Journal of Intelligent Systems and Internet of Things

Today, every nation strives for international recognition in a variety of sports. Governments invest in games and sports to raise the performance of their teams and athletes to get notoriety. Numerous people are involved in sports execution, including team management, coaches, and biomechanists who monitor athlete fitness and work to achieve remarkable results. Performance analysis is greatly aided by technological integration in sports management. The performance analysis of athletes is evaluated in this research using an upgraded machine learning approach on Improved Machine Learning approach on Wearable Devices (IMLA-WD). This design strategy utilizes wearable devices to collect health data, which is then fed into a machine-learning model to monitor athletes' progress. The athletes' performance is evaluated using standard machine learning methods, and the deep neural network monitors their health status. With a health prediction accuracy of 98.65%, the statistical findings of the proposed model demonstrate the highest performance compared to existing methodologies.

Currently, Machine Learning (ML) seems very attractive since it may speed up business functions in enterprises, lower costs for supplying goods and services, and manage information to promote enterprise efficiency. Essential technological domains nowadays are the explosive period of growth in enterprise solutions, which are progressively used in almost all business platforms. The ML sessions will receive a thorough summary, and the relevant organizations will be shown procedures for relevant business processes. The data management unit is already been striving to solve related issues in ML applications for more than a generation, creating numerous customized analytical techniques. The approach described in the study uses a weighted directed graph displayed in an industrial environment to identify the core part of the neural network structure and then trains them using the relevant data source. The article proposed ML-assisted Enterprise Data Management (ML-EDM) for identifying the trade-off between ML growth in the financial sector and its consequences in precision and interpretability. According to the experimental findings, the ratio of AI for decision-making is 84.25%, the Speed and Agility proportion is 92.70%, the result of Earlier Prediction Management is 93.80%, the  Infrastructure Development is 85.46%, with Data Efficiency is 84.5% and Performance efficiency of the system is 90.14%.

Wireless Sensor Networks (WSNs) play a vital role in Industrial 4.0 by facilitating significant data collection for monitoring and control purposes. However, their distributed and resource-constrained nature makes WSNs vulnerable to Denial-of-Service (DoS) attacks, which can impede their normal operation and jeopardize their functionality. To address this issue, we propose a new machine learning (ML) approach that enhances the security of WSNs against DoS attacks in Industrial 4.0. Our approach incorporates a spatial learning unit, which captures the positional information in WSN traffic flows, and a temporal learning unit which captures time interdependency features within periods of traffic flows. To evaluate the proposed approach, we tested it on a publicly available dataset. The results demonstrate that it achieves a high detection rate while maintaining a low false alarm rate. Moreover, our Intrusion Detection System (IDS) exhibits good scalability and robustness against various DoS attacks. Our approach provides a reliable and effective solution to secure WSNs in Industrial 4.0 against DoS attacks and can be further developed and tested in various real-world scenarios.

Edge computing is a distributed computing paradigm that involves processing data at or near the edge of the internet of things (IoT) network, instead of centralized server. This makes the cyber-attacks increasingly sophisticated, and traditional security measures become no longer sufficient to protect against them. Concurrently, privacy concerns arise when sensitive data is involved in Edge computing applications containing confidential information. In this paper, we propose a privacy-preserved federated learning (FL) approach for cyber-attack detection in edge based IoT ecosystem. A novel lightweight convolutional Transformer network (LCT) network is designed to precisely identify cyber-attacks though learning attack patterns from IoT traffics in local edge devices, where model is personalized though fine-tuning. The privacy of model and data is preserved in our system via incorporating differential privacy and secure aggregation during the cooperative training process on edge devices. We evaluate our proposed approach on a real-world dataset of network traffic (NSL-KDD) containing various types of attacks, and the experimental results show that our personalized FL approach outperforms traditional FL in terms of detection accuracy. We also show that our approach is effective in handling non-stationary data and adapting to changes in the network environment.

Business intelligence (BI) mentions to the technical and procedural structure which gathers, supplies, and examines the data formed by company action. BI is a wide term that includes descriptive analytics, procedure analysis, data mining, and performance benchmarking. Customer churn is a general problem across businesses from several sectors. Companies are working always for improving their supposed quality by way of providing timely and quality service to its customer. Customer churn is developed most initial challenges which several firms were facing currently. Many churn prediction techniques and methods were presented before in literature for predicting customer churn from the domains like telecom, finance, banking, and so on. Researchers are also working on customer churn prediction (CCP) from e-commerce utilizing data mining and machine learning (ML) approaches. This manuscript focuses on the development of Stacked Deep Learning with Wind Driven Optimization based Business Intelligence for Customer Churn Prediction model. The proposed model is considered an intelligent system that applies golden sine algorithm (GSA) based feature selection approach to derive a set of features. In addition, the stacked gated recurrent unit (SGRU) model is applied for the prediction of customer churns.

Melanoma is the kind of skin cancer that poses the greatest risk to one's life and has the maximum mortality rate within the group of skin cancer disorders. Even so, the automated placement and classification of skin lesions at initial phases remains a complicated task due to the lack of contrast melanoma molarity and skin fraction and a greater level of color similarity among melanoma-affected and -nonaffected areas. Contemporary technological improvements and research methods enabled it to recognize and distinguish this type of skin cancer more successfully. A clustering technique called neutrosophic c-means clustering (NCMC) is presented in this research to group ambiguous data in the detection of skin cancer. This algorithm takes its cues from both fuzzy c-means and the neutrosophic set structure. To arrive at such a structure, an appropriate objective function must first be created and then minimized. The clustering issue must then be stated as a restricted minimization problem, the solution of which is determined by the objective function. This paper made a comparison between NCMC and fuzzy c-means clustering (FCMC). The results show that the NCMC is more suitable than the FCMC.

Diabetic foot (DF) is one of the most common chronic complications of poorly controlled diabetes mellitus (DM). Early diagnosis of DF and effective treatment is usually difficult by traditional approaches. Lately, it has been found a strong relationship between temperature variation and diabetic foot ulcer emergence. Thus, the current study focused on monitoring the temperature of feet using thermal images and its analysis techniques. The proposed system was based on employing a deep convolutional neural network (CNN) on thermal foot images. Experimental results showed that the proposed CNN has a maximum accuracy of 99.3% with minimum losses. When comparing the proposed system to other relevant systems, the proposed system approved greater accuracy, lower elapsed and testing time, which offers an automatic diagnostic tool for the diabetic foot and differentiates between its types. Thus, a simple, cost-effective, and accurate computer aided design (CAD) system could be presented to get a valuable system for the clinicians in hospitals.

Because there is now so many Internet of Things–based service providers globally, it will be hard to choose an Internet of Things service that is appropriate for the demand from the huge pool of Internet of Things services that are already available and display comparable characteristics. When making an acceptable choice, one can take into account the quality-of-service, or QoS, factors that characterize a certain service. In this article, we consider the Internet of Things to be the combination of its three3 potential parts, which are things, a connectivity unit, and a computational object. A definition of an IoT may contain the quality of service metrics for every one of these elements. We suggest a methodology that creates utilizes multi-criteria decision-making (MCDM) as a known approach using the MABAC method for the goal of carrying out the choice process where the quality of service parameters of different components of the internet of things act as criteria. Together, the data and our demonstration of the efficiency of the suggested strategy form a coherent whole.