in the construction industry sometimes leads to various types of errors, delays, and even accidents at some stages. This chapter introduces the key to ending these disputes with the help of Corda, a distributed ledger platform for permissioned networks inspired by blockchain technology. This helps in maintaining transparency among the actors involved in this industry, thus avoiding any miscommunication.
– Chapter 3 analyzes the identity and access management challenges in the IoT, followed by a proposal of a cloud identity management model for the IoT using distributed ledger technology.
– Chapter 4 elucidates the development of an efficient deep neural network (DNN) with a reduced number of parameters to make it real-time implementable. The architecture was implemented on German traffic sign recognition benchmark (GTSRB) dataset. Four variations of neural network architectures—feedforward neural network (FFNN), radial basis function neural network (RBNN), convolutional neural network (CNN), and recurrent neural network (RNN)—are designed. The various hyperparameters of the architectures—batch size, number of epochs, momentum, initial learning rate—are tuned to achieve the best results.
– Chapter 5 deals with the basic aspects of honeypots, their importance in modern networks, types of honeypots, their level of interaction, and their advantages and disadvantages. Furthermore, this chapter also discusses how honeypots enhance the security architecture of a network.
– Chapter 6 provides an in depth review of the necessity for security in IoT platforms and applications of the industrial internet of things (IIoT). Over the past decade, cyberattacks have mostly occurred on IoT devices; therefore, cybersecurity is introduced to deal with these cyberattacks. Furthermore, one of the chief attack modes in the IIoT are botnets and denial-of-service attacks. These attacks happen in several ways, and once they have occurred it is hard to predict and stop them. This chapter highlights many suggestions from diverse authors, which are detailed in tabular form.
– Chapter 7 proposes an efficient navigation controller embedding hybrid Jaya-DE algorithm to obtain the optimum path of an individual robot. The efficiency of the proposed navigation controller was evaluated through simulation. The outcomes of the simulation revealed the efficacy of the suggested controller in monitoring the robots towards achieving a safe and optimal path. The strength of the suggested controller was further verified with a similar problem framework. The potency of the proposed controller can be seen from the outcome in resolving the navigation of mobile robots as compared to its competitor.
– Chapter 8 discusses a study conducted for diagnosing Parkinson’s disease using different machine learning (ML) algorithms for categorization and severity prediction through the measure of 16 voice and eight kinematic features accomplished with various archives. The dataset included 40 people with Parkinson’s disease and healthy patients generated with the help of spiral drawings and voice readings. Of the various ML algorithms used for estimating, the highest accuracy (94.87%) was demonstrated by ANN, while Naïve Bayes was the least precise (71.79%). The work also predicted a severity score by suggesting some scientific measures with a prototype dataset.
– Chapter 9 discusses the challenges faced in the development of a multi-sensor classification system and their possible solutions. Smart agriculture in rural areas can largely benefit from the low-power, low-cost sensors and aerial devices to sense (soil, temperature, salinity, water, light, insects, pests) and exchange data/images for monitoring and controlling crops.
– Chapter 10 builds a classification model that classifies whether a customer is going to buy a car with specific features. This research work consisted of four ML models and an analysis of their results. These classifying models were Gaussian Naïve Bayes, decision tree, Karnough Nearest Neighbors and neural networks. The author also attempted to find the best hyperparameter value to obtain the best result from these models. These results are used to compare the accuracies of every model and decide the best model for use in real-time prediction. Here, the author was predicting whether a customer was going to buy a car or not buy a car with particular features available in it. Hence, for this prediction the best accuracy we got was 97.4%, which was given by the decision tree classifier. Also, the neural network had about the same prediction accuracy. Therefore, this ML model can be used by a firm to determine whether or not a new car with specific features will sell well or by a customer wanting to know whether a particular car will be bought by other customers as well.
– Chapter 11 examines the use of AI and ML in political campaigns. It is divided into three sections—the first section explores internet penetration and the influence of social media on the Indian Lok Sabha election; the second section explores the forms of deepfake and automated social media bots and their use during the election campaign; and the final section explores the future of AI and ML in the election campaign in India.
– Chapter 12 attempts to explain the impact of segment routing (SR) in software-defined networking (SDN). For this, the authors implemented three algorithms known as multi-objective particle swarm optimization (MOPSO), advanced MOPSO (A-MOSPO) and minimum interference routing algorithm (MIRA) on a Waxman network topology created randomly having 100 nodes. For performance evaluation, MATLAB and parameters such as throughput, link utilization, and delay were taken as the key parameters for evaluating the above protocols in an SDN environment.
– Chapter 13 discusses the symptoms of COVID-19, precautionary measures against it, ways of spreading the corona-virus, and technologies used to fight it. Also discussed is the impact of COVID-19 on business, financial markets, supply-side and demand-side economics, and international trade on the Indian economy.
– Chapter 14 discusses the convolutional neural network (CNN) used for detecting skin cancer and compares the accuracy of the model by applying a vast dataset by varying the parameters, such as number of layers, activation functions, etc., to find the best suitable parameters for CNN to design the classifier that could give the best accuracy while classifying the images of the seven types of skin cancer.
– Chapter 15 presents the hybrid outcome of the firefly algorithm (FA) and artificial potential field (APF) algorithm for humanoid control, which is preferred in the present study for navigational tasks.
– Chapter 16 proposes a system that considers the student’s academic and behavioral characteristics. The data collected can help faculty members gain a better understanding of a student’s level of knowledge and personality. Based on the information collected, students are grouped into clusters using k-means clustering and a suitable partner is selected for group activities using Irving’s algorithm to enable active learning.
– Chapter 17 discusses how the workload prediction in cloud environments improves proper utilization of resources so that service level agreement remains at a stable level. Hence, the particle swarm optimization (PSO)-based hybrid wavelet weighted k-nearest neighbors (PHWkNN) algorithm is proposed to
