In-service Research Articles

IoT and Machine Learning Based Attacks Detection Model on Wearable Health Care Devices

The Internet of Things (IoT) in healthcare is becoming more and more popular in the field of research aimed at improving the effectiveness of intelligent healthcare networks and applications. Nonetheless, distinct risks affect the security and privacy of data in smart health (S-Health). IoT enables healthcare professionals to engage with patients more proactively and with greater vigilance. Smart gadgets with tiny sensors attached to them that communicate with one another to track each other's performance are part of the Internet of Things. To defend S-Health from MITM attacks. The suggested method employs two layers of machine learning algorithms for attack detection and security mechanisms, including low-cost access policies for SHRs (Smart Health Records), lightweight IoT detection schemes, and timely detection of to lessen their impact on the network. According to simulation data, the suggested Hybrid ML performs better than current methods and has a higher attack detection rate overall. The main goal of this research article is to develop an attack detection technique.

Allora: a Self-Improving, Decentralized Machine Intelligence Network

Recent advances in data access and computing power have enabled the first forms of machine intelligence capable of offering meaningful insights. However, the tremendous resources required have caused these solutions to be closed and siloed by industry monoliths. To achieve the full potential of machine intelligence, the data, algorithms, and participants must be maximally connected. Network solutions are needed, and the decentralized nature of blockchain technology is ideal for solving this problem. We introduce Allora, a self-improving, decentralized machine intelligence network that surpasses the capabilities of its individual participants by design. Allora achieves this through two main innovations. First, it lets network participants forecast each other's performance under the current conditions, thereby creating a form of context-awareness that enables the network to achieve the best inferences under any circumstances. Second, it introduces a differentiated incentive structure that rewards network participants for their unique contribution to the network goal, tailored to their specific task and purpose, avoiding any distracting incentives. We show that these innovations make Allora's inferences considerably more accurate than before. With its versatility and accessibility, Allora paves the way for machine intelligence to become fully commoditized and integrated with the economy, technology, and society.

The Potential of Neural Networks in Teacher Preparation for Foreign Language Classes and Educational Content Creation

Artificial intelligence (AI) technologies are now being extensively used in various fields of human activity, including science and education. Neural networks, which are AI algorithms, are widely used in language learning. The relevance of this study is due to the need to study the potential of neural networks in the process of preparing a teacher of higher school for foreign language classes. Using neural networks, the university teacher demonstrates the ability not only to choose teaching materials but also to adapt them to the real learning environment, as well as to create new materials, taking into account various available resources and educational needs. The use of neural networks helps to optimize the process of teacher’s preparation for classes. This study aims to analyze the didactic potential of Twee and Perplexity AI neural networks, as well as to examine their use in the preparation for foreign language classes and the creation of educational content for students in various fields of professional training. The research methods we applied involved analysis, comparison, generalization of scientific literature, observation, study of experience and products of activity, analysis of educational literature, modeling of educational process. The scientific novelty of the study is the development of methods of using Twee and Perplexity AI neural networks in the preparation of a set of exercises on the formation of main types of speech activity among non-linguistic students. The conclusion can be drawn that it is necessary to further develop the methodology of teaching foreign languages with the help of artificial intelligence technologies which would help in generating the unique educational content and a set of communication exercises. Twee and Perplexity AI neural networks are effective pedagogical and didactic tools in teacher preparation for foreign language classes. Keywords: artificial intelligence, Twee and Perplexity AI neural networks, creation of educational content, methods of foreign language teaching, AI technologies application

Focusing Forward: How to Traverse New Leadership While Emphasizing Continuity

Abstract: This article provides best practices for navigating leadership transitions when new senior executives are hired, drawing on research regarding organizational change, leadership, and human psychology to understand common reactions to change and offer evidence-based strategies. It recommends proactively reducing uncertainty through over-communicating regularly with varied forums, tailoring updates for different stakeholder groups, highlighting complementary strengths rather than differences in strategy or priorities, and addressing core human needs for security and belonging. A case study demonstrates these concepts as a director successfully handled new C-suite hires by holding weekly Q&As, discussing concerns privately with the new CTO, leveraging advisory networks for intelligence on new leaders’ focus, reassuring teams existing work remained strategically important, and emphasizing synergies between current and new areas of focus. Key implications explain how framing changes as natural evolution rather than disruption builds buy-in, and that with clarity of purpose and commitment to organizational mission, leaders can guide teams productively through transitions by managing both information and the human side of change.

Development and Evaluation of the DMIND Questionnaire: Preparing for AI Integration into an Effective Depression Screening Tool

Objective: Thailand’s mental health crisis is exacerbated by high demand and a shortage of mental health professionals. The research objective was to develop and validate the Detection and Monitoring Intelligence Network for Depression (DMIND) questionnaire, designed to be culturally relevant and easily administered in clinical settings. Crafted with expert input, items specifically conducive to artificial intelligence (AI) analysis were selected to facilitate the future development of an AI-assisted depression scoring model. This approach underscores the tool’s dual utility in both human-led and technology-enhanced diagnostics. Materials and Methods: We enrolled 81 participants from psychiatric and tertiary care hospitals in Bangkok. Participants were assessed using the DMIND questionnaire, followed by the Hamilton Depression Rating Scale (HDRS-17). Statistical analyses included the content validity index (CVI), Cronbach’s alpha, Pearson’s correlation coefficient, Cohen’s kappa, and receiver operating characteristic (ROC) analysis. The Liu method, Youden index, and nearest neighbor method were used to determine the optimal cut-off point Results: The DMIND questionnaire showed strong validity, with an item-level CVI (I-CVI) and scale-level CVI (S-CVI) exceeding 1.0, indicating strong consensus on its relevance and utility. The tool also demonstrated high internal consistency (Cronbach’s alpha = 0.96). ROC analysis showed an AUC of 0.88, indicating high accuracy in depression screening. An optimal cut-off score of 11.5 was identified, balancing predictive value and sensitivity. Conclusion: The DMIND questionnaire represents a significant advancement in innovative mental health diagnostics, addressing unmet clinical needs by providing accurate and efficient assessments capable of AI integration for further enhancing mental health service delivery in Thailand.

Neuro-computing analysis of model-based Casson hybrid nanofluid flow via three-dimensional radiative Riga plate with irregular heat source/sink

In this study, an analysis is conducted on a hybrid nanofluid using the Yamada-Ota and Hamilton-Crosser model in combination with the Casson fluid model. The hybrid nanofluid is analyzed in the context of a three-dimensional radiative Riga plate, which incorporates non-uniform heat source/sink effects. The thermal features and their behavior of the hybridized nanofluid is explored in the current investigation considering the flow over a radiative Riga plate with the impact of space and temperature dependent heat source/sink. The flow and thermal characteristics of the hybridized nanomaterial are presented for the consideration of Yamada-Ota & Hamilton Crosser models. Additionally, the analysis is enhanced by the implementation of Artificial neural network. Further the machine learning approach, a neuro-computing intelligent network is adopted for the computational techniques. The neural networks, which effectively uses the data and make predictions based on their acquired knowledge. The outcomes of the proposed analysis provide the flow behavior of the hybrid nanofluid. The effect of non-uniform heat source/sink of the Casson fluid model contributes to a greater hike in the fluid temperature. Moreover, the neuro-computing intelligent network shows efficient predictions of the involved thermal parameters.

3D mineral prospectivity modeling at the Axi epithermal gold deposit, NW China by using a feature adaptive fusion strategy

Mineralization distribution is always intricately affected by multiple ore-controlling geological units that play different roles in a mineral system (e.g., driver, trap, and throttle). How to effectively balance and integrate ore-controlling features from various 3D geological models during 3D mineral prospectivity modeling (MPM) is still a challenging task. In this paper, we introduce a novel approach, the feature adaptive fusion convolutional neural networks (CNN), which is designed to learn multiple 3D geological models with ore-controlling functions. The method is validated in the Axi epithermal gold deposit, northwestern China that mineralization distribution is jointly controlled by fault, volcanic phase, and phyllic alteration. The geology units are firstly constructed by explicit-implicit modeling and their ore-controlling features are subsequently described by high-frequency Laplace-Beltrami eigenfunctions and reassembled into multi-channel images as input to CNN. To learn the differences in ore-controlling effects among various geological units, we designed a fully connected layer to achieve adaptive quantification and weighted integration of the ore-controlling features by automatically optimizing weight allocation parameters and bias vectors using the neural network intelligence. Comparison results between the proposed method and other prospectivity methods suggest that the feature adaptive fusion CNN produces more reliable predictions, characterized by: (1) high consistency with known mineralization, (2) the highest AUC value and success rate, and (3) accurate prediction of deep voxels explored by drilling. Therefore, the proposed method effectively integrates the ore-controlling effects of multiple geological units and is suitable for complex scenarios of 3D MPM. Utilizing the prospectivity results generated by our method, we identified five potential mineralization in the Axi gold deposit, laying a robust foundation for future gold exploration.

Distributed virtual selective-forwarding units and SDN-assisted edge computing for optimization of multi-party WebRTC videoconferencing

Network service providers (NSP) have growing interest in placing network intelligence and services at network edges by deploying software-defined network (SDN) and network function virtualization infrastructure. In multi-party WebRTC videoconferencing using scalable video coding, a selective forwarding unit (SFU) provides connectivity between peers with heterogeneous bandwidth and terminals. An important question is where in the network to place the SFU service in order to minimize end-to-end delay between all pairs of peers. Clearly, there is no single optimal place for a cloud SFU for all possible peer locations. We propose placing virtual SFUs at network edges leveraging NSP edge datacenters to optimize end-to-end delay and usage of overall network resources. The main advantage of the distributed edge-SFU framework is that each peer video stream travels the shortest path to reach other peers similar to mesh connection model, whereas each peer uploads a single stream to its edge-SFU avoiding the upload bottleneck. While the proposed distributed edge-SFU framework applies to both best-effort and managed service models, this paper proposes a premium managed, edge-integrated multi-party WebRTC service architecture with bandwidth and delay guarantees within access networks by SDN-assisted slicing of edge networks. The performance of the proposed distributed edge-SFU service architecture is demonstrated by means of experimental results.

СКЛАДОВІ ПОТУЖНОСТІ ТИРИСТОРНОГО РЕГУЛЯТОРА НАПРУГИ З АКТИВНИМ НАВАНТАЖЕННЯМ

A study of the power balance of the semiconductor voltage regulator on an active load with phase control by two-operation thyristors was carried out. The suitability of its use for simultaneous obtaining of the reactive power resource for distribution electric networks has been confirmed. According to the concept of intelligent electrical networks, such devices belong to the class of flexible alternating current transmission systems, and they can be the basis for building virtual energy sources. The power components for individual components of the regulator, including the thyristor block, were studied. The spectral method of describing currents and voltages at the regulator sections is applied. The obtained results of theoretical studies were confirmed by an example with numerical calculations, which made it possible to clarify the physical essence of energy transformations during voltage regulation. Bibl. 8, fig. 3.

Datadog with Zigbee Wireless Communication Network Protocol for an Internal Implementation in an Educational Institution

Wireless Sensor Networks are the most powerful technique for monitoring Environmental factors. As the role of WSN relies totally on the service life of the sensor nodes, it is necessary to have complete monitoring of networks. Energy efficiency has always been a key concern for Wireless sensor networks. This paper presents a detailed literature review that examines how an artificial intelligence networking tool(datalog)can be used with ZIGBee-based network protocol for continuous Real-time data quality monitoring to detect bad data quality issues in an educational organization It also Tracks and improves application speed by following requests from beginning to end and monitoring application performance. By using AI algorithms for routing choices and optimization methods, the study aims to improve network performance, energy efficiency, and system scalability. The analysis of different routing strategies with AI implementations will be covered, emphasizing the potential advantages and challenges of this innovative approach in an organization.

Revolutionizing Network Intelligence: Innovative Data Mining and Learning Approaches for Knowledge Management in Next-Generation Networks

The Information and Communication (ICT) industry, a global giant among service sectors, is known for both its massive scale and its unforgiving demands. Here, downtime is unacceptable, requiring constant high availability – often at the stringent Sigma Six standard. Redundancy is a common solution, but it comes at a cost. To meet these demands proactively, the ability to predict load and growth becomes crucial. This project aims to develop a prototype, or proof of concept, that utilizes data mining to provide early warnings and growth forecasts for the ICT industry with good accuracy. Big data is key to making discoveries in any data analysis project. Normally, this data comes from real-time system logs. However, for this initial test, I used a dataset called MIT Reality Mining. This dataset is useful because real-world companies, especially in the tech industry (ICT), are often hesitant to share their current information. By using MIT Reality Mining, I could still find trends and potential reasons behind them in the ICT industry. It's important to remember that this is a limited functionality prototype. While it can serve as a guideline for Telcos looking to implement data warehouses, the actual implementation details will need to adapt to the specific needs of each industry.

Drone image recognition and intelligent power distribution network equipment fault detection based on the transformer model and transfer learning

Drone image recognition and intelligent power distribution network equipment fault detection based on the transformer model and transfer learning

The Guardian Node Slow DoS Detection Model for Real-Time Application in IoT Networks.

The pernicious impact of malicious Slow DoS (Denial of Service) attacks on the application layer and web-based Open Systems Interconnection model services like Hypertext Transfer Protocol (HTTP) has given impetus to a range of novel detection strategies, many of which use machine learning (ML) for computationally intensive full packet capture and post-event processing. In contrast, existing detection mechanisms, such as those found in various approaches including ML, artificial intelligence, and neural networks neither facilitate real-time detection nor consider the computational overhead within resource-constrained Internet of Things (IoT) networks. Slow DoS attacks are notoriously difficult to reliably identify, as they masquerade as legitimate application layer traffic, often resembling nodes with slow or intermittent connectivity. This means they often evade detection mechanisms because they appear as genuine node activity, which increases the likelihood of mistakenly being granted access by intrusion-detection systems. The original contribution of this paper is an innovative Guardian Node (GN) Slow DoS detection model, which analyses the two key network attributes of packet length and packet delta time in real time within a live IoT network. By designing the GN to operate within a narrow window of packet length and delta time values, accurate detection of all three main Slow DoS variants is achieved, even under the stealthiest malicious attack conditions. A unique feature of the GN model is its ability to reliably discriminate Slow DoS attack traffic from both genuine and slow nodes experiencing high latency or poor connectivity. A rigorous critical evaluation has consistently validated high, real-time detection accuracies of more than 98% for the GN model across a range of demanding traffic profiles. This performance is analogous to existing ML approaches, whilst being significantly more resource efficient, with computational and storage overheads being over 96% lower than full packet capture techniques, so it represents a very attractive alternative for deployment in resource-scarce IoT environments.

Research on the Subjective and Objective Evaluation Strategy of the Intelligent Vehicle Cockpit Based on the User Experience

The trend of intelligent networking in automobile industry is developing in full swing, with more and more functions. But feature richness does not necessarily reflect user needs. In order to perfect the evaluation system in the field of automobile intelligent cockpit, this paper constructs the subjective and objective evaluation strategy and tests the function of automobile intelligent cockpit. By building evaluation strategy, standardizing the tester, testing process pole test dimension, combining with subjective evaluation and eye movement, EEG device objective data, clear model and algorithm. Finally, the feasibility of the modified strategy is verified by a practical case.

Device-free Vehicle Type Identification for Next Generations Intelligent Transportation System

With the substantial progress in the automotive industry, intelligent vehicle identification systems have become imperative tools for next generation transportation system planning. The advent of internet of vehicles and intelligent cellular vehicular networks has entered the transportation system into a new era of technology. In smart transportation system of beyond 5G/6G vehicular networks, vehicle type identification is of vital importance for ensuring safe operability of autonomous driving environ ments, electronic toll, and traffic management. It can be used in high efficiency roadway services, vehicle theft prevention, road condition monitoring, automatic driving, access control security, etc. During the recent years, channel state information (CS I) of WiFi signal has gained escalating traction for activity or gesture recognition, and person identification with improved robustness and classification accuracy. Therefore, in this research work, a WiFi based vehicle type identification system is proposed for beyond 5G/6G vehicular networks. In this novel approach, we start from vehicle detection using the variation in CSI of WiFisignals caused by the moving vehicle. For vehicle detection we simply apply a threshold based algorithm and then extract meaningful features. Afterwards, the vehicle type is identified using machine learning algorithms. This is a wireless model that enables commercial WiFi devices to identify different types of vehicles using CSI measurements. The presented framework can identify six different types of vehicles with an accuracy of 91.8%

A Review of Exoplanet Detection Processes Using Artificial Intelligence and Neural Networks

A Review of Exoplanet Detection Processes Using Artificial Intelligence and Neural Networks

Campus Network Traffic Prediction and Anomaly Detection Based on Deep Learning

This paper proposes an intelligent solution for network traffic prediction and anomaly detection in campus networks, addressing the increasingly severe network security challenges. The proposed approach innovatively integrates Convolutional Neural Networks (CNN) and Long Short-Term Memory networks (LSTM) to simultaneously extract local features and capture dynamic temporal dependencies of network traffic, significantly improving prediction accuracy. Based on this, an adaptive threshold anomaly detection algorithm is designed to automatically adjust detection sensitivity according to traffic variations, achieving a better balance between accuracy and recall rates. Additionally, an anomaly visualization scheme is presented, intuitively displaying the spatiotemporal distribution of network anomalies through heatmaps, assisting administrators in decision-making. Large-scale experiments demonstrate that this approach can effectively identify various security threats such as DDoS attacks, scanning probes, and botnets, with an overall detection rate exceeding 90% while maintaining a low false positive rate. Compared to traditional statistical and machine learning methods, the proposed approach exhibits stronger adaptability and generalization capabilities, providing crucial support for building an intelligent, precise, and reliable campus network security protection system. Future work will focus on further improving the real-time performance and robustness of the solution, expanding its application in new network scenarios such as IoT and edge computing.

Road Network Intelligent Selection Method Based on Heterogeneous Graph Attention Neural Network

Selecting road networks in cartographic generalization has consistently posed formidable challenges, driving research toward the application of intelligent models. Despite previous efforts, the accuracy and connectivity preservation in these studies, particularly when dealing with road types of similar sample sizes, still warrant improvement. To address these shortcomings, we introduce a Heterogeneous Graph Attention Network (HAN) for road selection, where the feature masking method is initially utilized to assess the significance of road features. Concentrating on the most relevant features, two meta-paths are introduced within the HAN framework: one for aggregating features of the same road type within the first-order neighborhood, emphasizing local connectivity, and another for extending this aggregation to the second-order neighborhood, capturing a broader spatial context. For a comprehensive evaluation, we use a set of metrics considering both quantitative and qualitative aspects of the road network. On road types with similar sample sizes, the HAN model outperforms other models in both transductive and inductive tasks. Its accuracy (ACC) is higher by 1.62% and 0.67%, and its F1-score is higher by 1.43% and 0.81%, respectively. Additionally, it enhances the overall connectivity of the selected network. In summary, our HAN-based method provides an advanced solution for road network selection, surpassing previous approaches in terms of accuracy and connectivity preservation.

A Heterogeneous Glider System with Underwater Acoustic Communication and Positioning

Abstract Autonomous underwater vehicles are increasingly frequently used in the fields of ocean observation, data collection, and tactical surveillance. Underwater navigation and near real-time communication are the most important factors limiting the performance of these vehicles. In this article, an underwater acoustic communication and positioning system (UACPS) based on heterogeneous gliders is presented. The system consists of an acoustic wave glider and an acoustic underwater glider, and includes an ultrashort baseline and an acoustic modem. The program and communication protocol of the communication and positioning algorithm are independently developed, and the hardware circuit design of the communication and ultra-short baseline positioning, as well as the embedded surface wave gliders and gliders are independently implemented. Sea trial results indicate that the communication distance of the system is more than 3 km and the positioning error is less than 5%. With the further improvement of system performance, the low cost, long time, long distance and near real-time communication characteristics of our UACPS can be used to integrate multiple autonomous unmanned platforms as part of intelligent surveillance robot networks.

Риски и вызовы при внедрении искусственного интеллекта в систему высшего образования

<p style="text-align: justify;">The purpose of this article is to study possible risks and challenges in the higher education system that arise in connection with the growing speed of implementation of technologies, where artificial intelligence (AI) and neural network technologies are already part of its structure. The analysis is carried out with the aim of predicting the future nature of higher education, as well as exploring changes in the entire teaching paradigm under the influence of AI. The article identifies a number of expected problems when teaching students, in the pedagogical process, organizing their independent work, as well as in the management and administration of higher educational institutions when introducing AI capabilities into the process of higher education. The question is raised about the impact of restrictions on the development of Russian science, namely the disastrous consequences for its development due to the monopolization of AI algorithms and the subjective approach to managing a neural network by an organized technocratic group. In the course of analyzing the intersections of opposing issues of problematic issues and the positive dynamics of the influence of AI on the educational system of higher education, a vector has been set for further research on this topic.</p>