Artificial Intelligence and Natural Algorithms

There has been a tremendous revolution in computing technologies to handle the vast amount of data in recent years. Big data is the large-scale complex data in which real-time data is available and mushrooms the development of almost every field. In recent years, the demand and requirement of big data produced an opportunity to replace traditional data techniques due to their low efficiency and low accuracy. It shows adequate responsiveness, absence of versatility, execution, and precision for meeting the convolution of Big Data challenges. As an outcome, this created different dispersions and innovations. Big data does not mean that the data is humongous but additionally excessive in range and speed. This factor makes them tough to deal with the usage of conventional gear and techniques. Decision-makers read the extension and expansion of big data to understand and extract valuable information from rapidly varying data using big data analytics. In this chapter, we can analyze big data tools and techniques useful for big data. This chapter presents a literature survey covering various applications and technologies that play an indispensable role in offering new solutions dealing with large-scale, high-dimensional data. By summarizing different available technologies in one place from 2011 to 2019, it covers highly ranked international publications. Further, it extends in the context of computing challenges faced by significant Data Healthcare, Clinical Research, E-Commerce, Cloud Computing, Fog computing, Parallel Computing, Pervasive Computing, Reconfigurable Computing, Green Computing, Embedded Computing, Blockchain, Digital Image Processing and IoT and Computing Technology. The survey summarizes the large-scale data computing solutions that help in directing future research in a proper direction. This chapter shows that the popularity of data computing technology has steeply risen in the year 2015, and before 2011, the core research was more popular.

An IoT (Internet of Things) technology is a dynamic area of research, which has been growing at a remarkable rate for the last few years. The IoT is a mammoth network of associated things and people, which accumulates and shares data about how they are used and the environment around them. It is the discernment of connecting any device to the internet and other associated devices. When something is associated with the internet, it can propel or receive information, or both. This ability to propel and/or receive information makes things smart. IoT permits businesses and people to get more insights from the world around them and do more evocative higher-level work. Data fusion techniques are used to extract eloquent information from dissimilar IoT data. It ferocities dissimilar data from sensor sources to mutually find a consequence, which is more dependable, precise, and comprehensive. This chapter briefly designates the IoT by the characteristics of data procurement and data fusion. 

With the passing decades, Artificial Intelligence (AI) is gaining high popularity in various domains. In this chapter, we aim to present the current scenario of the application of AI in the field of medical science. Firstly, we will introduce the early and basic role of AI in the medical field. We preceded the chapter with a summary of the most current applications of AI in various areas of medicine and health care. In this review, we have discussed the latest developments of applications of AI in biomedicine while predicting the risk of disease. Estimating the success ratio of the therapy also manages or reduces the severity of complications, taking care of ongoing patients, living assistance, biomedical information processing, biomedical research, and medical imagining. We also present a survey on AI techniques, which were used by many authors with different objectives in medical science. Furthermore, we showcase the effects of the usage of AI by highlighting the reduction in the rate of mortality, and fast and accurate diagnostics which help in decreasing errors related to human fatigue and lessening medical costs. Finally, we draw attention to some of the possible weaknesses, apprehensions, and uncertainties in using AI in medical science. We briefly review the efforts being made to improve the healthcare industry by offering various AI-based healthcare products. 

We live in a world brimming with technology and crime. Even being in large numbers, law enforcers lack the required presence and resources. Although there are a lot of surveillance devices being installed in a general view of the public, most of them require an operator to monitor them. Even if they are smart devices, they lack the ability to cover all aspects of a situation. The proposed solution emphasizes computer vision to develop software that leverages the widely available network of surveilling cameras to detect criminal threats using object detection, violent activities such as CNN, detection of a person in need of medical aid, and sending the same to ground zero. Thus, effectively covering all 3 major aspects of a threat, namely, the crucial time before a crime is conducted, detecting an ongoing act of violence, and finally sending help as soon as possible to the victims in the aftermath. This would serve as an additional eye for law enforcement and will certainly aid in reducing the response time from authorities and mitigate most of the rising threats.

In cloud computing environments, parallel processing is required for largescale computing tasks. Two different tasks are taken, and these tasks are independent of each other. These tasks are independently applied to Virtual Machines (VM). We proposed Offbeat Load Balancing (LB) Machine Learning algorithm using a task scheduling algorithm in Cloud Computing (CC) environments to reduce execution time. In this paper, the proposed algorithm is based on the concept of Random Forest Classifier and Genetic Algorithm and K-Means clustering algorithm for optimized load. The proposed algorithm shows that the average execution time of 3.5104 seconds (20 jobs, 5 Machines) and 15.85 seconds (20 jobs, 10 machines) is based on a study of load balancing algorithms that needs less execution time than other algorithms.

 Stream data classification involves a predicament of new class generation through pattern evaluation. The evaluation process of the pattern raised new ways of data classification. The evolving decoration discrepancies dispensed the session for rivulet data arrangement. Now, the twisted pattern fashions innovative classes for cataloging progression. For this method of regulation, multi-class sapper method is used. A catastrophic spread of new decorating appraisal methods for multiclass mine workers is used nowadays. We cast off the pattern optimization performance using a transmissible algorithm aiming at the group of patterns and their heightened process for instructing multiclass. The enhanced pattern stables the new class while enhancing the successful multiclass miners. For the empirical appraisal, we used health care data such as cancer and some other deride for the evolutionary progression of the pattern optimization process.

In the last few decades, artificial intelligence (AI) has shown rapid growth in medicine with the evolution of computer vision, robotics, natural language processing, and deep neural networks. The technology has also been applied to healthcare with an inexact thought that AI will replace the workforce. AI works as a helping hand for human clinicians because a machine can never replace a human brain. Present healthcare systems can implement AI technology for diagnosing patients and their treatments, drug invention, prediction of disease outbreaks, real-time monitoring of critical patients, radiology, and many more. The latest achievements by Google for the diagnosis of cancer, and diabetic retinopathy by JAMA using deep learning algorithms and surgical robots show substantial shifts in medicine. A simple assessment of electronic health records (EHR) provides more opportunities for the medical experts during the invention and application of improving medicines. The coming future of health care depends on the advancement of AI. However, with ease comes difficulty, such as the privacy of data and causality problems which should be considered when deploying such strategies. 

 India is the land of agriculture with many varieties of plant species. These species have different uses in the medical, food, and harvesting industries. Despite having such a large collection of plants and agricultural assets, most of the Indian population is not aware of the goodness and properties of these precious plants except the usual ones. In this chapter, discussion and possibilities in this area are given and explored for the awareness of Indian people regarding the Indian plants. In this area, artificial intelligence and machine learning will likely develop an automated detecting machine that can classify and describe the plants by their images of leaves, bark, flowers, and stems. Looking forward in this direction, this chapter discusses an AI and ML based technique to recognize vegetation by the image of its leaves. In this approach, SIFT and ORB-based technique removes leaf image features and then tests the data set to match with a trained data set. The system is trained with 32 plant leaves. Henceforth, this system can recognize these plants by the image of their leaves. The uniqueness of this system is its data set. In the data set, the image of the leaf is prepared so that both sides of a leaf can be used to recognize the plant. This increase distinguishes the image irrespective of its color and shape. The system is still in an evolving phase that has the target of including all rare and useful plant information in this dataset. This system is very useful to preserve the information of all users, rare plants and those plants that are about to be extinct.

Rice is one of the staple crops in the world, as it is a rich source of protein, minerals, fibre, and vitamins. It is cultivated almost in every part of the world, but its productivity decreases due to several diseases. If these diseases are identified at the initial stages, then preventive measures can be taken, but their symptoms are quite similar for human eyes to recognize them correctly. Therefore, there is an immense need to apply automated techniques for recognizing rice diseases. Various Artificial Intelligence (AI) based prototypes have been surveyed in this chapter. These techniques were proposed by researchers for diagnosing rice disease. Here, our main goal is to present ideas on how Pretrained Neural Networks can be used in the recognition of rice diseases. Therefore, a brief description of AI techniques and their comparison is also outlined.

 Shallow clouds play a significant role in the earth’s radiation balance, but they’re still poorly represented in climatic models. Our project analyzes the cloud images taken from satellites and attempts to build a deep learning model to classify cloud patterns. This will help us to identify the cloud formations and help improve the earth’s climate understanding. We will use various deep learning and image segmentation techniques like UNet to produce a model which can classify the shallow layers of clouds into various labels (fish, flower, gravel, and sugar). Various data augmentation techniques are implemented to improve the proposed model. Additionally, transfer learning is implemented by using ResNet backbones to improve the performance of the segmentation model. This will help gain insights into the matter of shallow cloud effects on the earth’s climate, there by helping in the development of next-gen climate models without having to go through the tedious task of classifying the clouds present in the images first.

 Pattern classification is also called pattern reorganization. The classification of pattern is one of the critical problems in Artificial Intelligence. COVID-19 is the most viral lung disease that has put the whole world in such a difficult situation that it has become necessary to develop a machine-learning algorithm to classify lung disease. This research paper is ready to propose Artificial Intelligence Intelligence-based Lung Disease Classification by using an Evolutionary Deep Learning Paradigm to solve the said problem. It delineates an integrated bioinformatics approach in which different aspects of information from a continuum of structured and unstructured data sources are put together to form user-friendly platforms for physicians and researchers. The main objective of the proposed system is to find the probability, diagnosis, and treatment of the COVID-19 disease. Artificial Neural Network-based tool for challenges is associated with COVID-19. There is some specification of our platform as it includes various forms of input data containing medical as well as clinical data. This helps in improving the performance of the system. Experimental results are calculated and statistically analyzed. For benchmarking, the performance of the proposed approach is compared and statistically analyzed. 

The polysomnography test (sleep study) is used to diagnose several sleeping disorders. Sleep study is used to detect sleep disorders such as Insomnia, REM Sleep Behavior, Insomnia, Restless Leg Movement Syndrome, and Sleep Apnea. It measures different parameters such as heart rate, level of oxygen in your blood, body position, brain waves (EEG), breathing rate, eye movement, and electrical activities of muscles. In the world, 700 million people suffer from sleeping disorders. A wide range of sensors was attached to the body of the patient to measure the value of different parameters. However, in 2020, due to the exponential spread of COVID-19 coronavirus disease, the sleep study centers were closed, and it was very difficult to perform sleep studies on patients. Therefore, we developed a hybrid model based on deep learning techniques like Convolutional Neural Network (CNN) and Deep Belief Network (DBN) architectures. Numerous cameras were mounted in rooms at certain angles, which provide live surveillance data and record a patient’s movements after a short periodic interval of time. This research paper concludes that non-contact-based hybrid models are highly accurate in detecting sleep disorders based on polysomnography tests.

 Today, the coronavirus has widely affected the entire world. In late 2019, a virus with the pandemic potential was reported in the city of Wuhan, situated on the mainland of China. In no time, the virus had spread all over the world, multiplying from person to person. Undoubtedly, COVID-19 has become an important research topic, and many research works are coming forward daily. Thus, COVID-19 patient’s detection has become the most personified research for the researchers. CT scanning has been an important and widely used approach for detecting COVID-19 patients. In this work, the identification of COVID-19 patients is performed using two different deep neural network methods. For the image accession, the dataset having 746 samples was used. The entire dataset has been bifurcated into two different classes, i.e., COVID-19 and non-COVID-19. COVID-19 class contains samples of the COVID-19 positive cases, whereas the non-COVID-19 class contains the sample of COVID-19 negative cases. In total, 506 images are used for training purposes, whereas 240 images are used for validation. The identification is performed using MobileNet-V2 and Modified LeNet5 convolutional neural network (CNN) models having a fixed number of convolutional and fully connected layers. The term modified is added before the LeNet architecture because an extra convolutional layer was created for the experiment. As per the details and requirements, the architecture for Modified LeNet was designed, whereas, for the MobileNet-V2, it is imported from predefined libraries and is used further as per the author’s need. After the successful completion of the experiment, it has been found that the accuracy of MobileNet-V2 and Modified LeNet5 came out to 85.86%, and 84.38%, respectively.

This chapter builds the foundation of software testing techniques by classifying the various testing approaches and testing coverage criteria. It gradually advances in the concepts and process of Mutation Testing and its application areas. Mutation testing has been applied at both the source code level and specification level of the software under test. Mutation testing, when applied to the source code, is named as Program Mutation. Similarly, when applied to the specifications, it is named as Specification Mutation. The relevant Mutation Testing tools available for different programming languages for both program and Specification Mutations are hereby listed. Owing to the high cost incurred in applying Mutation Testing to industrial needs, the on-going endeavors of the researchers in the area are elaborated here. Applying nature-inspired algorithms along with Mutation Testing for data generation/selection/minimization is an upcoming area of research. Search based Mutation Testing (SBMT) applies evolutionary techniques like Genetic Algorithms or other metaheuristic approaches for automating the tasks associated with mutation testing, which otherwise requires a lot of human effort, thus, making it a practical approach. This chapter concludes by giving the seminal recent advancements in the area.

Extreme learning machine (ELM) is a rapid classifier evolved for batch learning mode unsuitable for sequential input. Retrieving data from the new inventory leads to a time-extended process. Therefore, online sequential extreme learning machine (OSELM) algorithms were proposed by Liang et al.. The OSELM is able to handle the sequential input by reading data 1 by 1 or chunk by chunk mode. The overall system generalization performance may devalue because of the amalgamation of the random initialization of OS-ELM and the presence of redundant and irrelevant features. To resolve the said problem, this paper proposes a correspondence multimodal genetic optimized feature selection paradigm for sequential input (MG-OSELM) for radial basis function by using clinical datasets. For performance comparison, the proposed paradigm is implemented and evaluated for ELM, multimodal genetic optimized for ELM classifier (MG-ELM), OS-ELM, MG-OSELM. Experimental results are calculated and analysed accordingly. The comparative results analysis illustrates that MG-ELM provides 10.94% improved accuracy with 43.25% features compared to ELM.

Due to the continuous evolution of the Big Data phenomenon, data processing in Business Big Data Analytics (BBDA) needs new advanced load balancing techniques. This chapter proposes a new algorithm based on a nonstigmergic approach to address these concerns. The algorithm imitates a specific species of ants that communicate by the acoustics in situations of threats. Besides, the research methodology in this study presents a methodic filtration of the relevant metrics before carrying out the benchmarking trials of several ant-colony algorithms (i.e., makespan, response time, throughput, memory and CPU utilization, etc.). The experimentations' outcomes show the effectiveness of the proposed approach that might empower the research efforts in big data analytics, business intelligence, and intelligent autonomous software agents. The main objective of this research is to contribute to reinforcing the resilience of the Big Data processing environment for enterprises.

The concept of computational algorithmic and sustainable elastic artery evaluation and its impact on different variables such as structural and morphological variations and its applications are explored in this chapter. The Crank-Nicolson approach has solved the mathematical goal of the equation. The result is explained in the case of blood supply in elastic vessels while the electromagnetic effect is established. In this computational analysis, the discrepancy between the arteries and veins attached to the elastic arteries in the blood vessel is easier than determining the presence or absence of the elastic layer within the vessel. The obtained results in the analysis are in relatively accurate compliance with the computational findings in this chapter. The findings may be applicable to cases of pulmonary edema, etc.

Nowadays, in the automation industries, the Cruise Control System (CCS) is one of the essential aspects, and it is necessary to have a well-designed controller that can suit a new improvement in innovation. The CCS is a very famous and important model in control system engineering. The fundamental objective of CCS is to regulate vehicle speed depending upon the chosen speed. The CCS is an example of a close loop control system. Speedometer is utilized in the feedback path for measurement of the speed. This is a simple model used to solve the many problems of drivers like road accidents, weariness, etc. In this paper, we analyze the performance of different controllers such as Proportional-Integral-Derivative (PID) controller, the fuzzy logic controller (FLC) and the fuzzy-PID (F-PID) controller in the different situations on the road, such as friction, road grad, or angle of inclination to attain the chosen speed of the vehicles. The tuning of PID parameters is done using the method of Ziegler-Nichols, and FLC uses the gaussian Membership Function (MF) in this paper. The MF is a graph that lies between zero and one. It indicates the mapping of every point in the input state and the values of MF. The mathematical model of this system is considers the road grad and the friction. Finally, in this paper, we see the response of models with and without a controller in different situations on the road.