Edge service monitoring is essential for ensuring the robustness and efficiency of service executions, where predictive monitoring enables proactive detection of potential service violations. Current approaches for predictive monitoring, which mostly adoptSignalTemporalLogic (STL) specifications for requirements representation and evaluation, primarily focus on deterministic signals, and thus, may lack probabilistic guarantees for uncertainty interpretation. To address these challenges, this paper proposesBayesianSTL(BSTL), an extension ofSTLthat enables probabilistic reasoning over stochastic signals. Specifically,BayesianNeuralNetworks (BNNs) are employed to generate sequences of posterior probability distributions, offering more comprehensive predictive insights compared to traditional point- or interval-based methods with deterministic sequential predictions. Uncertainty interpretation over these distribution predictions is achieved by a novel expected robustness metric that jointly quantifies both the degree and probability of service satisfaction. Thereafter, aBSTL-based predictive monitoring framework is developed, where a service constraint is formally specified by aBSTLformula and interpreted with both qualitative and quantitative semantics. Besides, confidence levels and constraint thresholds ensuring robust satisfaction of aBSTLformula are rigorously estimated. Extensive experiments on publicly available datasets demonstrate thatBSTLoutperforms baseline techniques in terms of expressiveness, robustness, and applicability.

Edge service monitoring is crucial for ensuring the robustness and reliability of service executions. Predictive monitoring, in particular, enables proactive detection of potential service violations. Existing predictive monitoring approaches, often leveraging Signal Temporal Logic (STL) for requirement specification, primarily focus on deterministic signals, and thus, lack probabilistic guarantees for uncertainty interpretation. To address these challenges, this paper introduces Bayesian STL (BSTL), an extension of STL that enables probabilistic reasoning over stochastic signals. Specifically, Bayesian Neural Networks (BNNs) are utilized to transform deterministic sequential predictions into sequences of posterior probability distributions. Uncertainty interpretation over these distribution predictions is achieved by a novel expected robustness metric that jointly quantifies both the degree and probability of service satisfaction. Thereafter, a BSTL-based predictive monitoring framework is developed, wherein service constraints are formally specified by BSTL formulae and interpreted with both qualitative and quantitative semantics. Besides, confidence levels and constraint thresholds ensuring robust satisfaction of BSTL formulae are rigorously estimated. Extensive experiments on publicly available datasets demonstrate that BSTL outperforms baseline techniques in expressiveness, robustness, and applicability.

Parameter-efficient fine-tuning (PEFT) methods, such as LoRA, have enabled the efficient adaptation of large language models (LLMs) by updating only a small subset of parameters. However, their robustness under out-of-distribution (OOD) conditions remains insufficiently studied. In this paper, we identify the limitations of conventional LoRA in handling distributional shifts and propose $\textbf{HiMoLE}$($\textbf{Hi}$erarchical $\textbf{M}$ixture of $\textbf{L}$oRA $\textbf{E}$xperts), a new framework designed to improve OOD generalization. HiMoLE integrates hierarchical expert modules and hierarchical routing strategies into the LoRA architecture and introduces a two-phase training procedure enhanced by a diversity-driven loss. This design mitigates negative transfer and promotes effective knowledge adaptation across diverse data distributions. We evaluate HiMoLE on three representative tasks in natural language processing. Experimental results evidence that HiMoLE consistently outperforms existing LoRA-based approaches, significantly reducing performance degradation on OOD data while improving in-distribution performance. Our work bridges the gap between parameter efficiency and distributional robustness, advancing the practical deployment of LLMs in real-world applications.

Link prediction is a fundamental task of graph machine learning, and Graph Neural Network (GNN) based methods have become the mainstream approach due to their good performance. However, the typical practice learns node representations through neighborhood aggregation, lacking awareness of the structural relationships between target nodes. Recently, some methods have attempted to address this issue by node labeling tricks. However, they still rely on the node-centric neighborhood message passing of GNNs, which we believe involves two limitations in terms of information perception and transmission for link prediction. First, it cannot perceive long-range structural information due to the restricted receptive fields. Second, there may be information loss of node-centric model on link-centric task. In addition, we empirically find that the neighbor node features could introduce noise for link prediction. To address these issues, we propose a structural information enhanced link prediction framework, which involves removing the neighbor node features while fitting neighborhood graph structures more focused through GNN. Furthermore, we introduce Binary Structural Transformer (BST) to encode the structural relationships between target nodes, complementing the deficiency of GNN. Our approach achieves remarkable results on multiple popular benchmarks, including ranking first on ogbl-ppa, ogbl-citation2 and Pubmed.

Empowered by edge computing, resources and computation capabilities provided by edge devices can be encapsulated as containerized services. When burst requests are coming, there may have edge devices which are overloaded, since most requests are spatially and temporally constrained, and edge devices are resource-scarceness and capacity-limited. Overloaded devices should be relieved through optimally migrating one or more activated services to contiguous edge devices. Besides, sensory data gathered by original edge devices should be periodically transmitted to migrated devices for data analysis purpose. To mitigate this issue, this paper proposes an Energy-efficient Online Service Migration (EOSM) mechanism to conduct the migration of multiple services simultaneously. Extensive experimental results show that our EOSM mechanism outperforms the state of arts techniques in mitigating overloaded services in terms of access latency, energy consumption, and request success rate.

Empowered by edge computing, resources and computation capabilities provided by edge devices can be encapsulated as containerized services, and domain applications can be achieved through service compositions. When burst requests are coming to be satisfied, there may exist edge devices which are overloaded, since requests are mostly spatially and temporally constrained, and edge devices are resource-scarceness and capacity-limited. In this setting, overloaded devices should be relieved through optimally migrating one or more activated services to contiguous edge devices. Besides, sensory data gathered by original edge devices should be periodically transmitted to migrated devices for data analysis purpose. To mitigate this issue, this paper proposes an Energy-efficient Online Service Migration (EOSM) mechanism to conduct the migration of multiple services simultaneously. Specifically, a light service sharing strategy is developed to only transmit the top container layer, and a modified NSGA-II algorithm is adopted to generate one or multiple paths for the container layer and time-series sensory data migration of each migrated service. Extensive experimental results show that our EOSM strategy outperforms the state of arts techniques in mitigating overloading devices in terms of access latency, energy consumption, and request success rate.

Edge service monitoring is essential to guarantee the healthy of service compositions at runtime. Current techniques focus mostly on the monitoring of atomic edge services, but they are inadequate for that of inter- and composite services. Besides, constraints to be monitored are usually pre-specified, although certain parameters may have to be adapted online according to the execution context. To address these challenges, this paper formulates the problem of edge service monitoring as the interpretation of temporal constraints and time-dependent QoS constraints upon intra-, inter-, and composite services. Leveraging our proposed Compositional Signal Temporal Logic (CSTL) with extended compositional modalities and online parameter settings, an adaptive monitoring mechanism is developed, where constraints are converted to CSTL formulae, and QoS variations and temporal violations are interpreted qualitatively and quantitatively at runtime. Extensive experiments are conducted upon publicly-available datasets, and evaluation results show that CSTL performs better than baseline techniques in terms of expressiveness, applicability, and robustness.

Anomaly detection is an important topic in the Internet of Things (IoT). Recently, some anomaly detection methods based on graph neural networks (GNNs) have gained much attention. However, such methods require a large amount of sensory data for inference, which leads to high energy consumption for data transmission and can hardly be applied in IoT scenarios. This paper proposes a subgraph-based anomaly detection strategy as an energy-efficient anomaly detection method. To accomplish this task, we use graph structure generation to divide subgraphs by feature similarity and reduce energy consumption for data transmission. To validate the effectiveness of our mechanism, we use real-world IoT multivariate time-series data for modelling. The results show that our scheme is more energy efficient and has higher precision compared to other methods in anomaly detection.

With the wide-adoption of edge computing, the functionalities of Internet of Things (IoT) devices can be encapsulated as edge services, to facilitate domain applications through edge service compositions. Considering the capacity-fluctuating and resource-varying of IoT devices, edge service monitoring is essential to guarantee the healthy of their compositions at runtime. Current techniques focus mostly on the monitoring of atomic edge services, which, however, are inadequate for that of inter-and composite services. Besides, constraints to be monitored are usually pre-specified, although certain parameters may have to be adapted online according to execution context. To address these challenges, this paper proposes a novel logic-based adaptive monitoring mechanism, to achieve the interpretation of temporal constraints and time-dependent QoS constraints upon intra-, inter-, and composite services. Leveraging our proposed Compositional Signal Temporal Logic (CSTL) with extended compositional modalities and online parameter settings, constraints can be converted to CSTL formulae, and QoS variations and temporal violations are interpreted qualitatively and quantitatively at runtime. Extensive experiments are conducted upon publicly-available datasets, and evaluation results demonstrate that our CSTL performs better than baseline techniques in terms of expressiveness, applicability, and robustness.

With the applicability of edge intelligence in various domains, anomaly detection, which aims to identify unusual and infrequent circumstances, is regarded as a regularly performed task to guarantee the health of the Internet of Things (IoT) applications. Generally, sensory data are gathered at the network edge and completely transmitted to the cloud, where computational-heavy algorithms are mostly adopted to determine the locations of anomaly. Considering the occurrence infrequency of anomalies, this strategy may transmit relatively huge volume of sensory data, which may reflect a healthy situation indeed, to the cloud. To mitigate this problem, this article proposes an accurate anomaly detection mechanism with energy efficiency in three-tier IoT–edge–cloud collaborative networks. Specifically, after gathering sensory data provided by IoT nodes in certain edge networks, the edge node applies the marching squares algorithm to generate isopleths, where an isopleth may capture the boundary of anomaly. A sensory data filtering mechanism is conducted at the edge tier, such that anomaly-relevant sensory data are transmitted to the cloud and, thus, the network traffic is decreased significantly. Thereafter, the boundary of anomaly is obtained, and the locations of candidate boundary nodes are determined by adopting the Kriging spatial interpolation algorithm at the cloud tier. These locations are traversed by mobile sensing nodes at edge networks, and their sensory data are gathered for boundary refinement. Extensive experiments are conducted on an air quality hazardous gas data set from Toward Data Science, and evaluation results show that our technique outperforms the state-of-the-art counterparts in boundary accuracy and energy consumption.

Financing needs exploration (FNE), which explores financially constrained small- and medium-sized enterprises (SMEs), has become increasingly important in industry for financial institutions to facilitate SMEs’ development. In this paper, we first perform an insightful exploratory analysis to exploit the transfer phenomenon of financing needs among SMEs, which motivates us to fully exploit the multi-relation enterprise social network for boosting the effectiveness of FNE. The main challenge lies in modeling two kinds of heterogeneity, i.e., transfer heterogeneity and SMEs’ behavior heterogeneity, under different relation types simultaneously. To address these challenges, we propose a graph neural network named Multi-relation tRanslatIonal GrapH aTtention network (M-RIGHT), which not only models the transfer heterogeneity of financing needs along different relation types based on a novel entity—relation composition operator but also enables heterogeneous SMEs’ representations based on a translation mechanism on relational hyperplanes to distinguish SMEs’ heterogeneous behaviors under different relation types. Extensive experiments on two large-scale real-world datasets demonstrate M-RIGHT’s superiority over the state-of-the-art methods in the FNE task.

The Internet of Things (IoT) has been widely deployed to support versatile applications, where an application can be satisfied by functionally compatible and non-functionally satisfiableIoTservices. Considering the fact that the capacities ofIoTdevices may change dynamically, whether or not, and to what extent, certain constraints can be satisfied during their execution, are to be explored. This observation motivates us to formalize the interpretation of qualitative and quantitative satisfaction for prescribed constraints, and thus, to achieveIoTservice monitoring at runtime. Specifically, we formulate the problem ofIoTservice monitoring as a constraint satisfaction problem, where multiple constraints, including spatial-temporal constraints, energy limitation, and capacity restrictions, are considered. Specification-based monitoring is developed based onSignalTemporalLogic (STL), where a novel accumulativerobustnessmetric is proposed, denotedAccumulativeSTL(ASTL), to emphasize the robust satisfaction over the entire time domain. Thereafter,IoTservice monitoring is converted toASTLformulae, and its constraint satisfaction is interpreted with qualitative and quantitative semantics at runtime. Case studies and extensive evaluations are conducted upon publicly-available datasets, where various influential factors are considered. Experimental results show that ourASTLperforms better than the state-of-the-art's techniques with more robust satisfaction.

With the recent adoption of edge computing,Internet ofThings (IoT) devices collaborate at the network edge to facilitate edge-native applications. In this setting,IoTdevices are typically encapsulated asIoTservices to encode their functionalities, and their collaboration is achieved throughIoTservice composition. Due to the continuous resource occupancy, release, and consumption ofIoTdevices at runtime, a composition, which is functionally compatible and non-functionally optimal at this moment, may not hold in the forthcoming time durations, when certainIoTservices may significantly downgrade in theirQuality-of-Services (QoS). To guarantee the compatibility of compositions withQoSvariations, this article proposes an adaptive composition mechanism leveragingComputationTreeLogic (CTL) specifications. Specifically, we formalize the composition as a temporal task, and convert it toCTLformulae with the abstractions of required functionalities and composite structures. Functional compatibility is formally interpreted byCTLsemantics during the execution of compositions. Besides, we construct aQoSDependencyGraph (QoSDG) to captureQoSvariations, and achieve adaptive composition with dynamicQoSsatisfactions. Extensive experiments are conducted upon publicly-available datasets, and comparison results demonstrate that our technique outperforms the state-of-the-art counterparts in heterogenous scenarios with higherQoSdependencies ranging from 0.3$\%$to 27.8$\%$.

With the service-oriented encapsulation of Internet of Things (IoT) devices, IoT services, which are functionally compatible and non-functionally satisfiable, are composed to support domain applications. The execution of IoT services may last for a relatively long time duration in which their capacities may vary significantly. In this setting, whether or not, and to what extent, certain constraints specified upon certain IoT services can always be satisfied during their execution, are to be explored. This observation motivates us to formalize the interpretation of qualitative and quantitative satisfaction for prescribed constraints, and thus, to conduct IoT service monitoring at runtime. Specifically, the requirement of IoT service monitoring is formulated as a constraint satisfaction problem. Specification-based monitoring is developed leveraging Signal Temporal Logic (STL), where a novel accumulative robustness metric, denoted Accumulative STL (ASTL), is proposed to emphasize the robust satisfaction over the entire time domain. Hence, an ASTL-based mechanism is proposed to support IoT service monitoring, where prescribed constraints are converted to ASTL formulae, and interpreted with qualitative and quantitative semantics at runtime. Case studies and extensive evaluations are conducted upon publicly-available datasets. Experimental results show that ASTL performs better than the state-of-the-art techniques with more robust satisfaction.

The collaboration of contiguous Internet of Things (IoT) devices is envisioned to satisfy complex applications which are beyond the capacity of single devices. The functionalities of IoT devices are encapsulated as IoT services, and their collaboration is implemented in terms of IoT service composition. Considering the capacity occupancy, release, and consumption caused by the implementation of IoT services, their composition is challenging in capacity-dynamically fluctuating IoT networks. This paper proposes a dynamic IoT service composition mechanism with inter-service dependencies adopted to capture the dynamic changes of IoT devices, and this change is specified by various Quality-of-Service factors. IoT service composition is formalized under Computation Tree Logic specification with certain composite structures and dynamic dependencies, and this composition is formally achieved by an optimized model checking method. Extensive experiments are conducted on publicly available datasets, and evaluation results show that our technique outperforms the state-of-the-art's approaches in relevant performance metrics.

In order to better serve the safety protection of urban residents during cycling, the KANO model is used to analyze the problems encountered by users wearing helmets during cycling. Combined with QFD, the weighted average analysis of user needs and the importance of engineering measures of product technical characteristics is carried out and users are found. We used TRIZ innovative invention principles to analyze the key requirements and converted them into TRIZ standard problems. We used the corresponding invention measures to improve the existing problems of the helmet and verified the rationality of the design through finite element analysis. Through the KANO-QFD model combined with the TRIZ innovative invention principle, the helmet is optimized to improve its safety.

This paper proposes to derive temporal constraints and granularities corresponding to individual activities, collaborative activities and their connecting edges from event logs. Specifically, a timed hierarchical business process model is constructed. Temporal anomalies are measured with time-constrained and granularity-aware bounds according to user's acceptance of deviant executions. Temporal interestingness, as the complement to anomaly detection, is evaluated as the most probable execution times that are partitioned into user-defined granules and ranked by probability. Experimental evaluations upon public event logs demonstrate the effectiveness and applicability of our proposed model for temporal anomaly and interestingness detection in terms of accuracy and recall, in comparison with the state-of-art`s techniques.

Device-free hand gesture is one of the most natural ways to interact with everyday objects. However, existing WiFi-based gesture recognition solutions are typically restricted to indoor environments due to limited outdoor coverage. Furthermore, to achieve high sampling rates, they may interfere with normal data transmissions. In this paper, we aim to develop a robust dynamic gesture interaction system that can be ubiquitously deployed using Long-term Evolution (LTE) mobile terminals. Through both empirical studies and in-depth analysis using the Fresnel zone model, we reveal the key factors that contribute to the repeatability and discernibility of gestures. We show that the optimal location and orientation to perform gestures indeed exist and can be identified without prior knowledge of the position of LTE base stations (BSs) relative to a terminal. Guided by the design principles derived from Fresnel zone characteristics around a 4G terminal, we design highly repeatable and discernible gestures with salient received signal profiles. A gesture interaction system has been developed and implemented to achieve robust recognition with this careful design. Extensive experiments have been conducted in both indoor and outdoor environments, for different relative placements of mobile terminal and BS, and with different users. The proposed system can automatically identify the direction of BSs with a median error of less than 15 degrees and achieve gesture recognition accuracy as high as 98% in all scenarios without the need to acquire any training data.

This paper proposes a multi-request cooperative-integrating mechanism to optimize concurrent multi-applications in service-oriented wireless sensor networks (WSNs). Specifically, a sensor node is encapsulated as one or multiple WSN services, which can be categorized into service classes. A service network is constructed by considering the invocation relationship between service classes. Candidate service class chains are recommended. These service classes chains will be instantiated by available WSN services, which can be reduced to a multi-objective and multi-constraint optimization problem, where the spatial-and temporal-constraints, and energy efficiency of the network, are taken into consideration. This combinational optimization problem is solved by adopting heuristic algorithms. Experimental results show that this technique improves the shareability and energy efficiency for supporting concurrent applications.

Hosting multiple applications in a shared infrastructure of wireless sensor networks is a trend nowadays, and sharing sensory data for answering concurrent applications is a promising and energy-efficient strategy. To address this challenge, this paper proposes an energy-efficient query optimization mechanism for supporting multiple concurrent applications leveraging our two-tier cooperative caching mechanism. Specifically, query requests for concurrent applications are represented as binary strings, which are reduced to a single one for avoiding the reprocessing of shared subquery requests. This reduced query request is answered through our cooperative caching mechanism, where sensory data, which are highly possible to be reused for answering forthcoming query requests, are cached at the sink node (SN). Besides, the gray model GM(1, 1) is adopted for forecasting sensory data units which may be interested mostly by forthcoming query requests. These units of sensory data may be prefetched from the network and cached at the SN. Experimental evaluation shows that this approach can reduce the energy consumption significantly, and improve the network capacity to an extent, especially when the number of concurrent query requests is relatively large.