Agriculture is a fundamental and basic source of living of many people and it is considered as a main part in economy and finance of any country. It is highly dependent on climate, weather and environment factors. Efficiency in crop productivity depends on agricultural elements embracing soil, water, temperature, and climate changes. The challenges in agriculture field, specifically crop prediction are complex and a machine-learning approach for crop cultivation is proposed for the achievement of production goals. This paper is proposed to design a machine learning-based system for crop selection. It can analyze the quality of soil and water, agro-climatic conditions, and also the necessary crop requirements for selecting suitable crops using the database. Also, supervising crop growth is proposed for monitoring the proper growth of the crop. A machine learning technique to develop crop prediction and soil quality check models. This model works on Machine Learning Techniques like Support Vector Machine (SVM) algorithm. The classifiers like Random Forest (RF), Gaussian Naive Baye's (NB), and k Nearest Neighbor (kNN) are used in crop prediction process. Based on machine learning algorithm and prediction accuracy, this model can help farmers at different locations for crop cultivation management.

Industrial 4.0 (4th industrial revolution) embodies rising technological advances withinside the improvement of clever manufacturing strategies. Industry 4.0 has great potential for many manufacturing companies to allow customization of products, provide flexibility to meet new needs in real-time and produce highly efficient jobs. The 4th Industrial Revolution and rising technologies—which include the Internet of Things, artificial intelligence, robots, and greater productions—affect the emergence of the latest manufacturing techniques and commercial enterprise fashions that transform fundamental manufacturing. The next generation of our industry is Industry 4.0—with the guarantee of improved production flexibility, as well as greater customization, improved productivity, and better quality. This allows companies to access additional products designed for each in a shorter period of time and better market standards. Intelligent manufacturing performs a critical position in Industry 4.0. Ordinary tools are transformed into intelligent objects that you will hear, handle, and perform in an intelligent environment. This makes industrial autonomy a reality where different countries work together to develop technology and start the next generation which is a 5.0 industrial revolution. Here in our work, we’ve mentioned the effect of Industry 4.0 on making the arena digital. Industry 4.0 is ubiquitous; however, we goal to take a more in-depth study of the cutting-edge enterprise imaginative and prescient and display the enterprise 4.0 destiny trends. Additionally, this chapter introduces generation to transport from Industry 4.0 to Society 5.0 and anticipates the destiny from Industry 4.0 to Industry 5.0.

Wireless sensor networks (WSNs) are designed to sense, collect, and transmit information from the environment to the base station. Under applications such as the Internet of Things and mobile WSNs, intelligent routing is important for achieving better quality of service (QoS) performance of the network. The real-time scenario is dynamic and contains heterogeneous nodes (especially in terms of energy) for a prolonged lifetime of the WSN. Despite these recent advancements in WSNs, energy efficiency is still an essential component that needs researchers to focus their efforts to increase the lifetime of the network. Many researchers have presented their contributions with the integration of optimization algorithms with cluster-based routing protocols to enhance such energy constraints in real-time applications. The objective of this chapter is to review existing energy-efficient routing protocols for homogeneous as well as heterogeneous WSNs. However, these existing algorithms have shown limitations in terms of dynamic, mobile, and heterogeneous network scenarios. Further, this chapter proposes the application of machine learning, swarm optimization, and an evolutionary approach to compensate for these limitations and to make routing decisions more cost-effective in terms of energy.

IONT is a new way of idea about how wireless communication networks should work (AI). Spectrum scarcity, advanced security, and not having enough intelligent, autonomous security are big problems for traditional cellular networks. Next-generation communication networks make it possible to build IONT networks based on sensors that can handle a lot of data. We create a safe 5G Internet of Nano Things device communication to improve communication routes and make sense of huge amounts of data. To proposed an AI- and Blockchain-based architecture for securing 5G-enabled nano IoT things. In 5G, the current AI algorithm is not as good as the physical layer (PHY), but it gives the air interface flexibility, agility, efficiency, and security. AI-based quantum Boltzmann machines and a blockchain-secured 5G are interface are part of our proposed paradigm. Lastly, To perform the simulation using Python to compare our proposed paradigm with 5G-IONT and 5G smart IONT.

With the rapid development of communication tools and techniques as well as intelligent processing and analysis had led to the development of smart Internet of Things (IoT) applications. The IoT application generates a huge amount of data every day as its application is vast. One of the most important applications of IoT is in the healthcare sector that generates large and sensitive data. So, to process such a large amount of data over an insecure network is an issue of concern. In this chapter, an overview of IoT architecture, issues, and solutions are discussed. The existing privacy-preserving inference approaches have problems such as high computation and communication overheads. In this chapter an overview of security challenges and security models using machine learning for the Internet of Healthcare Things (IoHT) is given along with a framework is proposed that can enforce security and trustworthiness on the Internet of Healthcare Things (IoHT) by amalgamating privacy and deep learning. This chapter will give a direction for future research work while designing a secure IoHT framework with low latency, and fast processing for accurate end-to-end data delivery.

With the application of wireless body area networks, patients can be remotely monitored by doctors. WBANs collect the medical data and transmit it over the internet for further processing. There is need to ensure security of such highly sensitive data over the network. This has deliberately attracted researcher interest to provide WBAN security by integrating blockchain. This chapter discusses internet of things (IoT) architecture of WBANs and proposes a lightweight secure access control using blockchain to achieve higher performance.

With the rapid development of big data technologies, their applications are growing rapidly. Conventional big data management is vulnerable. If any tampering occurs over stored database or sometimes untrusted data gathering occurs, then it causes serious issues with conventional data management. To resolve these problems, this chapter is focused on the development of blockchain technology for secure educational big data (EBD). Blockchain techniques divide the data blocks into secure data blocks using cryptography algorithms. The secure data blocks are connected in the chain with each other. This ensures security as well as flexibility. So, in this chapter, the proposed methodology had adopted blockchain technology and ensures its efficiency due to its decentralized and flexible approach.

Network protection becomes the key concern of all organizations in defending them from cyber threats of all kinds. The analysis of traffic in the network is an important activity for this reason. Network traffic analysis is achieved via the implementation of various intrusion detection systems. A honeypot is one of the most critical instruments to detect network intrusions. Honeypots communicate with the attacker and gather data that can be analyzed to collect information about the attacks and network attackers. Honeypots are security tools by which the attacker targets and generates attack data logs. Honeypots provide small quantities of relevant data so that security vigilance can easily understand, and future research can be carried out. To resolve the issues of new attacks in the network, this paper is focused to focus on the integration of honeypot to identify and treat suspicious network traffic flow. This paper gives a brief overview of the implementation of honeypot in the cloud. In this work, deep learning trained IDS is proposed on the honeypot server for network traffic analysis.

Edge computing is a type of distributed computing that was designed especially for internet of things (IoT) users to provide computational resources and data management nearby to users' devices. By introducing edge computing for IoT, networks have reduced the bandwidth and latency issue while handling real-time applications. The major benefit of edge computing is that it reduces the communication overhead between IoT user and server. With integration of IoT in our daily lives, it has attracted researchers towards its performance management such as complexity minimization, latency minimization, memory management, energy consumption minimization, etc. In this chapter, deep reinforcement learning is focused to minimize the computational complexity at IoT user end. The task offloading decision process is designed using Q-Learning, which minimizes the system cost and curse of high dimensional data. In addition, the proposed methodology will perform better as compared to existing algorithms with respect to system costs.

Transport sector has its importance in almost every area. Among the various kind of transport sector, the road transport sector is one of them. As a continuous function of automobile leads to some major or minor damage, efficiency decrement, defects in various elements of the automobile, etc. To fix all these problems, timely maintenance is required which either prevents severe damage or cures the problem. If the problem is not anticipated or fixed when damage is less than there are huge chances of major road accidents or breakdown of automobile which increases the maintenance cost and also risks human life. Therefore, automobile monitoring is vital to prevent road accidents, maintain efficiency, increase the life of automobiles, and decrease maintenance costs. This is achieved by integrating the advance and smart monitoring system with various automobile parts that need to be monitored. It is feasible to use analyze data of sensors attached to the automobile parts, by machine learning techniques to forecast failure. Data is collected by sensors when the automobile is moving as well as at rest. The data is then transmitted to machine learning which divides the data used to train the algorithms developed and for testing. The patterns are learned through classifiers and these patterns use in failure detection of other vehicles also which shows the same behavior as the vehicle from which data is collected. The chapter presents the role of artificial intelligence (AI), machine learning (ML), and the internet of things (IoT) in automobile monitoring specifically for fuel-efficiency monitoring, air inflation monitoring, and suspension adjustment. Further, various kinds of existing technologies recently developed regarding suspension adjustment monitoring, fuel efficiency monitoring, and air inflation monitoring have been discussed.

Knowledge management (KM) is a technique that is used to develop and manage organizational performance in such a competitive scenario due to current fundamental economic resources. This chapter is aimed at analyzing KM linked to decision support systems for human resource management to argue the relevance of knowledge in organizations. The implementation of KM has increased rapidly with the development of Web 3.0. So, the management information systems (MIS) of an organization must be upgraded and information stored in it can be made available easily to those who need it. The proposed framework will deploy KM to achieve high performance and to achieve a sustainable advantage. This framework will focus on the organization’s functional structure, operational and marketing aspects, and technical aspects.

Autism spectrum disorder (ASD) is one of the most common diseases that cause difficulties for an individual to express his/her emotions or to understand other's emotions. ASD has become a challenging problem as its symptoms are unpredictable. The main symptoms of ASD include problems such as abnormal social reciprocity, nonverbal communication, sensory abnormalities, etc. To understand such abnormalities, there is a requirement of some learning tools. It has been witnessed that facial expression images, eye tracking, and neuroimage have been shown as effective tools for analysis of abnormalities that had occurred in both grey and white matter of the brain. Many researchers focused their work on the classification problem of ASD disorder from healthy subjects but still didn't reach effective diagnosis and healing tools. As with the advancement of digital image processing, it has become feasible to use these technologies for accurate diagnosis of ASD subjects. These technologies are integrated with deep learning for the identification and treatment of ASD.

The basic concepts of training and the model structure of deep belief networks (DBNs) in deep analysis are studied to apply image recognition in the area of deep learning. Random propound is provided with the parameter in the fine-tuning stage and the randomly hidden layer eliminated to maintain unchanged weights. The results show that the layered DBN training system reduces training problems and training times significantly. In the small sample, the deep faith network has improved significantly after introducing the down sample and random dropdown and effectively alleviates the over-fitting phenomenon. Design a new Deep Learning Image Recognition and Classification Algorithm. Novel Algorithm for Image Classification Using Cross Deep Learning Technique.

The World Wide Web has taken seriously new ways for individuals to convey their views and conclusions on different topics, models and issues. Clients create content that resides in a variety of media, such as web gathering, conversation gathering, and weblogs, and provide a solid and generous foundation for gaining momentum in different areas such as advertising and research. Policy, logic research, market forecasts and business outlook. Hypothesis research extracts inferences from information available online and orders the emotions that the author conveys for a particular item into up to three predefined categories (good, negative, and unbiased). Identify the problem. This article outlines a hypothesis review cycle for quickly ordering unstructured news on Twitter. In addition, we are exploring different ways to perform a detailed emotional survey on Twitter News. In addition, it presents a parametric correlation of strategies considered according to recognized boundaries. This work tends to make the case enjoy investigating on Twitter; The values communicated in them represent the tweets: positive, negative or fair. Twitter is an online thumbnail that contributes to a blog and a wide range of interactions, allowing customers to create short 140-character short instructions. It is a fast growing association with more than 200 million subscribers, of which 100 million are dynamic customers and half of them constantly sign up for Twitter, generating around 250 million tweets every day. Due to this overwhelming use, we plan to achieve a biased impression of the public by breaking the estimates communicated in the tweets. Researching public opinion is important for some applications, for example, when companies are looking to respond to their material, predict political careers, and anticipate economic wonders like stock trading. The function of this to build a useful classifier for the command in a precise and programmed way of the stream of fuzzy tweets.

Cyber security is the most important part of our technologically dependent lives today. Government institutions, financial firms, public and private services, nuclear power stations, power grid providers, water supplies or wastewater disposal corporations use information technology in their day-to-day activities. Anything that uses technology is focused on connectivity and computer networks, which ensures that it relies on cyber security. Every year, the public and private sectors spend millions of dollars on technology, protection tools and hardware that can improve cyber security within their businesses, but are still vulnerable. The key problem with this situation is that data defense is now generally viewed as a technological component or technology that can be quickly introduced within the enterprise, and this application would ensure cyber protection. This mentality needs to change, because cyber security is more than mere technology nowadays. This post presents the taxonomy of critical infrastructure threats, analyses attack vectors and attack tactics used to destroy critical infrastructure, as well as the most frequent cyber security failures that companies create in the area of cyber security while attempting to make themselves safer from vulnerabilities. The main purpose of this article is to include technical elements of the cyber security management model that can be used to ensure the protection of sensitive infrastructure in an enterprise or business. The network protection management model discussed in this article is examined from a management viewpoint and does not discuss the technical issues and products used to secure sensitive infrastructure from cyber security threats and vulnerabilities.

In recent advancement of computational techniques, there is an exponential increase in amount of data. Learning on such large amount of data is a major area of concern with application of machine learning algorithms. Therefore, it is considered to be a complicated task to handle and perform computation on such large, complex, and heterogenous dataset. In this paper, a brief discussion about different dimension reduction or feature selection algorithms is given. A brief review about contribution of researchers for designing feature selection algorithms for large dataset is given. By analyzing exisitng problems, this paper is motivated to design a hybrid, robust, flexible, and dynamic feature selection model for classification of large datasets. For this, multi-objective optimized feature selection is proposed with an objective to minimize the error rate and execution time as well maximize accuracy of problem and to generate solution with high probability.

Security is a key problem to each computer and computer networks. Intrusion detection System (IDS) is one of the most important research problems in community safety. IDSs are advanced to stumble on each acknowledged and unknown assaults. IDS employs many methods to secure information systems and networks against community-based and host-based threats. IDS utilises different machine learning methods. This thesis analyses IDS machine-learning methods. It also discusses several similar research completed between 2000 and 2012 and specialises in engineering techniques. Linked experiments include used unmarried, hybrid, ensemble, baseline, and datasets.

Recent developments in information management and corporate computerization of company processes have made data processing faster, easier, and more accurate. Data mining and machine learning techniques have been used increasingly in different areas, from medical to technological, training, and energy applications to analyze data. Machine learning techniques allow significant additional knowledge to be deducted from data processed by data mining. This critical and practical knowledge allows companies to develop their plans on a sound basis and reap substantial time and expense benefits. This article implements the classification methods used for data mining and computer training for the collected data during technical advice processes and aims to find the most powerful algorithm.

In the recent scenario, there is a drastic improvement in transportation, infrastructure, and communication technology which increases the number of commercial as well as non-commercial vehicles. Therefore, there is also an increase in the number of accident incidence. This ultimately results in a high death rate due to a road accident. More than half of accident incidence results in death due to delayed medical aid to the victim. If medical aid or services received at the proper time, then the victim may survive. With the application of machine learning processes and communication advancements, there is scope for the development of a more accurate system. In this chapter, a model is presented based on IoT devices that can sense and predict the pre-accident/pre-collision state and generates an alarm message about the collision is going to occur. This model is designed to extracts image/video features to determine the possibility of occurrence of a collision. This model is also efficient for post-collision.