Thillaiarasu Nadesan

Electricity theft in smart grids poses a significant threat to energy security, leading to billions in financial losses and grid instability worldwide. Traditional detection methods, including hardware-based solutions and machine learning (ML) models, are often costly, reliant on labeled data, and lack scalability. Deep learning (DL) approaches, while more advanced, face challenges such as overfitting to static datasets and inefficiency in adapting to evolving consumption patterns and new cyberattacks, requiring frequent and computationally expensive retraining. In this context, we propose a novel deep learning framework, Deep Adaptive Feature Learning for Theft Detection (DAFL-TD), tailored for smart grid environments. The architecture of DAFL-TD integrates a Temporal Feature Extraction Network (TFEN), which captures temporal dependencies in electricity usage, with an Adaptive Feature Learning Network (AFLN) that leverages both labeled and unlabeled data for adaptive feature extraction and classification. The novelty of DAFL-TD lies in its ability to handle fluctuating, imbalanced data and dynamically update its feature representation without the need for extensive retraining, making it highly scalable for real-world smart grid applications. Extensive evaluations on the State Grid Corporation of China (SGCC) dataset demonstrate that DAFL-TD achieves a 13.84% improvement in AUC compared to state-of-the-art models, alongside superior precision as measured by MAP metrics. These results underline the efficacy of DAFL-TD as a robust, scalable, and efficient solution for real-time electricity theft detection, significantly enhancing the resilience and security of smart grids.

Deep learning-based handwriting personality prediction uses neural networks (CNNs) to consider and predict personality characters from handwritten text to improve applications such as adaptive technology and automated data access. This method requires high computational resources and widely classified data, and its accuracy may suffer from signature variation and overfitting. This study introduces a novel optimizeddeep learning for Hand Written Personality Prediction. First the input data is collected from givendataset and an enhanced Gaussian filter is used in pre-processing to improve image quality by reducing noise and increasing contrast.After pre-processing, features are extracted by PCA-based normalized GIST, which standardizes the features to improve image representation.These extracted features are generated using an adaptive horse herd optimization algorithm to select the most important features. Based on selected features Improved Generative Adversarial Networks with artificial hummingbird optimization (IGAN_AHb) enables fast and efficient convergence of GANs. This reduce the mode collapse and ensures training with consistently fine-tuned parameters.The selected features are considered by characters such as openness, conscientiousness, extroversion, agreeableness and neuroticism, and finally classified using the IGAN_AHb classifier. In the result section, the proposed model is compared with various models with the matrix precision, recall, F1-score. The proposed model attained the value of 97.3% of precision and a recall rate of 96%, respectively. By comparing with other models, the proposed model attained highest values.

Smart cities aim to sustainably increase urbanization, lower energy consumption, improve inhabitants' economic and quality of life, and protect the environment. The idea of smart cities revolves around ICT since it makes it easier to formulate policies, make decisions, execution, and ultimately the supply of helpful offerings. In addition to utilizing digital technologies, the concept of the “smart city” is a response to the political, social, and economic issues that post-industrial civilizations are facing at the beginning of the new millennium. In this chapter, the authors discuss several IoT-based machine learning techniques that are applied, among other things, in the aforementioned domains. Furthermore, a review of the assessments and a report on the lessons learned are presented, with an emphasis on the fundamental role that machine learning techniques should play in internet of things networks.

Cyber-physical systems (CPSs) are the intelligent systems that offer an interaction among computational, software, and networking resources in a continuous and dynamic fashion. Future systems are likely to be created and developed using CPSs, which have been recognized as a significant area of research. The electric power grid, energy systems, body area networks, modern vehicles, smart homes, cooperative robotics, and smart transportation are the examples for CPS. The security aspects of CPSs can be enhanced with blockchain (BC) technology. For instance, with the combination of CPSs and blockchain, a peer-to-peer energy market is made possible where machines may automatically buy and sell energy based on parameters specified by the user. In this chapter, we summarize recent developments in the creation and applications of CPS, the state-of-the-art and pertinent concepts, numerous CPS applications that have employed blockchain, relevant solutions, and open challenging issues.

With the evolution of Internet standards and advancements in various Internet and mobile technologies, especially since web 4.0, more and more web and mobile applications emerge such as e-commerce, social networks, online gaming applications and Internet of Things based applications. Due to the deployment and concurrent access of these applications on the Internet and mobile devices, the amount of data and the kind of data generated increases exponentially and the new era of Big Data has come into existence. Presently available data structures and data analyzing algorithms are not capable to handle such Big Data. Hence, there is a need for scalable, flexible, parallel and intelligent data analyzing algorithms to handle and analyze the complex massive data. In this article, we have proposed a novel distributed supervised machine learning algorithm based on the MapReduce programming model and Distance Weighted k-Nearest Neighbor algorithm called MR-DWkNN to process and analyze the Big Data in the Hadoop cluster environment. The proposed distributed algorithm is based on supervised learning performs both regression tasks as well as classification tasks on large-volume of Big Data applications. Three performance metrics, such as Root Mean Squared Error (RMSE), Determination coefficient (R2) for regression task, and Accuracy for classification tasks are utilized for the performance measure of the proposed MR-DWkNN algorithm. The extensive experimental results shows that there is an average increase of 3% to 4.5% prediction and classification performances as compared to standard distributed k-NN algorithm and a considerable decrease of Root Mean Squared Error (RMSE) with good parallelism characteristics of scalability and speedup thus, proves its effectiveness in Big Data predictive and classification applications.

Universities are currently confronted with changing student needs, a competitive labour market, and a fast-paced environment. The advancement of communication technology has enabled us to address these issues. Collaboration advances are critical to the current learning process because they train students to work in groups on tasks. In this chapter, the authors present a thorough foundation for an e-collaboration platform that was established during the successful implementation of an e-collaboration solution at the management information systems. The solution makes use of cutting-edge web portal technology and a digital asset management system to create a uniform, centralised platform for system users to collaborate, communicate, and exchange information.