LFs with minimal retrieval. However, generating complete LFs with LLMs continues to pose a challenge due to the complexity of the required graph structures and constraints, leading to the significant issue of non-executability. To address these challenges, we propose GCA-KBQA, a step-wise fine-tuned LLM-based framework that employs hop-wise generation, knowledge-assisted calibration, and path-level assembly to construct complete LFs for KBQA. Specifically, we decompose the complex SP process into manageable steps: first, we iteratively generate LFs for each topic entity one hop at a time using a fine-tuned LLM, leveraging KB knowledge to calibrate intermediate outputs and mitigate error propagation. Subsequently, we guide the LLM in assembling path-level LFs from different topic entities, resulting in optimized final LF. We evaluate the proposed method on four KBQA benchmarks spanning two distinct KBs, demonstrating its superior performance compared to state-of-the-art baselines. The code is available at https://github.com/pvfeldt/GCA-KBQA.

Drone swarms are increasingly important in IoT applications such as agriculture, disaster response, and industrial inspection. However, effective service discovery remains challenging due to drones’ limited resources and the swarm’s dynamic topology. Existing solutions often suffer from congestion, single points of failure, and poor adaptability. To overcome these limitations, we propose a decentralized and dynamic service discovery framework tailored for drone swarms. Our approach models services using fine-grained sensor-level decomposition and leverages spatiotemporal information from drones to enable timely coordination. The core of the framework is a two-phase affinity propagation mechanism: a fully distributed clustering phase based on spatiotemporal leadership to provide a decentralized service registry, followed by a local adaptation phase for dynamic registry updates. To enhance reliability, a spatiotemporal-driven priority chain is used for service replication and failover. Extensive simulations and a case study in a wildfire suppression scenario across various swarm sizes show that our framework significantly outperforms centralized and existing clustering-based methods in efficiency, robustness with limited resources. This makes it a promising solution for reliable service discovery and flexible, fine-grained collaboration in drone swarms.

Knowledge base question answering (KBQA) refers to the task of answering natural language questions using factual information from large-scale knowledge bases (KBs). To obtain accurate answers, recent research optimizes semantic parsing methods, a major KBQA approach, with large language models (LLMs), where concise logical forms (LFs) are generated by LLMs and executed in KBs. Although these methods demonstrate superior performance, they still encounter the problem that some generated LFs fail to yield answers when executed, significantly limiting their effectiveness. To mitigate this issue, we propose KARV, a Knowledge-Assisted reasoning path Reconstruction and hierarchical Voting approach for non-executable LFs. This method extracts semantic knowledge from KBs as guidance to correct and reconstruct reasoning paths, deriving answers through a voting-based strategy. The insight is that non-executable LFs generated by LLMs still contain rich semantic information, and the knowledge retrieved from KBs can effectively correct them. Specifically, we fine-tune LLMs to generate high-quality LFs, and the nonexecutable LFs are decomposed into multiple path branches based on mentioned entities. Semantic knowledge from KBs is then leveraged to correct the entities and relations within these branches, effectively reconstructing the reasoning paths. To obtain precise final answers, we apply a hierarchical voting strategy both within and across the non-executable LFs. Our proposed method achieves state-of-the-art performance on benchmarks including WebQuestionSP (WebQSP), ComplexWebQuestions (CWQ), and FreebaseQA.

Deep neural network(DNN) services are now widely used in industrial defect detection applications. With the increasing number of pre-trained model services on MaaS platforms like HuggingFace and inside smart enterprises, fine-tuning or directly applying DNN services has become a new solution for building intelligent applications. However, selecting appropriate services for tasks with various industrial requirements is also challenging work. Existing DNN model recommendation systems typically categorize models based on a limited set of task types or leverage the training data similarities. However, they fail to reflect the DNN service's native transferability and dynamic ability in the specific industrial scenario (i.e., pragmatics). In this paper, we introduce DSR, a novel pragmatic-information-model-based DNN service recommendation approach, designed to retrieve the most suitable services by incorporating information across the scene of industrial tasks and the ability of services. Through graph convolutional networks, DSR embeds the pragmatic information model of services into unified vectors and applies a regression model for usefulness-oriented recommendation towards specific industrial tasks. Additionally, we established a benchmark dataset with hundreds of customized tasks derived from public datasets with open-source services, on which we evaluate DSR compared to existing methodologies, including ImageDataset2Vec, AutoMRM, and TransferGraph. Our results demonstrate DSR's superior performance in terms of accuracy, efficiency, and generality. We also conduct a case study on an industrial surface defect detection scenario, which illustrates the feasibility of the system.

DatalogMTL is a powerful rule-based language for temporal reasoning. Due to its high expressive power and flexible modeling capabilities, it is suitable for a wide range of applications, including tasks from industrial and financial sectors. However, due its high computational complexity, practical reasoning in DatalogMTL is highly challenging. To address this difficulty, we introduce a new reasoning method for DatalogMTL which exploits the magic sets technique—a rewriting approach developed for (non-temporal) Datalog to simulate top-down evaluation with bottom-up reasoning. We have implemented this approach and evaluated it on publicly available benchmarks, showing that the proposed approach significantly and consistently outperformed state-of-the-art reasoning techniques.

Unmanned aerial vehicle (UAV) swarms are useful for mobile and collaborative applications due to their flexibility, scalability, and reliability. However, managing their communication and collaboration in complex environments is challenging. Service mesh has demonstrated excellent performance in managing communication between microservices in cloudnative environments. However, its centralized and static network structure design hinders its adaptability to dynamic topologies, increases vulnerability to single points of failure, and exacerbates resource constraints when applied to UAV swarms. To address these challenges, we propose UAV-Mesh, a graph-based decentralized service mesh framework for UAV swarms. It models the swarm as a dynamic graph for enabling the data plane to adapt to changing topologies, mitigates the risk of single points of failure through a decentralized control plane, and addresses resource constraints by optimizing consensus mechanism and algorithm. UAV Mesh offers a decentralized perspective for the application of service mesh in UAV swarms. Through experiments and analysis involving varying numbers of UAVs in a complex scenario, we demonstrate the effectiveness and efficiency of UAV Mesh in managing and controlling UAV services.

Whole Slide Images (WSIs) are high-resolution digital representations of histopathological slides that contain a wealth of information crucial for disease diagnosis. However, analyzing these images manually is a time-consuming and labor-intensive task. Therefore, the uses of AI algorithms for WSIs analysis have become a trend. Building WSIs analysis models, however, are very challenging. Under specific conditions, the scales of lesions for diseases vary greatly and the local texture features of different lesions are highly similar. In this paper, we propose a Local-Global Multi-Scale Feature Fusion Learning Network (LoGl-Net) to achieve segmentation of various types of lesions. Firstly, pre-processing algorithms are designed to segment regions of interest based on color features in the original WSIs. Next, a three-layer patch network is constructed for extracting features of each region of interest in multiple scales. Cross-layer attention is established via pooling operations, using local histological features to adjust global morphological feature classification. Experiments conducted on a dataset built by this research and a public dataset obtained promising results, demonstrating the potential of our model in improving the accuracy and efficiency of lesion segmentation.

In response to the demand for automated analysis and optimized reasoning of part model data faced in the process of ship research, design and manufacturing, this paper combines the knowledge graph with the model design rule base and manufacturing process library to construct a model process generation framework based on the knowledge graph. The paper first standardizes heterogeneous models for each type of CAD model in research design to construct a component hierarchy diagrams for multi-level assembly of models. Then the decomposed models are subjected to feature extraction and association. Then the process analysis based on model features is carried out with the help of knowledge graph. Finally, the automated generation of process plans is realized through the reasoning and consistency verification of process combinations of single components. The project explores the common components in ship research, design and manufacturing, and the case application results show that the project system has high accuracy and good adaptability, and has practical value and reference significance for the interoperability of design and manufacturing process data and the realization of automated process planning.

In the domain of ship design and manufacturing, complex plates are widely used in ship structures, and the rapid retrieval of plate classifications is crucial for effective ship design management. This paper proposes a novel method for similar ship plate retrieval, which is based on semantic fuzzy classification. Commencing with the reconstruction of plate surface structures from point clouds, the method then proceeds to extract semantic features via multi-scale geometric feature extraction, feature line detection, and high-level semantic label extraction. To address the fuzzy boundaries between plate categories, the method constructs fuzzy category vectors to characterize the features of plates. Finally, by integrating fuzzy classification results with fine-grained geometric feature differences, the method achieves the retrieval of similar plates. Experimental results demonstrate that this approach significantly improves the efficiency and accuracy of similarity retrieval among complex ship plates, thereby providing robust support for efficient ship design, and holding important application value.

Multimodal recommendation has become a key technology for social media platforms. It is widely used in content recommendation, user preference analysis, advertisement placement, etc. Existing recommendation methods mainly focus on learning multimodal embeddings from direct interactions between users and items, ignoring indirect interactions among users-to-users and items-to-items. This limits the further exploration of potential interests between users and items. Moreover, during the model training, classical recommendation methods usually randomly select uninteracted items of a user as their negative samples. This may introduce significant learning bias, as uninteracted items could be false negatives and still potentially interest the user. To this end, we propose a novel indirect interactions discovery and true negative sampling multimodal recommendation (ITMRec) method to further explore potential user interests and mitigate the issue of false negative samples during learning. Specifically, we propose an indirect interactions discovering (IID) model to explore the latent interests among users-to-users and items-to-items. Then, we propose a true negative sampling (TNS) model to refine negative sampling that can alleviate the false negative sample problem. Finally, we enhance existing collaborative filtering methods by integrating representations derived from multimodal content, indirect interactions discovery, and refined negative sampling strategies, allowing for more precise alignment with users’ latent interests. Extensive experiments on three benchmark datasets demonstrate that our ITMRec significantly outperforms state-of-the-art recommendation baselines, achieving a 3.64% improvement over peer methods. The code is available athttps://github.com/long-best/ITMRec.git.

An important task in IoT application scenarios is to perform synergy reasoning on the phenomenal data and events of strong temporal semantics and complex correlations characteristics, with the help of associated knowledge and rules. However, current reasoning methods suffer from difficult rule representation with poor readability, lack of temporal semantics, and high reasoning complexity and inefficiency. To address these problems, this paper proposes parallel collaborative reasoning approaches based on DatalogMTL. Firstly, a series of collaborative access control mechanisms are designed for the concurrent conflict problems. Then, the rule-level parallel and fact-level parallel reasoning methods are presented based on materialization algorithm respectively. In this paper, we take experiments on two relative datasets and verify that our approaches greatly improve the reasoning efficiency and have good scalability in IoT scenarios.

Heterogeneous Information Networks (HINs) are crucial in various intelligent systems. The latest advancements in HIN learning aim to combine meta-paths and hypergraphs, capitalizing on their strengths for further success. However, existing methods typically transform meta-paths into hypergraphs by simply removing the original edges from the meta-paths to integrate two semantics. This will inevitably encounter semantic ambiguity, a so-called semantic-shift problem, during the “meta-path → hyperedges” transforming, causing limited improvements. To address this, we introduce a novel fusion framework that distills knowledge from meta-paths into hypergraphs, mitigating such a problem. Specifically, we propose a unique hyperedge extraction method for constructing the hypergraph, incorporating various aspects instead of relying solely on one type of meta-path. Subsequently, we introduce a shallow student model to capture high-order information from the hypergraph, complementing a teacher model that focuses on encoding low-order information from meta-paths. Then, a distillation framework is employed to integrate explicitly multi-order information into the student. Experimental results across diverse datasets demonstrate a substantial improvement in node classification tasks, with an average accuracy increase of 2.1% over existing state-of-the-art methods.

Tracking the process of remote task execution is critical to timely process analysis by collecting the evidence of correct execution or failure, which generates a process digital twin (DT) for remote supervision. Generally, it will encounter the challenge of constrained communication, high overhead, and high traceability demand, leading to the efficient remote process tracking issue. Existing approaches can address the issue by monitoring or simulating remote task execution. Nevertheless, they do not provide a cost-effective solution, especially when unexpected situation occurs. Thus, we proposed a new cloud-edge collaboration framework for process DT generation. It addresses the efficient remote process tracking issue with a real-virtual collaborative process tracking (RVCPT) approach. The approach contains three patterns of real-virtual collaboration for tracking the entire process of task execution with a coevolution pattern, identifying unexpected situations with a discrimination pattern, and generating a process DT with a real-virtual fusion pattern. This approach can minimize tracking overhead, and meanwhile maintains high traceability, which maximizes the overall cost-effectiveness. With prototype development, case study and experimental evaluation show the applicability and performance advantage of the new cloud-edge collaboration framework in remote supervision.

Internet-of-Things (IoTs) entity discovery plays an important role in the Industrial IoTs, especially with the rapidly increasing and updating of IoT sensors in the industrial environment driven by the era of Industry 4.0 and intelligent manufacturing. However, large numbers of nonsmart sensors are required in the industrial environment, causing IoT entity discovery challenges. Unlike the smart sensor, the nonsmart sensor with limited computation and communication ability is hard to discover and recognize by traditional IoT platforms. Aiming at the challenge, this work proposes a novel IoT entity discovery middleware for nonsmart sensor discovery in the industrial environment. The proposed middleware combines both sensor knowledge graphs and sensor data values to build an IoT entity discovery and recognition model. A knowledge–data fused learning network is proposed for the model to identify the data type, function, and other information of the nonsmart sensor. At last, a prototype middleware with the discovery and recognition model is produced to implement nonsmart sensor discovery. In the experimental evaluations, the prototype middleware tests various nonsmart sensors and achieves 87.6% recognition accuracy. In real-world case studies, the prototype middleware proves the feasibility and effectiveness of nonsmart sensor discovery in the industrial environment.

Real-Time Traffic Flow Prediction (RT-TFP) is one of the critical technologies for implementing the Intelligent Transportation System (ITS), enabling rapid and accurate prediction of real-time traffic flow at intersections. RT-TFP typically needs to be deployed on-site edge devices for real-time traffic flow calculation that requires low inference latency and minimal computational resources. However, the existing Traffic Flow Prediction (TFP) models are generally based on spatiotemporal graph neural networks (STGNNs), which are complex and require high computational resources and relatively high inference times that can hardly be deployed on edge devices. To this end, this work proposes a simple RT-TFP model, SpatioTemporal-MultiLayer Perceptron (ST-MLP), which requires low computational resources and inference times. The base idea of this work is to establish a spatio-temporal MLP model to replace the STGNN model for conducting the TFP, which is much faster and simpler. Specifically, first, a TempEncoder is proposed to encode the temporal information into the MLP features. Then, a Spatiotemporal Mixer is proposed to mix spatial information into the temporal-enriched MLP features. After, MLP features are distilled from a complex STGNN model to obtain a simple MLP that inherits complete Spatial-Temporal information of the traffic graph. The experimental results on four real-world datasets show the proposed model achieves competitive prediction accuracy with STGNN models in much fewer computational resources and lower prediction time costs. It is worth noting that, the proposed method is faster than the compared STGNNs by an average of 21.62 times (~10.81s$\rightsquigarrow ~\sim 0.50$s). Interestingly, the proposed ST-MLP even has a −3.23% error rate decreasing on average compared to the corresponding STGNN model. Moreover, the error rate of the proposed ST-MLP decreases over pure MLPs by −3.92%$\sim -42.62$%. The source code is available at:https://github.com/zhangjunfeng1234/ST-MLP

The three-dimensional (3D) curved plate manufacturing is performed by constructing surfaces corresponding to the shape of the curved plate for multi-point forming. However, in the manufacturing process, the rebound restricts the forming accuracy, and the currently adopted rebound control methods cannot predict the rebound amount accurately. Meanwhile, the process involves multi-role collaboration and multiple data conversions and comparisons. These problems lead to a high degree of manual dependence, which affects manufacturing efficiency and accuracy. To address the above problems, this paper proposes a collaborative platform for the intelligent manufacturing of curved plates based on graph matching. Firstly, this paper establishes information models covering the whole process of curved plate manufacturing and forms a unified topology graph model. Then, the intelligent generation method of processing parameters based on graph matching is proposed, which realizes similar case recommendation and case-based processing parameters generation. Finally, we design and develop a collaboration platform based on micro-service architecture to support efficient collaboration among various departments and roles. In this paper, we use sail-shaped curved plates as a case of processing parameters generation and verify that this intelligent method can improve the accuracy of rebound control by comparison with related work, which shows that our method can be effectively applied to curved plate manufacturing.

Since 2020, the COVID-19 pandemic severely disrupted regular off-line business activities. This unprecedented situation inspires the valuable research on facilitating off-line business under pandemics. In this article, we conceptualized the problem as travel management in pandemic (TMiP) and analyzed it from the technological perspective. Enabling travel in a pandemic not only needs a health certificate to prove that the traveler is safe but also entry/exit permissions from both the origin and the destination regions, determined by the local situation and measures. Thus, TMiP is related to technical, social, economic, and administrative factors. By conducting a review on the literature covering the health certificate technology, its adoption in practice, and the exchange system technology published during the COVID-19 pandemic, we learned about their usefulness and limitations in TMiP. Second, we analyzed the review outcomes to infer the six distinctive technical challenges of TMiP. Third, we analyzed the feasibility of referential solutions to these challenges and showed their applicability and limitations. Finally, we offered the perspectives on new TMiP solutions and concluded that they rely on adapting existing solutions, creating new ones, and integrating all of them. We also presented future research directions in a holistic view of TMiP technical solutions. Overall, the findings of the study will stimulate more research on a more coordinated, comprehensive, and intelligent TMiP solution. We also hope this article can help practitioners to restart economies in a pandemic.

One of the main goals of the Internet of Things(IoT) systems is to achieve intelligent interaction of IoT devices. Event prediction is one of the approaches to achieve intelligent interaction. Event logic graph can effectively represent the relationship between events and be used for event prediction. However, in IoT systems, the data generated by IoT devices are usually incomplete and there are complex relationships between events, which in turn affect the accuracy of event prediction in the event logic graph. To address the above problems, this paper proposes a scenario-aware event prediction approach based on event logic graph in IoT systems. First, a flexible paradigm is designed for recognizing events and scenarios in IoT devices. Then, a scenario collaboration-based event context extraction method is proposed for extracting event contexts with similar scenario attributes in the event logic graph. Finally, a scenario-based event prediction method is designed to predict the events that will occur subsequently. In this paper, we verify that our approach can improve the accuracy of event prediction through the case of driving, which shows that our approach in this paper can be effectively applied in IoT systems.

Recent trends in the Web of Things (WoT) have led to data explosion. Data lake (DL), as a flexible on-demand heterogeneous data management architecture, has become a feasible solution in data management. Metadata modeling for DLs is the key basis for smart analysis and processing. However, the varieties in structures and semantics of industrial WoT data hinder metadata modeling and maintenance. Moreover, the lack of textual descriptions and the semantics hidden in value streams make it hard to automatically construct semantic metadata. The dynamic nature of WoT requires on-time evolution on metadata. To overcome these challenges, we propose an automated bottom-up metadata generation approach for DL of WoT applications. Applying a data-driven framework, raw data are notated as linked data and self-organizing map-based online clustering is applied to real timely extract data characteristics. To recognize entities, concepts and relations, semantics-based entity discovery approach from short texts is proposed according to the feature of WoT data. The numerical analysis is performed to find the hidden relations from raw values. Full-dimensional metadata with rich semantic knowledge are finally built. Experiments on a real-world dataset are conducted to verify the effectiveness of methods and a case study on an energy WoT system is provided to demonstrate the feasibility of the approach.

Currently user authentication and identity monitoring are required in various collaborative computer supported systems, such as online assessments and examinations. However, existing authentication methods such as passwords checking are less reliable. In addition, identity monitoring is hard to be realized effectively and efficiently for applications based on collaborative architecture. In this paper, we leverage keystroke dynamics to explore biometrics security and propose a user authentication framework based on edge computing architecture to address these issues. To support both static and continuous authentications with high accuracy and efficiency, dynamically improved keystroke profiles, Gaussian model based anomaly detector and keystroke stream processing are designed in the framework. The feasibility and effectiveness of the framework are verified by three representative public data sets and a real-world case study. The results show that the authentication can proceeds efficiently to enable an undisturbed and secure environment for online examinations.

One of the primary goals of cyber-physical system is to deeply integrate cyberspace and the physical world to realize intelligent interaction of the system. Event prediction technique is a powerful means to fulfill this goal. Recently, a novel knowledge graph, the event graph, is widely studied in the field of event analysis due to its excellent ability in event relationship modeling. Therefore, this paper proposes constructing the event graph to model the sequential event evolution in the physical world for event prediction. To this end, the sequential event graph construction method and related event prediction mechanism for CPSs are proposed. First, a flexible and universal paradigm is designed to assist in extracting event instances from the data generated by physical devices. Then, an automatic event graph construction method based on frequent episode mining is proposed. Finally, the related prediction mechanism is designed, including the identification of contextual information a nd the prediction of subsequent events. A case study on car usage illustrates the feasibility of our approach. The flexibility and support for complexity are demonstrated by a comparative discussion.

While recent research in music generation has mostly focused on encoder decoder architectures and self-attention mechanisms, prominent advancements regarding GANs have not yet been incorporated for the creation of music. These include solutions for major challenges when training GANs, most importantly training instability. In this work, we aim to apply this new knowledge to music generation, in order to make it more efficient and enable the automatic creation of music of higher quality. We utilize the progressive approach towards GANs, and implement it to train on symbolic music data. For best results, we process this data to obtain a new dataset, which matches the progressive approach. To achieve this, we propose a new way of downsampling fit for musical data. We furthermore conduct a user study to evaluate our results, and compute an FID score of 12.30 as objective metric.

Coordinating of energy networks to form a city-level multidimensional integrated energy system becomes a new trend in Energy Internet (EI). The collaborating in the information layer is a core issue to achieve smart integration. However, the heterogeneity of multiagent data, the volatility of components, and the real-time analysis requirement in EI bring significant challenges. To solve these problems, in this article we propose a stream processing framework based on linked data for information collaboration among multiple energy networks. The framework provides a universal data representation based on linked data and semantic relation discovery approach to model and semantically fuse heterogeneous data. Semantics-based information transmission contracts and channels are automatically generated to adapt to structural changes in EI. A multimodel-based dynamic adjusting stream processing is implemented using data semantics. A real-world case study is implemented to demonstrate the adaptability, feasibility, and flexibility of the proposed framework.

With the transformation of Internet of Things to Web of Things (WoT), a variety of applications are required to deal with huge volumes of real-time and heterogeneous data. However, in most applications, due to the weak semantics of data itself and the loose combination with specific scenarios, it is sometimes difficult to depict the spatiotemporal feature of the scenario in an application only through the data. In this paper, a spatiotemporal semantic scenario meta-model-based configurable platform is proposed for the development of WoT applications to address this issue, based on the data configuration, event stream configuration, and service encapsulation, the entire WoT scenario can be depicted with an abstract data model and related rules, and the business process can be changed by redefining corresponding rules when requirements change. A case study is given to verify the feasibility of our platform. The result shows that the platform can provide background support for WoT applications in a promising way.

Instance matching frameworks that identify links between instances, expressed as owl: sameAs assertions, have achieved a high performance while the performance of property matching lags behind. In this paper we leverage owl: sameAs links and show how these links can help for property matching. First, we extract all owl: sameAs instance pairs together with their properties and transform them into tables. Then, we apply table matching techniques and propose matching criteria to find relationships between properties. The experiments with real world LOD datasets show the efficiency and effectiveness of the proposed approach to deal with property matching.

Cloud of Things (CoT) is an integration of Internet of Things (IoT) and cloud computing for intelligent and smart application especially in mobile environment. Model Driven Architecture (MDA) is used to develop Software as a Service (SaaS) so as to facilitate mobile application development by relieving developers from technical details. However, traditional service composition or mashup are somewhat unavailable due to complex relations and heterogeneous deployed environments. For the purpose of building cloud-enabled mobile applications in a configurable and adaptive way, Model-Driven Development Patterns based on semantic reasoning mechanism are provided towards CoT application development. Firstly, a meta-model covering both multi-view business elements and service components are provided for model transformation. Then, based on formal representation of models, three patterns from different tiers of Model-View-Controller (MVC) framework are used to transform business models into service component system so as to configure cloud services rapidly. Lastly, a related software platform is also provided for verification. The result shows that the platform is applicable for rapid system development by means of various service integration patterns.

With the wide spreading of mobile and Internet of Things (IoT) devices, music cognition as a meaningful task for music promotion has attracted a lot of attention around the world. How to automatically generate music score is an important part in music cognition, which acts as an important carrier so as to disposing huge quantity of music data in IoT networks or Internet. For the reason that the computers lack of the domain knowledge and cognitive ability, it is hard for computers to recognize the melody of music or write score while listening to the music. Therefore, a music cognition system is introduced to cognate music and automatically write score based on machine learning methods. First, considering large-scale data processing is needed by machine learning algorithms and a number of music devices are involved in the cognition system through Internet, fog computing is adopted in the proposed architecture to efficiently allocate computing resources. Then, the system can collect, preprocess, and store raw music data on the fringe nodes. Meanwhile, these data will be transmitted from fog nodes to cloud servers to form music databases. Then, machine learning algorithms, such as hidden Markov model and Gaussian mixture model, are performed in cloud servers to recognize music melody. Finally, a case study of music score generation demonstrates the proposed system. It is shown that the method provides an effective support to generate music score, and also proposed a promising way for the research and application of music cognition.

Driven by the automation technologies and health informatics of Industry 4.0, hospitals in China have deployed a complete automation system/platform for healthcare services accessing. Without much more Internet knowledge, elderlies usually seek the third-party to assist them to get healthcare services from Web or APPs, it consequently results in an unexpected situation that scalpers could grab all healthcare services booking by unrighteous means in order to resell to elderlies for a much higher price. Moreover, it is hard for physicians to identify the scalpers due to the complexity, ad-hoc, and multiscenario nature of healthcare processes. In this paper, a novel method is proposed for the identification and creation of user groups of scalpers in mobile healthcare services. The approach utilizes and extends state of the art data analysis approaches in the event-logs of the mobile system to identify user groups. Based on the user groups, user profiles are extracted by identifying representative eventcases from hierarchical user-event clusters. A comprehensive evaluation is conducted in a selected test-set from the event-logs of a mobile healthcare APP. The result shows its accuracy and effectiveness in scalper detection in mobile healthcare APP. Further, a complete case study is deployed in a real word hospital to ensure its utility, efficacy, and reliability.

Distributed computing technologies, as popularized by Hadoop, have been proliferating in Cloud and enterprise computing over ten years, with the capability of processing data across thousands of machines. There is a wide diversity of workloads in such large scale clusters shared by multi-tenant. Hence, resource utilization and task scheduling become vital to performance and bring challenges to architecture designers. We present Daphne, a hybrid scheduling framework that strikes a tradeoff among three universal scheduling frameworks: 1) centralized scheduling; 2) loose coordination scheduling; and 3) fully distributed scheduling. Daphne defines a matching tree that forwards the application to the fittest scheduler according to the characteristic and priority of the application. Besides, Daphne utilizes resource prediction to increase task throughput significantly while it does not interfere with any running workload. We implement Daphne based on YARN and demonstrate that Daphne improves task throughput by nearly 17%.

Internet of Things (IoT) related applications have emerged as an important field for both engineers and researchers, reflecting the magnitude and impact of data-related problems to be solved in contemporary business organizations especially in cloud computing. This paper first provides a functional framework that identifies the acquisition, management, processing and mining areas of IoT big data, and several associated technical modules are defined and described in terms of their key characteristics and capabilities. Then current research in IoT application is analyzed, moreover, the challenges and opportunities associated with IoT big data research are identified. We also report a study of critical IoT application publications and research topics based on related academic and industry publications. Finally, some open issues and some typical examples are given under the proposed IoT-related research framework.

Information resource services are the key element for resource sharing in cloud manufacturing. Traditional resource service models focus on modeling the attributes, interfaces, and descriptions of the resources into resource information services. Such resource services are suitable for local environment but suffer semantic heterogeneities in open cloud environment. Recently, well-designed ontologies are applied in resource service models to unify the schema and eliminate the semantic heterogeneities among the services. However, the effectiveness of ontology-based models mainly depends on the expertise of the ontology experts in ontology designing. Moreover, it is difficult to catch the dynamic changes in the cloud once the ontology has been embedded. In this paper, a semantic model is presented for information resource service modeling that uses semantic links instead of ontologies. The model takes advantage of semantic links to enable automated integrating and distributed updating in resource service cloud. In the experiment, the model is applied on practical manufacturing resources from a wheel manufacturing company. The case study and experimental results show that the proposed model is suitable for modeling manufacturing resources into cloud services and enables the flexible and distributed manipulation on resource services in the cloud environment.

Ontology matching is an important part of enabling the semantic web to reach its full potential. Most existing ontology matching methods are mainly based on linguistic information (label, name, title and comment) but from the results achieved it is realized that this information is not sufficient. The latest ontology matching research works are trying to deeply dig into the structural information of ontologies by using "similarityflooding" method. However, there are several innate issues in similarity-flooding methods that lead to wrong matching results. In this paper, we report the problems of similarity-flooding in ontology matching and propose a novel method to effectively leverage the structural information of the ontology. The evaluation is conducted on OAEI ontology matching benchmarks from 2011 to 2015. The result shows that the proposed approach performs comparatively well with other state of the art matching systems.

For bridging the semantic gap between the low-level features of videos and high-level semantic concepts in videos, we propose a multi-semantic video annotation method with semantic network. First, we use the semantic network to represent the highlevel semantic knowledge and model the relationships between the concepts. Then we divide the videos to key frames and use Convolutional Neural Networks (CNNs) to extract low-level visual features and detect the concepts in the videos. Finally, we combine the low-level features with the high-level knowledge to perform a two-level reasoning to optimize the result. Experiment results show that the proposed method significantly outperforms existing video annotation techniques in terms of precision value.

Emergency clinical decision making is a challenging issue in healthcare services, notably in the environment of complicated data processing. Effective and efficient clinical decision making highly depends on the sufficient information sharing of the involved working teams. However, emergency decision support systems are usually hard to be developed because that the problems of emergency decision are always unexpected and unstructured. This paper focuses on the developing of decision support system to coordinate actions carried out in emergency situations. A framework is proposed based on MDA (Model-Driven Architecture) approach and resource model to dynamically build decision support system when emergency events occur. The effectiveness of our method is discussed and verified in a case study of collaborative clinical decision making on traffic accident emergency rescuing. The result shows that the MDA approach combined with resource model has the potential to support information system evolution along with the emergency events.

Management of ship designing is difficult because of the complexity in modeling of the process and large amounts of data throughout the process. Though existing technologies of BPMN can solve the first problem, BPMN does not provide sufficient supports on dealing with complex dependencies in the process, e.g., storage and search problems of correlated data with “many to many” relations. In this paper, we introduce a series of annotations of data objects to solve this problem. First, we extend the annotations of data objects based on BPMN and utilize foreign key in relational database to manage the relations between data objects. Second, we use SQL queries to execute the common operations to data objects in the process. Our approach is the extension of mature BPMN and database technologies, so it is standard and reusable. We implemented our approach on the Produce Management System to verify the availability and efficiency.

Internet of Things (IoT) software is required not only to dispose of huge volumes of real-time and heterogeneous data, but also to support different complex applications for business purposes. Using an ontology approach, a Configurable Information Service Platform is proposed for the development of IoT-based application. Based on an abstract information model, information encapsulating, composing, discomposing, transferring, tracing, and interacting in Product Lifecycle Management could be carried out. Combining ontology and representational state transfer (REST)-ful service, the platform provides an information support base both for data integration and intelligent interaction. A case study is given to verify the platform. It is shown that the platform provides a promising way to realize IoT application in semantic level.

The Internet of Things (IoT) has provided a promising opportunity to build powerful industrial systems and applications by leveraging the growing ubiquity of Radio Frequency IDentification (RFID) and wireless sensors devices. Benefiting from RFID and sensor network technology, common physical objects can be connected, and are able to be monitored and managed by a single system. Such a network brings a series of challenges for data storage and processing in a cloud platform. IoT data can be generated quite rapidly, the volume of data can be huge and the types of data can be various. In order to address these potential problems, this paper proposes a data storage framework not only enabling efficient storing of massive IoT data, but also integrating both structured and unstructured data. This data storage framework is able to combine and extend multiple databases and Hadoop to store and manage diverse types of data collected by sensors and RFID readers. In addition, some components are developed to extend the Hadoop to realize a distributed file repository, which is able to process massive unstructured files efficiently. A prototype system based on the proposed framework is also developed to illustrate the framework's effectiveness.

The rapid development of Internet of things (IoT) technology makes it possible for connecting various smart objects together through the Internet and providing more data interoperability methods for application purpose. Recent research shows more potential applications of IoT in information intensive industrial sectors such as healthcare services. However, the diversity of the objects in IoT causes the heterogeneity problem of the data format in IoT platform. Meanwhile, the use of IoT technology in applications has spurred the increase of real-time data, which makes the information storage and accessing more difficult and challenging. In this research, first a semantic data model is proposed to store and interpret IoT data. Then a resource-based data accessing method (UDA-IoT) is designed to acquire and process IoT data ubiquitously to improve the accessibility to IoT data resources. Finally, we present an IoT-based system for emergency medical services to demonstrate how to collect, integrate, and interoperate IoT data flexibly in order to provide support to emergency medical services. The result shows that the resource-based IoT data accessing method is effective in a distributed heterogeneous data environment for supporting data accessing timely and ubiquitously in a cloud and mobile computing platform.

Online visual steering technology provides users a 3D cooperative working environment in the complex or multi-steps task. Due to the difference of role or authority, new demands have risen such as multi-view presentations and conflict solving mechanism for users in the process of cooperative work. Thus, a configurable architecture based on scene composition is proposed for visual steering. It mainly contains several points: (1) Configurable scene structure including geometry data, concept model, user profile, enabling scene composition to build multi-view for different uses in working environment. (2) Browser-based 3D scene rendering page using X3Dom. (3) Loose coupling model sources on web services. A demo is also provided to test our architecture and verified its usability. By comparing with other implement methods, this architecture provides more flexible customizability and expandability, together with light-weight web-based clients.

On the cloud computing platform, there are more and more composition of different web services to get an added value. However, because of the ambiguity between different forms of Service Level Agreement (SLA), to get an integrated service quality of several services is a complex task. Based on this fact, a SLA-based Service Composition Model with Semantic Support (SSCMSS) is proposed in this paper. First, a SLA management ontology model is built to provide semantic support. And then the framework composes the SLAs of different services to get an integrated service quality level based on the composition pattern. At last, it automatically searches the integrated SLAs to get the compositions that can match the customer's request. Moreover, a case study is implemented and the results show that this model is effective and useful in searching the SLA satisfied services.

Integration and unified management of data scattering along the lifecycle chain is the primary problem to be solved in PLM field. In this paper, a framework based on resource meta-model is proposed to integrate and manage product lifecycle data. Firstly, the structure of this framework is presented. Then, the resource meta-model is defined and the functions, structure and generation method of the model are given in detail. To better control resource accessing operations, a resource accessing control mechanism is proposed to make sure that product lifecycle information in different stages is accessed by authorized roles in valid way. Finally, a case study is presented to demonstrate the application of our framework for PLM. The result shows that our proposed method could integrate and manage heterogeneous data during product lifecycle more flexibly.

Due to the complexity of describing and execute business requirements, it is difficult for enterprises to adapt to rapidly changing environment. Thus a Configurable Resource-Oriented Service Architecture (CROSA) is proposed to bridge business modeling in build time and services execution in run time seamlessly. Firstly, based on business model analysis, a business meta-model is built to act as a referred model to encapsulate enterprise information resources. Then these resources act as the basic description to generate services by means of service transformation. And WADL and BPEL are involved in this step. Next, referred to Model-View-Controller pattern of software development, IT elements including service, web pages and data sources are mapping to resources oriented service architecture. Lastly, a state space defined by a resource array is built as the control mechanism for services integration so as to build a completed IT system. And a prototype system is implemented to develop data-centre information system for verification. The approach provides a way to realize service cooperation in a more flexibly pattern.

Product lifecycle data is considered valuable resource to improve quality of product designing. However, at present, under the environment of E-manufacturing, product data is scattered across different companies during product lifecycle. Because enterprise information systems are heterogeneous, it becomes a challenging task managing product lifecycle data through the complete supply chain. In this paper, firstly, the product lifecycle management process is analyzed from the viewpoint of supply chain. Then, a data model is proposed to represent business activities related to product data generating. In the data model, all business activities are discomposed to three elements which are target, work flow and resource. All product data is composited by these three kinds of elements. Furthermore, in order to integrate heterogeneous data across supply chain, ontology is used to represent the proposed data model. Because that the ontology is abstracted from different companies in supply chain, multi-views modeling method is adopted to construct ontology collaboratively. Finally, based on the data model of ontology, a prototype of product lifecycle management is designed and implemented in the application of a manufacturing company. The case study shows that the ontology-based data model could support the integration of heterogeneous data during product lifecycle flexibly.

The recent years have witnessed a rapid development in virtual reality technology and computer aided design, which accompanied by a vast amount of 3D content springing up from various domains, has given rise to an emerging need for methods by which they can be efficiently annotated before being searched and retrieved. In this paper a new approach to semantic annotation on 3D content is proposed. In this approach, much of the manual work on annotation is replaced with a semi-automatic process based on prior gathering of domain expertise. The geometric properties of objects and their spatial relationships are first automatically extracted and associated with a general ontology, the result of which is then matched against a set of user-defined rules to create annotations represented in an application ontology. The approach is Web3D oriented so that it can take advantage of features of X3D format. It will be shown using a prototype system how this approach can succeed in accelerating the annotation process.

With the arrival of cloud computing and the internet of things, the difficulty of extracting data from the cloud services is preventing some organizations from adopting cloud computing. At the same time, the number of web services increases dramatically, and the granularity of service is different, so how to choose a suitable granularity service also becomes a problem that needs to be resolved. So a storage strategy for cloud computing services is given and the method based on event stream is used to construct the hierarchal model of services which can provide different functional services or different granularity services to meet the business requirements. Firstly, the storage strategy for cloud computing services is that, when registering the services, the meta-data of services will be extracted and stored into database according to certain structure. Secondly, the Event Cloud can be constructed as the one-way event processing, which provide the raw data for more-way event processing. Then the more-way event processing can construct the hierarchical model of services with certain composition goal. The prototype system has been developed in order to fulfill the application, both human events and RFID events can trigger other events in the cloud events, which can create new business models, improve business processes, and reduce costs and risks.

RBAC as a kind of permission access control technologies supports enterprise information security effectively. However, in many cases, traditional RBAC can only establish a permission access control mechanism based on discrete group-role or user-role management inside an organization. And the user group whose organization structure is more complicated is not supported by RBAC. It is also lack of the adaptability of dynamic changes to the complex organization structure. To solve these problems, a permission model called Flexible Organization Structure-Based Access Control (FOSBAC) is proposed, which combines the flexible organization structure with the access control to achieve the dynamic management of permissions. First, the general framework and the formal description of FOSBAC are given. Then, the application template using the XACML specification is constructed and an analysis on a case of accessing financial statements is used to demonstrate the feasibility of the application. Finally, it is shown that FOSBAC possesses better adaptability to complex organization structure and higher management efficiency in comparison with RBAC and ROBAC.

Nowadays many companies built enterprise level data warehouses (DW) for decision making support. However explosive data accumulated in distributed databases in company or across companies with the widely use of Computer Supported Cooperative Work in Design (CSCWD) technologies. Therefore it becomes a time costing task for engineers to construct the multidimensional model of data warehouse. This research presents an ontology approach to eliminate data source heterogeneity aiming to design the conceptual structure of data warehouse automatically. The proposed approach includes a domain ontology mete-data model, which consists of data, concept, ontology and resource repositories, to describe the semantic meaning of the data sources. The supply-driven and demand-driven methodologies are combined together to construct the concept model of DW. By supply-driven method, domain ontology system is extracted bottom-up from data sources; on the other hand, by demand-driven method, relationships of the concepts in the ontology system are extended top-down according to the business process. Furthermore, candidate of the data warehouse concept model is derived according to the relationships of the concepts in the ontology system. After discussing the process of data warehouse designing, a case study is given to show how our method is used in clinic domain. The result shows that ontology method could help users designing data warehouse more easily.

In order to gain the competitive advantage, more and more hospitals put their attention on determining and optimizing the standard clinical pathway. However, there are many abnormal instances in the event logs which disturb the effect of the process mining and the process analysis. Also, the prescription and the medical test items may have a large difference in the specific disease, where has the similar clinical pathways due to the multi-factor in the clinical pathways. In this paper, an abnormal process instances identification method (APIIM) is proposed. Given the event logs and the standard clinical pathway, the method classifies the instances based on the classification attributes and identifies the abnormal instances by the outlier detection technology. Moreover, a case study using the real data in one hospital is implemented and the result shows that the method is effective and efficient in discovering the abnormal process instances in the healthcare environment.

Service matching is a key research area in Web Service Application. Most existing matching methods gain low precision so that they are not satisfied to use for automatic service discovery in various applications, such as dynamic and automatic service substitution. To solve the problem, this paper proposes a high-precision matching method based on the formal semantic description of services. In our method, we use input, output and the internal logic process to clarify the function of the service. Domain Conception Ontology is used to describe the input and output. SOFL and Domain Process Ontology are introduced to describe the internal logic process. Based on these descriptions, Domain Ontology matching, IO matching and Process matching are conducted successively to give out the high-precision matching result. We define the matching rules and give a case study to illustrate the matching method.

Business Process Modeling (BPM) is widely acknowledged as the specific technology to analyze and control workflow with flexibility and convenience. With the development of SOA theory and practical exploration, it is challenging for BPM to involve web services and to maintain complex interaction information. Since the lack of method to involve web service semantic in business process, we present the Semantic Event-driven Process Chains (SEPCs) that describe the business process of semantic web services composition. A function metamodel is defined to solve semantic inadequacy in a modeling phase and transforming phase. A mapping rule is defined to identify structure patterns and transform the SEPCs model into WS-BPEL business process model. The experiment takes purchase process of one shipbuilding company as an example to show the practical usability of the integrated modeling approach.

With rising accumulation of data and applications in enterprise IT environment, the business data management becomes increasingly important. But the data in the IT environment is usually organized by static modeling, which could not meet the business changes need. In our study the enterprise ontology is adopted as conception backbone to model for the business scenario, and the use of ontology evolution can reflect the business process changes. So we explore the data mode mapping and the modeling of enterprise ontology. Then the analysis of changes derived from application requirement is forced specially. Finally, the changes realization enables the whole evolution process to be implemented. The prototype system has been developed now in order to fulfill the application. It can help connect the data and business, and obtain business changes to direct the ontology evolution. As there are an increasing research and application tendency in the field of business data modeling and ontology evolution, our study pay more attention to the modeling, mapping and analysis between dynamic business changes and ontology.

An open resource-based integrated platform system is built for the execution of business processes, assuming that the business process model is the fundamental factor for the next generation of integrated business information systems. First of all a resource meta-model is constructed, which represents the basic information unit. Then the concept of domain ontology is used to organize and manage resources obtained from distributed and heterogeneous sources. Based on the resource, Bite is used to compose RESTful services in order to control business processes in an orderly way. Therefore, the way of services to call resources within the limits of traditional web service architectures is being transferred into a resource flow. A software architecture is proposed and tested, based on an exemplary business process scenario. The result indicates a new research direction for the application of business process models based on the platform ORIPS.

Considering the low efficient of resource re-use for lack of consideration of user's role, team, design stage, and individuation in design activity, an agent-based interactive mechanism is proposed to build CSCW supported environment. Firstly, an Interactive Agent model is used to describe user feature in CSCW design activity. The definition of role, people preference, ability description as well as the definition of action, mental state, and other descriptions is formally represented in ABI agent structure. Then, based on analysis of design activity, a general design meta-resource (DMR) structure is constructed as the basic meta-unit to extract, organize, manage, and operate resources. Therefore, by combining user feature with design resource feature into resource matching calculation, design resources are accessed and pushed to the people with more pertinence. A practical resource-based system has been implemented for a jewel design enterprise. The result shows that the system makes design activity more effectively by pushing design resources with a high precision.

On the purpose, of sharing the distributing, various format, and low relevance design resources among people of different design teams and in different stags, a CSCW design supporting environment is proposed based on design resource modeling, managing, searching, and using so as to shorten product development period. First, a general design meta-resource (DMR) structure was built to act as the united basic unit to extract resource information and build index. Based on binary large object (BLOB), some common-used resource cases are also stored in platform database to decrease the access numbers of distributing database. Then three kinds of semantics relations were given to build hierarchy semantics networks so as to organize, manage the design resource. At last, in order to carry out association-based searching, resource evaluating and filtering, a similar-degree algorithm between two DMR cases was presented. In practical, a resource-based integrated system is implemented for a jewel design enterprise. The application shows that the system enhances the relevancy and sharing among resources of different design team, and makes design activity rapidly.

This paper proposes a new approach to converting irregular genus-0 meshes into those with subdivision connectivity. The original mesh is parameterized onto the unit sphere firstly. Then a spherical base mesh is constructed with only four vertices. After that, a 1-to-4 subdivision operation and an iterative vertex relocation operation are applied alternately over the base mesh to produce a spherical subdivision mesh, which has similar vertex distribution as the spherical parameterized one. Finally, the remesh with subdivision connectivity is obtained by sampling the original surface. The experimental results show that our method can not only make the number of irregular vertices in the remesh as small as possible (only four), but also preserve the details of original model well.