In intelligent transportation systems (ITS), real-time vehicle detection based on aerial images is crucial for effective traffic monitoring and decision-making. However, detecting small vehicles with varying orientations in complex backgrounds remains technically challenging, as existing models often struggle to balance the requirements of detection accuracy and computational efficiency. In this letter, we introduce the Vehicle Detection eXtended Network (VDXNet), a lightweight model that is capable of achieving high detection performance while minimizing computational complexity. VDXNet incorporates the novel Residual Cross Depth Fusion (RxDF) module to enhance feature extraction in the backbone. Furthermore, it employs newly proposed Lightweight Feature Pyramid Pooling (LiteFPP) and Channel Reduction downsampling (CRDown) modules to support multi-scale detection and spatial dimensionality reduction. These innovations streamline the model’s neck, reducing complexity while ensuring accurate detection of vehicles across diverse scales, angles, and backgrounds. Evaluations on the UCAS-AOD, VEDAI, UAV-ROD, and UAVDT datasets demonstrate that VDXNet achieves substantial reductions in model complexity, with 1.608M parameters (a decrease of 37.72%) and 5.9 GFLOPs (a decrease of 6.35%) compared to the YOLO11n model. Despite these efficiency gains, VDXNet also improves mAP by 0.52%, achieving 96.3% mAP on the UCAS AOD dataset.

Integrated Development Environments (IDEs) are pivotal in enhancing productivity with features like code completion in modern software development. Recent advancements in Natural Language Processing (NLP) have empowered neural language models for code completion. In this study, we present an extensive investigation of the impact of open and closed vocabulary systems on the task of code completion. Specifically, we compare open and closed vocabulary systems with various vocabulary sizes to observe their impact on code completion performance. We experiment with three different open vocabulary systems: byte pair encoding (BPE), WordPiece and Unigram to compare them with closed-vocabulary systems to analyze their modeling performance. We also conduct experiments with different context sizes to study their impact on code completion performance. We have experimented using various prominent language models, including one from recurrent neural networks and five from transformers. Our results indicate that vocabulary size significantly impacts modeling performance and can artificially boost the accuracy of code completion models, especially in the case of a closed-vocabulary system. Moreover, we find that different vocabulary systems have varying impacts on token coverage, whereas open-vocabulary systems exhibit better token coverage. Our findings offer valuable insights for building effective code completion models, aiding researchers and practitioners in this field.

Leakage and tampering problems in collection and transmission of biomedical data have attracted much attention as these concerns instigates negative impression regarding privacy, security, and reputation of medical networks. This article presents a novel security model that establishes a threat-vector database based on the dynamic behaviours of smart healthcare systems. Then, an improved and privacy-preserved SRU network is designed that aims to alleviate fading gradient issue and enhance the learning process by reducing computational cost. Then, an intelligent federated learning algorithm is deployed to enable multiple healthcare networks to form a collaborative security model in a personalized manner without the loss of privacy. The proposed security method is both parallelizable and computationally effective since the dynamic behaviour aggregation strategy empowers the model to work collaboratively and reduce communication overhead by dynamically adjusting the number of participating clients. Additionally, the visualization of the decision process based on the explainability of features enhances the understanding of security experts by enabling them to comprehend the underlying data evidence and causal reasoning. Compared to existing methods, the proposed security method is capable of thoroughly analyzing and detecting severe security threats with high accuracy, reduce overhead and lower computation cost along with enhanced privacy of biomedical data.

Studies have substantiated the efficacy of deep learning-based models in various source code modeling tasks. These models are usually trained on large datasets that are divided into smaller units, known as tokens, utilizing either an open or closed vocabulary system. The selection of a tokenization method can have a profound impact on the number of tokens generated, which in turn can significantly influence the performance of the model. This study investigates the effect of different tokenization methods on source code modeling and proposes an optimized tokenizer to enhance the tokenization performance. The proposed tokenizer employs a hybrid approach that initializes with a global vocabulary based on the most frequent unigrams and incrementally builds an open-vocabulary system. The proposed tokenizer is evaluated against popular tokenization methods such as Closed, Unigram, WordPiece, and BPE tokenizers, as well as tokenizers provided by large pre-trained models such as PolyCoder and CodeGen. The results indicate that the choice of tokenization method can significantly impact the number of sub-tokens generated, which can ultimately influence the modeling performance of a model. Furthermore, our empirical evaluation demonstrates that the proposed tokenizer outperforms other baselines, achieving improved tokenization performance both in terms of a reduced number of sub-tokens and time cost. In conclusion, this study highlights the significance of the choice of tokenization method in source code modeling and the potential for improvement through optimized tokenization techniques.

The security of industrial control systems (ICSs) against cyber-attacks is essential in modern era since ICSs are vital constituent of modern societies and smart cities. However, the augmentation of legacy ICS networks with smart computing and networking technologies [such as Internet of Things (IoT)] has intensely enlarged the surface of attacks against these critical infrastructures. This augmentation makes these networks more vulnerable to cyber-attacks and despite the current security solutions, attackers still find ways to proliferate these networks. The intrusion detection system (IDS) is one of the key security aspect to prevent these networks from contemporary cyber-attacks. Therefore, this article proposes a new IDS model named federated-simple recurrent units (SRUs) for the security of IoT-based ICSs. Specifically, the federated-SRUs IDS model uses an improved simple recurrent units architecture to reduce computational cost and alleviate the gradient vanishing issue in recurrent networks. Then, it performs data aggregation through several communication rounds in the federated architecture which allows multiple ICS networks and stakeholders to build a comprehensive IDS model in a privacy-preserving manner. The performance of the federated-SRUs IDS model is validated through experiments using real-world gas pipeline-based ICS network data, which indicates that it is able to accurately detect intrusions in real time without compromising privacy and security. Experiments also verify that the federated-SRUs model outperforms existing state-of-the-art approaches and thus can serve as a viable IDS method in IoT-based ICS networks.

The management of contemporary communication networks of supply chain (SC) 4.0 is becoming more complex due to the heterogeneity requirements of new devices concerning the integration of the Internet of Things in the legacy industry networks. Hence, it becomes a challenging task to secure networks of SC 4.0 from cyber-attacks and provide a robust and efficient defence framework that can resist sophisticated attacks. Machine learning-based intelligent detection algorithms are often trained at either a centralized or single server, which makes it difficult to train an effective model and also it violates privacy concerns if gathering data from other servers at the edge. Classical machine learning approaches function on the legacy group of data placed on a central or single server, which brands it the least favored choice for supply chain networks, with data privacy issues. To address these problems, this article proposes a federated learning-based efficient detection model named, DFF-SC4N, to proactively identify intrusions from SC 4.0 networks using distributed local data training. DFF-SC4N uses communication rounds in a federated learning manner having gated recurrent units by only sharing the learned parameters and keeps the data intact on local servers. The accuracy of the global model is optimized by an aggregating model, which updates from multiple servers and multiple SC 4.0 networks. Extensive experiments on real industrial network data demonstrate that the DFF-SC4N outperforms both centralized training models and state-of-the-art peer methods in protecting SC 4.0 networks.

Device Classification (DC) is one of the critical network management means in the Internet of Things (IoT). Most machine learning-driven DC methods are implemented in two steps: firstly, extracting device features, and then training the classifier. Thus, for the same classifier, the quality of device features plays a decisive role. However, for one thing, these methods usually merely consider device traffic or device attributed features, ignoring the complex relationship between devices in Social IoT. For another thing, these features fed to the classifier are often selected manually without automatic learning. These two factors seriously affect the performance and flexibility of the algorithm. To address the above issues, we propose a novel Graph Contrastive Neural Network-based algorithm for DC dubbed GCNNDC. Specifically, based on the complex relationship between heterogeneous objects in Social IoT and the attribute information of the device itself, we first construct four kinds of homogeneous device-device attributed networks as the input of the neural network. Then, we train a novel GCNN model to learn informative device features in a self-supervised learning fashion. Finally, we leverage the features to train a classifier to complete the DC task. Experimental results on real-world IoT network dataset demonstrate the superiority of GCNNDC.

The Internet is a vital part of today’s ecosystem. The speedy evolution of the Internet has brought up practical issues such as the problem of information retrieval. Several methods have been proposed to solve this issue. Such approaches retrieve the information by using SPARQL queries over the Resource Description Framework (RDF) content which requires a precise match concerning the query structure and the RDF content. In this work, we propose a transfer learning-based neural learning method that helps to search RDF graphs to provide probabilistic reasoning between the queries and their results. The problem is formulated as a classification task where RDF graphs are preprocessed to abstract the N-Triples, then encode the abstracted N-triples into a transitional state that is suitable for neural transfer learning. Next, we fine-tune the neural learner to learn the semantic relationships between the N-triples. To validate the proposed approach, we employ ten-fold cross-validation. The results have shown that the anticipated approach is accurate by acquiring the average accuracy, recall, precision, and f-measure. The achieved scores are 97.52%, 96.31%, 98.45%, and 97.37%, respectively, and outperforms the baseline approaches.

Studies have confirmed the robust performance of machine learning classifiers for various source code modeling tasks. In general, machine learning approaches are incapable of handling imbalanced datasets, since they are sensitive to the choice of diverse classes. Therefore, these approaches may lean towards the classes with a large percentage of observations. In this work, we investigate and explore the impact of balanced and imbalanced learning on source code suggestion task otherwise known as code completion, covering a large number of imbalanced classes. We further explore the impact of vocabulary size on modeling performance. First, we provide the essentials to formulate the problem of source code suggestion as a classification task and investigate the level of imbalanced classes. Second, we train the four most adapted neural language models as a baseline to assess the modeling performance. Third, we impose two diverse class balancing techniques, TomekLinks and AllKNN, to balance the datasets and evaluate their impact on the modeling performance. Finally, we trained these models with a weighted imbalanced learning approach and compared the performance with balanced learning approaches. Additionally, we train models by varying the vocabulary size to study their impact. In total, we trained 230 models on 10 real-world software projects and extensively evaluated these models with widely used performance metrics such as Precision, Recall, FScore, mean reciprocal rank (MRR), and Receiver operating characteristics (ROC). Additionally, we employed ANOVA statistical analysis to study the statistical significance and differences between these approaches. This study has demonstrated that the modeling performance decreases during balanced model training, whereas the weighted imbalance training produces comparable results and is more efficient in terms of time cost. Additionally, this study exhibits that a large size of vocabulary does not necessarily improve the modeling performance when out-of-vocabulary predictions are disregarded.

Industrial Internet of Things (IIoT) and Industry 4.0 empower interrelation among manufacturing processes, industrial machines, and utility services. The time-critical data collected from heterogeneous sensing devices are usually communicated to processing points for analysis and aggregation as the basis of IIoT. The IIoTs’ service quality typically depends on data integrity and accuracy, which could be exploited by injecting malicious events, such as false data injection and data poisoning attacks. Thus, effective anomaly recognition and explanation are critical for ensuring quality services and empowering security administrators to interpret the causal reasoning of prediction decisions and underlying data evidence. This study proposes an autoencoder-based detection framework using convolutional and recurrent networks to discover cyber threats in IIoT networks and explain the model. A two-step sliding window (SW) is applied to learn the latent representations of data features better. Malicious points from the raw time series are transformed into fixed-length series through the first-step SW. Every series is converted into continuous-time-reliant subseries via another smaller SW to learn latent representations of malicious events. Fully connected networks use the extracted temporal and spatial features for the classification and explanation of attack events. The empirical results revealed that this framework effectively extracts features that include contexts of malicious patterns. This demonstrated that the proposed framework is robust in detecting malicious events using multiple evaluation metrics and outperforming the contemporary state-of-the-art methods, indicating its suitability as an operative application method in real-world IIoT-based networks.

Intelligent Transportation Systems (ITS), particularly Autonomous Vehicles (AVs), are susceptible to safety and security concerns that impend people’s lives. Nothing like manually controlled vehicles, the safekeeping of communications and computing constituents of AVs can be threatened using sophisticated hacking techniques, consequently disrupting AVs from the operative usage in our daily life routines. Once manually controlled vehicles are linked to the Internet, so-called the Internet of Vehicles (IoVs), they would be misused by cyberattacks. In this paper, we present a multi-stage intrusion detection framework to identify intrusions from ITSs and produce low rate of false alarms. The proposed framework can automatically distinguish intrusions in real-time. The proposed framework is based on normal state-based and a deep learning-centered bidirectional Long Short Term Memory (LSTM) architecture to efficiently discover intrusions from the fundamental network gateways and communication networks of AVs. The designed framework is evaluated through two benchmark datasources, that is, the UNSWNB-15 datasource for exterior network communications and the car hacking datasource for in-vehicle communications. The outcomes indicated that the proposed framework achieves high performance that outperforms various current state-of-the-art systems with an accuracy rate of 98.88% for the UNSWNB-15 dataset and 99.11% for the car hacking dataset. Besides, the proposed framework is furthermore capable to detect zero-day (concealed) outbreaks from IoVs networks.

Biological network classification is an eminently challenging task in the domain of data mining since the networks contain complex structural information. Conventional biochemical experimental methods and the existing intelligent algorithms still suffer from some limitations such as immense experimental cost and inferior accuracy rate. To solve these problems, in this paper, we propose a novel framework for Biological graph classification named Biogc, which is specifically developed to predict the label of both small-scale and large-scale biological network data flexibly and efficiently. Our framework firstly presents a simplified graph kernel method to capture the structural information of each graph. Then, the obtained informative features are adopted to train different scale biological network data-oriented classifiers to construct the prediction model. Extensive experiments on five benchmark biological network datasets on graph classification task show that the proposed model Biogc outperforms the state-of-the-art methods with an accuracy rate of 98.90% on a larger dataset and 99.32% on a smaller dataset.

The resource description framework (RDF) was adopted by the World Wide Web (W3C) as an essential semantic web standard and the RDF scheme. It accords the hard semantics in the description and wields the crisp metadata. However, it usually produces vague or ambiguous information. Consequently, fuzzy RDF helps deal with such special data by transforming the crisp values into a fuzzy set. A method for analyzing fuzzy RDF data is proposed in this paper. To this end, first, we decompose the RDF into fuzzy RDF variables. Second, we are designing a model for global sensitivity analysis based on the decomposition of fuzzy RDF. It figures out the ambiguities of fuzzy RDF data. The proposed global sensitivity analysis model provides the importance of fuzzy RDF data by considering the response function’s structure and reselects it to a certain degree. A practical tool for sensitivity analysis of fuzzy RDF data has also been implemented based on the proposed model.