Emerging Technologies in Computer Engineering: Cognitive Computing and Intelligent IoT

Game-based learning approaches and 21st-century skills have been gaining a lot of attention from researchers. Given there are numerous researches and papers that support the effect of games on learning, a growing number of researchers are determined to implement educational games to develop 21st-century skills in students. This review shows how Game-Based learning techniques impact 21st Century skills. 22 recent papers have been analyzed and categorized according to learning outcome, age group, game design elements incorporated, and the type of game (Game genre) implemented. The range of game genres and game design elements as well as learning theories used in these studies are discussed. The impact of implementing machine learning strategies and techniques in educational games has also been discussed. This study contributes to the ongoing research on the use of gaming features for the development of innovative methodologies in teaching and learning. This study aims to shed light on the factors and characteristics to be considered when implementing a game-based learning system. This study provides valuable insight to future researchers and game designers with issues and problems related to educational game design and implementation.

A method has been developed that solves digit recognition problem through the abstract idea of lines and curves. The human brain can easily classify digits because it learns from the formation of the digit, it learns from its geometry. Human brain will view the digits as combination of lines and curves and not just mere arrangement of pixel values. For the neuron network to exhibit human level expertise, it must also learn from the formation of digits. Traditional techniques learn from pixel values of the image and hence lack the sense of the geometry of the number. With the help the of this developed method the program begins to have a gist of the geometry and the shape of the number. It develops the idea that a combination of some lines and curves forms the digit. The curves can be easily represented by tangent vectors. The program uses this set of vectors instead of pixel values for training. Results show that the program, which is learning through the abstract idea of curves classifies digits far better than the program learning from mere pixel values and is even able to recognize hugely deformed digits. It also becomes capable of recognizing digits invariant to the size of the number.

Now day’s people opinion or sentiment matters a lot in the field of research. There are many patients who stay away from their family, friends and live in the place where hospital, facilities are not available. Then if they have any health related problems and they have to take a medicine in emergency then which medicine will be better for them. Therefore to overcome the problem developed the proposed predictive classifier model in which experience patients who used a medicine in the past, their reviews or feedback will be mined. In the developed model the worthless model has been removed by the stop word removing and stemming algorithms. The developed model will train the simplex decision tree classification algorithm rather than the complex classification algorithm employed in the model used in the previous years. The model used in the previous years were processed over the entire sentence of the reviews, which caused the model to take a long time to process the drug reviews because the worthless words exists in the sentence.

The rapid growth in count of patients suffering from liver disease is a major concern all over the globe. Identification of persons having liver disease is done through liver biopsy and by visual checking of MRI by trained experts which is a tedious and time-consuming process. Therefore, there is a need to develop automated diagnosis system which can provide results in less time and with high accuracy. Researchers worked on this domain and came up with various models for detection of liver disease and its severity using machine learning algorithms. This paper presents a systematic and comprehensive review of the work done in this domain focusing on various machine learning techniques developed by various authors for prediction of liver disease. The performance comparison of the various algorithms is also discussed. This study also explores the datasets used by the various authors for liver disease prediction. Finally, in the conclusion section the challenges involved in liver disease prediction and future scope is discussed.

Dysgraphia is a common learning disability in children worldwide. It is characterized as a disturbance or difficulty in the production of written language presented through visual graphics. Almost 10–20% of school going children do face this issue. The child’s functional limitation in creating correct formation of letters or words, insufficient speed and legibility of written text is considered as Developmental Dysgrphia Disorder. Also the term developmental dysgraphia refers to the fact that a child is unable to get writing skills, in spite of the sufficient opportunities to learn in the absence of any neurological disorder. Because of dysgraphia children may have serious issues in their day to day life. It is proposed by various researchers that there might be serious consequences in a child’s academic, social and emotional behaviour because of handwriting difficulties. It is therefore required to detect it in the earlier phase.There are various scales which are developed to assess the handwriting quality. The Objective of this paper is to present various methods available for automatic detection of dysgraphia. We also presented a comparative study of existing research work for early detection of dysgraphia based on some already available measures.

A neural network for handwritten digit recognition is dependent on the dataset used to train it. Most of the techniques available in the literature recognize the handwritten characters of any script based upon the moment of inertia of the character uses a central point to it and suffers from certain drawbacks. In this paper, we develop methods that can be applied to any neural network, to make the dataset insensitive to brush size and insensitive to rotation applied to the digit based on the moment of inertia that the digit produces about an axis. All images in the dataset undergo pre-processing. First, the image is made independent of brush size. Then, the image is rotated to negate the rotation that is applied to the image according to the axis of the least moment of inertia. This pre-processing makes the dataset independent of brush size and rotation. The results show that when this preprocessing technique is applied to neural a network, it does makes it insensitive to rotation and brush size.

Since the corona virus has emerged, genuine clinical resources, such as a paucity of experts and healthcare workers, a lack of adequate equipment and medications, and so on, have reached their peak of inaccessibility. Several people have died as a consequence of the medical profession’s concern. Individuals began self-medicating due to a lack of supply, which exacerbated an already precarious health situation. A rise in new ideas for automation is being spurred by machine learning’s recent success in a varied variety of applications. In this paper, we have proposed a two-phase Decision Tree Classifier based on Artificial Neural networks (DTNN). The work is based on the satisfaction of the drugs among patients with the help of their comments as positive or negative polarity. The dataset of drugs used in this paper is Cymablta and Depopovera. The proposed results are compared with the existing methodology of Support Vector Machine Neural Network (SVMNN). The results are shown in graphical and tabular form which shows the efficiency of the proposed methodology.

The use of artificial intelligence (AI) within the finance industry can be considered as a transformative approach as it enables the financial institutions to enhance their performance capacity. The use of artificial intelligence within the finance sector helps the industries to streamline the processes and optimise their management efficiently for various types of operations pertaining to credit decisions-making, financial risk assessment and management and quantitative trading. The paper aims at analysing the proactive approach that can be taken with the use of AI in order to enhance effective management within the financial sector. The empirical study conducted in the paper utilizes various types of secondary materials with a qualitative approach. The findings of the study demonstrate the enhanced capacity of AI that can be used for a proactive approach, utilised for the assessment of risks or threats prior to any mismanagement incident. In this regard, fintech companies such as Enova, Ocrolus, ZestFinance, and DataRobot and so on have taken a predominant position in aiding the financial industries to use AI-based systems that aids the management process. However, the inclusion of AI within the financial sector is faced with certain challenges such as lack of knowledge regarding technological infrastructure, poor financial investment especially for government aided banks, unawareness of the employees and weak collaboration with the IT industry. Regardless, AI technologies in recent years have achieved great advancement, leading to the enhancement of its capacity to assist the effective management within the financial sector.

With the rapid increase use of WWW people from all age group shares their views, thought, ideas, suggestions on the internet. This leads the internet to become a hub for user-generated data. If this extracted data from the web is analyzed properly, then this may become the largest platform for making decisions for new product reviews, political issues, public issues, etc. But manually extracting and analyzing this content becomes a drastic task.So, for solving this problem a new area of research has been introduced that is Sentiment Analysis. It is the part of data mining where it extracts and analyzes the unstructured data. It is a domain where there is a huge scope of research. Various researchers already had researched different datasets with different Tools and Techniques on various domains like politics, sports, public issues and etc. This paper is focused on a literature survey for Sentiment Analysis on public transportation using different Tools and Techniques.

Nature has a great capability of utilizing waste in its ecosystem through recycling. Earth is a significant example of a circular economy. At the same time generation of waste is inevitable as well as it has association with various disposal principles. These days, by mimicking the various natural sources, the waste is treated as a possible resource, and the conversion of waste materials into a useful product, gaining a lot of focus. The growing awareness of usage of wastages, the depletion of limited natural sources, the consciousness of the environment of human health are key players by extending the life of waste materials and reusing them again after quality enhancements. The environmental, social, and governance (ESG) are the pillars of sustainability that encourages industries as well as nations to adopt circular economy concepts to achieve a zero-carbon economy in the coming years. Resources extraction and processing, and subsequent waste management are the major causes of carbon emissions. In this perspective, the adoption of circular economy principles by big technology players and their conscious shift towards zero carbon emissions pledge through 3R (reduce, reuse and recycling) principles were discussed.

The tremors caused by the COVID-19 epidemic have ushered in a new era of healthcare problems for people all across the globe. Significant problems and challenges have been observed in areas such as victim assistance, remote monitoring, health resources, and healthcare staff, among others. The goal of this study is to provide a comprehensive perspective of digital healthcare during the COVID-19 pandemic, which is now underway. Various efforts such as mobile applications, digital sites, and sophisticated analytics are being used in order to improve early diagnosis and overall healthcare management. While briefly describing the major aspects leading to widespread implementation of e - healthcare concepts, This research also sheds light on some key aspects of artificial intelligence and the Internet of Things, as well as their practical applications such as clinical decision support systems and predictive risk modelling, which are particularly relevant in the context of combating the COVID-19 pandemic’s practical difficulties. The potential uses of artificial intelligence and Internet of Things technology in the battle against the COVID-19 pandemic are thoroughly examined in this paper. The present and future applications of artificial intelligence and the Internet of Things are covered in depth, as well as a thorough assessment of the supporting tools and methodologies. An in-depth analysis.

The term deep learning is seen as an important part of artificial intelligence that allows the system to understand and make decisions without special human intervention. In-depth learning uses a variety of statistical models and programs that allow different computational properties to reach the highest point. It is estimated that the market development of artificial intelligence and technology for deep learning will amount to USD 500 billion by 2026. The use of advanced technology, such as neural networks, enables better image recognition and the use of automated processes for deep operations. The main purpose of the study is to understand the critical determinants of Deep Learning in Creating a better City through Intelligent Smart Waste Management, the major determinants cover: System usability scale, Implementation of RFID sensors and Optimizing route selection. The proposed work is that implementation of advanced tools like deep learning methodologies and machine learning tools can support in managing the waste in a smart way, this will enable in creating better cities, enhance the environment and support sustainable living. Smart cities today need to use tools like deep learning and other artificial intelligence to effectively manage waste. Smart vessels are mainly controlled and implemented, which makes it easier for users to open vessels, it is also suitable for storing solid and dry waste, but provides information on the total degree of filling, can share data and information with central waste management service, you can collect waste quickly and avoid flooding. To achieve this, governments, administrators and communities are introducing sensors that transmit data and information to the waste management company in real-time and take appropriate action.

The use of the Internet of Things (IoT) has brought about radical changes in the construction and business sectors, and companies are now using technology to remain competitive, support the exploitation of competitive advantages and increase growth and profitability. The use of the next generation of computers facilitated industrial change in all areas, and IoT helped shape the production structure, build an efficient value chain and achieve economic growth points. It can be argued that the introduction of IoT has changed the way we create value in the supply chain, which creates better opportunities for companies to improve scalability, make faster decisions and achieve better profits and growth. Although there are few challenges for the company, such as optimizing resources, investing in IoT and related digital technology, changing the production process and supply chain, etc., these new problems tend to change the organization’s bases and change the traditional way of doing things. Business. digital environment for effective customer engagement.

The Internet of Things (IoT) is a computer concept in which common items are enhanced with computational and wireless communication capabilities, generally through the inclusion of resource-constrained components such as sensors and actuators that allow them to access the internet. The Internet of Things (IoT) is regarded as a critical component in the implementation of intelligent environments. Nonetheless, the present IoT ecosystem provides a plethora of different connectivity options with varying performance characteristics. This circumstance makes determining the best IoT connectivity solution for a certain intelligent environment extremely difficult. In this article, we look at the specific requirements of major smart settings, such as the home automation, smart healthcare, smart urbanization, and advanced manufacturing, and compare them to modern Iot communication solutions. We define the key features of these smart settings before providing a detailed assessment of applicable IoT communication technologies and systems. The Internet of Things (IoT) has the potential to change organizations by automating operations ranging from inventory management to robotics and automation, therefore resulting in cost savings. Can, meanwhile, tiny businesses profit from IoT? This study investigated the increasing importance of the Internet of Things (IoT) in small companies, its influence on their capacity to engage in a fast-changing digital world, and their awareness, attitudes, opinions, and desire to embrace it. An initial exploratory method is used in the research, which is predicated on a study of instance studies in the literature, interviews with many economic development employees, and several small and medium-sized business executives. The adoption of IoT can lead to greater operational efficiency and cost savings in enterprises. These advantages have been confirmed by the medium-sized firms surveyed.

The Internet of Things (IoT) field has grown exponentially and, it is a significant emerging research field in this era. IoT consists of sensors, actuators, radio frequency identifier (RFID), and machine-to-machine communication. A failure may occur in IoT at layering architecture such as applications: sensor and actuator nodes can be lost, missed, and failed, network links can be down processing, and storage components of IoT can fail to perform appropriately. That is the reason for which data dissemination fault-tolerant (F.T.) has become a significant role for IoT-enabled systems to provide Quality of service (QoS). It can reduce the end-to-end delay, low energy consumption, and maximize throughput. The authors aim to identify and systematically review data aggregation, data dissemination, and F.T. mechanisms. Data dissemination has become effective when data aggregation, fault detection, and fault-tolerant has done at the same level of communication. The authors also discuss the theoretical aspect of efficient and fault-tolerant data dissemination using a group formation framework for IoT-enabled systems and different fault-tolerant techniques with a comparative survey.

The Internet of Things (IoT) is a worldwide network of physically interconnected devices that communicate with one another over an online platform. The Internet of Things envisions the connectivity of a few billion and billions of smart things around each other, each completely unique and accessible on a regular basis. Those very same objects will be able to collect, process, and share information with themselves and their surroundings. Medical services, developing smart city projects with the latest Industry 5.0 systems, civil and military monitoring, and collection of data are some of the biggest examples of IoT technologies and their applications. Intelligent sensors and actuators, as well as RFIDs, have been developed recently, resulting in a huge number of wireless networks with smart and intelligent equipment (objects, or things) interconnected to the Internet continually transmitting data. As a result, providing privacy and security for all this data in the IoT is a highly difficult challenge that must be prioritised for numerous present and future applications. Poor equipment updates, a shortage of effective security measures, consumer low awareness, and well-known active device surveillance are just a few of the difficulties that IoT is struggling with. IoT security solutions prevent unauthorised access by securing unwanted modifications or damage. Privacy methods retain the right to restrict the acquired information that is available for utilisation and objective. This article presents a comprehensive survey of IoT systems’ security and privacy with countermeasures.

The healthcare industry is developing rapidly, and innovations are now considered as the significant game-changer. IoT (Internet of Things) is shaping the healthcare industry in a new form with promising advances in testing, monitoring processes. Monitoring the health issues of the patients, organizing the treatment initiatives, and empowering the physicians it is providing superlative measures. The invention of the IoT through internet based artificial intelligence is determining the bright future of the medical field. Whether IoT is diagnosing the disease, or analyzing the past history of a certain disease the implementation of artificial intelligence is great. Here in this study the roles of internet based artificial intelligence are illustrated. Furthermore, it has described the current working features in health monitoring. Key aim of this study is to analyze this new innovative implementation in health monitoring. The article is developed including secondary qualitative analysis. Data collection, diagnosing health issues, and in monitoring the preventive care of IoT is compared with the traditional way of heath monitoring. Many experts see that artificial intelligence is more able than the conventional method to work in a more organized way. This study targets to analyze both the advantages, and disadvantages of implementation of artificial intelligence. Various components are addressed along with the gap to predict the increasing use of it in the near future. Comparing with the traditional; ways in giving better service experience is discussed. Including both the gaps, and benefits this study would be beneficial to give a better and effective understanding about the chosen topic.

The Research on the Internet of Things (IoT) has paved the way for a revolution in community services. It was found that the application of IoT in a smart city is mainly carried out without major human intervention. The different uses of IoT devices allow interfaces to available systems, improve communication and perform a variety of tasks. In addition, the availability of multiple IoT devices has been shown to be compatible to protect integrity, collect and analyze information, and improve functionality. The study is focused in understanding the critical role Artificial Intelligence empowered Internet of Things (IoT) for creating Smart city. The researchers have used extensive data analysis by collecting the data from the respondents to understand the importance of AI empowered IoT in creating better smart city. The analysis is focused in using regression tools and other key data analysis to test the hypothesis. Based on the overall analysis it is concluded that enhanced security and privacy; implementing smart sensors; implementation of Intelligent analytics and better collaboration and Networking has supported the organization and other stakeholders in creating better smart city.

The Internet of Things (IoT) has found extensive use in areas such as water management, waste management, and sustainable development. With its broad connectivity, the Internet of Things is a new and promising technology that has the potential to positively transform human existence globally. IoT allows low-energy devices to exchange information and interact with one another. Waste management is a daily chore that necessitates a huge number of labour resources and has an impact on natural, fiscal, efficient, and social elements. The rate of garbage generation has been magnified at an alarming rate as a result of fast urbanisation and expanding population. As a result, as the world faces global environmental problems, it is necessary to develop changes in waste management systems and technology to address issues that have never been addressed in such innovative ways. Many applications throughout the world have been using IoT to conduct various activities to provide unique services for wastage handling and maximise energy efficiency. These improvements allow IoT technologies to serve as a bridge between basic network-based systems and technologies that scan and gather data from the real environment, as well as deliver new services and applications that help people in a variety of ways. As a result, this research conducts a study of existing IoT-enabled waste management solutions and sustainable development. The goal is to get an understanding of the strengths and weaknesses so that changes and innovations may be made to effectively and efficiently manage waste while also maintaining a healthy environment in the communities. The impact of waste management on long-term economic, social, and environmental sustainability is also depicted in the literature review. The secondary data has been used in the research to gain a reliable conclusion of the study.

This article presents a review of the Internet of Things in the health area, focusing on the solutions that currently exist in home-oriented health. A very promising future is predicted with the appearance of portable smart devices, using protocols such as 6LoWPAN, which will allow the development of many applications for solving every-day problems in the health sector and the rapid implementation of the home-centered health model. In the various works and solutions consulted, the use of wireless technologies pre-dominates, such as: WPAN, WBAN, MBAN, Wi-Fi, WiMAX, ZigBee, Bluetooth, ANT, ultrawideband, ingestible sensors, and epidermal electronics, smart bandages, smartphone applications, RFID, RTLS and IPS. These technologies are widely used for biomedical census systems. This reflects a broad advance at the level of technologies and network architecture based on IoT, which provide specific solutions to problems in the health sector, especially in scenarios focused on home-centered health, which allows the maximum use of technology IoT in this area, very commonly called IoT Health. The above sounds promising for the health sector and ICT industry in general because it allows personalizing the health service, and accelerating its evolution.

Climate change, as well as the continuous and increasing exploitation of the planet’s natural resources, have forced future generations to reconsider their behavior. Because the availability of fossil fuels is finite, and their usage as an energy source causes pollution and harms the ozone layer, it is critical to take steps to mitigate or eliminate the environmental effect. Improving the network’s operation, lowering electrical losses, and incorporating a substantial share of renewable energy sources are just a few of the options available. These are the key reasons for the introduction of smart grids as a solution to the old power grid’s architecture, which have sparked international interest and support. Information was kept in a SQLite database for this project. The measuring nodes’ communications network was built utilizing the low-power, fault-tolerant ZigBee protocol. The developed supervision system allows for the reduction or deletion of third-party-managed communication nodes in some circumstances, resulting in increased independence and dependability. Through the growth of the National Software Industry and the usage of open-source platforms, it also provides for technical sovereignty.

The brain tumor is deadly disease. The correct and early identification is required to cure and improve the brain tumor patients’ quality of life. The manual identification of brain tumor is very difficult. Sometimes, it leads to wrong prediction also. Computer aided identification and classification helps the doctors to identify and classify tumors accurately. To mitigate the above issues, Convolutional Neural Network (CNN) along with autoencoder based feature selection was proposed. The auto encoder was used to extract the relevant feature and eliminate the noise. The CNN was correct choice for prediction of brain tumor. The MRI brain images were used for classification and prediction of brain tumors. The CNN has divided into e layers namely pooling layer, convolutional layer and dense layer. The Convolutional layer divides the images in to segments and extract the relevant features. The pooling layer downsize the image to reduce the computational complexity. The tumors are identified and classified in the dense layer. The BRATS 2013 and BRATS 2015 data set is taken for experimentation. The accuracy, sensitivity, specificity, precision and Dice Similarity Coefficient matrices is considered for performance assessment. The proposed methodology performed better when compared to traditional CNN, Decision tree and Bayesian classification.

Rumors are misleading information that are not sustained at the time of circulation and are not true at the time of verification. In other words, Rumors are set of linguistic, symbolic or tactile propositions whose veracity is not quickly or ever confirmed. As the use of social media platform has grown in recent years, incorrect information and rumors have circulated widely causing a significant influence on people’s lives. Rumors spreads faster than righteous news and spreads through social media. Because of the expansion of Internet and web technologies, it is now possible for anybody to post anything on online platforms such as blogs, comments on articles, post on social media, and so on, where false news, rumors, and true news are swiftly conveyed. This rapid and expansive spread of rumors has encouraged researchers to differentiate between rumors and non-rumors data. In this work, we have used stylometric and word vector features and put them into machine learning models. These features are extracted from the twitter-16 dataset and by applying SVM, we have attain the highest accuracy in compare to existing state-of-the-art studies.

We live in an era of smart phones, and the number of smart phone users is growing by the day, resulting in a rapid increase in the number of people with access to the internet and social media. According to reports, the average person spends about five hours per day on his or her mobile phone. This extensive use of phone and internet had a significant impact on the amount of data exchanged on social media platforms such as Instagram, Facebook, Whatsapp, Snapchat, and others. Images are one type of data. Every day, millions and billions of images are shared, and many of them may contain information that compromises an individual’s privacy. Hackers can use such information for malicious purposes and personal gain. They can even use this person’s personal information to blackmail or threaten them. Cybercriminals can also use that information to open a bank account in the victim’s name, create a forged driving licence, and other forms of identification. Although many researchers have addressed this issue by using tools such as Tensorflow and OpenCV to detect and obfuscate sensitive information in images. In this paper, we will use the YOLOv3 object detector to solve this problem. First, we’ll label the data that needs to be blurred, and then we’ll train our object.

Computer security is defined as the defense of computing systems against external threats in order to protect resource confidentiality, integrity, and availability. When an intrusion occurs, both network resources and the victim server are put at risk. When an intrusion occurs in a computer system or network, the Intrusion Detection System (IDS) tracks it and notifies the system administrator, allowing the appropriate action to be taken. People’s faith in the Internet has declined as the frequency of cyber-attacks has increased. A security attack known as Denial of Service (DoS) is quite successful (DoS). When an intrusion detection system (IDS) detects external attacks as well as system abuse or internal attacks, it sends a signal to a central monitoring station. In terms of functioning, an intrusion detection system is similar to a burglar alarm. This article provides a machine learning based intrusion detection system. NSL KDD data set is used as input for experimental work. ANN, SVM and ID3 algorithms are used in analytical investigation.

Identification and classification of fruits of different qualities are vital for fruit industries. Traditional techniques, such as visual inspection of fruits by handpicking, are time-consuming, tiresome, and error-prone. To automate the fruit inspection process and segregating them into different classes, computer vision and machine learning approaches has been applied and researched. For segregating fruits into different classes or qualities, Transfer Learning is one of the popular technique to build a fruit classifier. This paper evaluates the performance of VGG16, InceptionV3, Xception, ResNet152V2, and DenseNet by training and testing models using transfer learning. The experiment is conducted on two fruits datasets. Further, to improve the accuracy of fruit classification, the pre-trained model DenseNet is partially unfreezed and retrained. The results show that this model archives accuracy of 99.61% for fruit classification.

One of the most serious and overarching issues in our society is crime, which involves many crimes daily. Crime represents today’s greatest threat to humanity. The crime rate is steadily increasing; it is impossible to predict because crimes are not systematic or random but grow and spread rapidly. The victims of crimes are all small towns and large cities. It is the police department’s duty to control and reduce the crime that has increased the rate of crime. To do that, we need to deal with criminal matters much faster. This dataset includes the date and crime rate for each year. The crime rate of this project is based on murder, violence against women and auto theft. We use historical data to predict the percentage of crime in the future using a variety of regression algorithms using the accuracy scores.

This extensive review aims to classify the Intrusion Detection System (IDS) and various machine learning and deep learning (ML/DL) approaches used for IDS. The survey also addresses security, which is a concern with the Internet of Things. Several types of intrusion detection systems (IDSs), including shallow and deep learning methods and various learning algorithms to aid intrusion detection, are also categorized. This research expands on Network Intrusion Detection Systems and investigates techniques for improving their performance. It provides a more comprehensive understanding of deep and shallow learning methodologies with their benefits and drawbacks. The study component examines IDS classification, feature extraction techniques, machine learning, deep learning, and examples of how these may be applied. The essence of this review will establish a viable approach to assist professionals in modeling trustworthy and powerful IDS based on real-time requirements. Because the methods of intrusions and cyberattacks in networks are constantly evolving, it attracted the interest of many scholars and industrial professionals. However, cyber specialists struggle to develop an accurate and effective Intrusion Detection System (IDS). In addition, an increasing number of devices has resulted in more complicated network topology, raising security risks. As a result, a lengthy and exhaustive review is indispensable while developing a secure communication system.

A CNN (Convolutional Neural Network) is an artificial neural network used to evaluate visual pictures. It is used for visual image processing and is categorised as a deep neural network in deep learning. So, using real-time image processing, an AI autonomous driving model was built using a road crossing picture as an impediment. Based on the CNN model, we created a low-cost approach that can realistically perform autonomous driving. An end-to-end model is applied to the most widely used deep neural network technology for autonomous driving. It was shown that viable lane identification and maintaining techniques may be used to train and self-drive on a virtual road.

In recent years, all firms have been concerned with marketing analytics. They’re using numerous advanced technologies to analyse marketing analytics. Artificial intelligence (AI) and Deep Learning (DL) technology are highly capable of examining large databases for patterns and insights. These technologies enable the marketing function to encompass its reach and analytics empower a deeper understanding of how the market responds to actions. With the help of Deep learning, businesses are now able to connect a wide range of datasets to better understand what customers want with greater sophistication and analytic capacity, and then use that information to gain a competitive advantage. In addition, Deep learning uses numerous technical tools that excel at extracting perceptions and patterns from huge amounts of data and then predicting the future for marketing. Thus, Deep learning can potentially be used to create products that are tailored to what customers want. It has been found that marketing analytics play a significant role in HRM. It helps in examining the employees’ skill sets and developing a training programme based on the market demands. AI assists firms in determining target audiences and devising a strategy to meet its objectives. Also, AI technology adapts and learns from data to make data-driven decisions. Many time-consuming and managerial chores will be automated by HR software that includes artificial intelligence. A lot of administrative activities are automated and speeded up using AI. Moreover, to gain in-depth knowledge of Artificial Intelligence and Deep learning this paper is conducted. Furthermore, the article examines how artificial intelligence and deep learning technologies are utilised to assess marketing statistics, as well as their impact on human resource management systems. For this paper, descriptive research methodology has been used for this study, and secondary data has been used to obtain reliable conclusions.

Plant diseases are the significant elements impacting food supply and minimizing production losses; hence crop illnesses must be detected and recognized quickly. Deep learning approaches have recently expanded their applicability in plant disease identification, providing a comprehensive instrument with accurate findings. In this paper, we present a thorough assessment of the literature to determine the art of present state in the application of deep learning techniques for noticing and classification of diseases of plants and identifying trends and gaps.

The aim is to estimate a Windows device’s risk of being corrupted by different malware families based on the machine’s various properties. By merging movement and studies of threats obtained by Microsoft’s endpoint security solution, Windows Defender, the simple data containing these properties and system infections was created. A Machine Identifier uniquely identifies each row throughout this dataset, which relates to a machine. The similarity measure has Detections, which means that Malware was found on the computer. You should estimate the value for Detections of each system in test.csv using the details and labels in train.csv. Using the Malware data collection, this analysis tests the output of the new proposed classifier algorithm. That proposed updated version of the Random forest classifiers provides improved outcomes in case of intrusion detection and false incidence, according to the empirical findings of our research.

Urdu Nastaliq script is a cursive and ligature-based script that is very difficult to recognize and detect using current computer algorithms dues to its complicated writing structure . Handwritten text imposes much more difficulty since all writers have their own writing styles. Many researchers have worked in this field on various aspects, from creating datasets to improving the recognition rate for its text. This paper is a literature survey on the work done for Urdu handwritten text recognition by optical character recognition using deep learning and machine learning techniques. It also compares the accuracy rates of respective algorithms over different datasets.

The spread of false news on an online social media platform has been a major concern in recent years. Many sources, such as news stations, websites, and even newspaper websites, post news pieces on social media. Meanwhile, most of the new material on social media is suspect and, in some circumstances, deliberately misleading. Fake news is a term used to describe this type of information. Large volumes of bogus news on the internet have the potential to generate major societal issues. Accepting the stories and pretending that they are true is extremely harmful for our community. Many people believe that false news affected the 2016 presidential election in the United States. The term has since become commonplace as a result of the election. It has also attracted the interest of industry and academics, who are trying to figure out where it comes from, how it spreads, and what impacts it has. In this work, we looked at a number of different papers and compared all of the strategies for detecting false news.

Sentiment analysis or opinion mining is a prominent and most demanding research topic in today’s world. The main idea behind this research topic is to recognize the user’s opinions and emotions towards the aspect of service or product via a text basis. Sentiment analysis involves mining text, lexicon construction, extracting features and finally finding polarity of text. Even though numerous amounts of research work were conducted in this field through different methods, opinion mining is still considered a challenging field for research.Most of the prior research concentrated on the binary or ternary classification of sentiments such as positive, negative, neutral. Some studies have done an analysis of Twitter sentiment based on ordinal regression, but by turning the problem of ordinal regression into a problem of binary classification. The aim of this study is to review the multiclass sentiment analysis of Twitter text data using an automated i.e., machine learning approach. This review paper intends to focus on existing work for Twitter sentiment analysis with multiple polarity categorization and explore gaps with future scope in the said research area.

Social media streams have brought about a new way of information dissemination in today’s world. People use social media streams such as Twitter, Facebook and so on, and see it as an important means of communication. However, huge amount of data generated real time is difficult to manage. One such challenge is the efficient clustering of data to check for a potential event. To tackle this problem, we applied different approaches to find the optimal clustering for event detection using Twitter as our primary source of data. Also, we proposed a theoretical model for purging of dead events. Although the work is experimental, it holds great potential to be further established as an efficient and effective event detection approach.

With the advancement in recommendation techniques, focus is diverted from just making them more accurate to making them fairer and diverse, thus catering to the set of less-popular items (the long tail) that often get neglected due to inherent biases in recommender systems. Popularity bias has been recently acknowledged as a major bias of critical concern in the field of recommender systems. Although research on popularity bias has gained pace from the last couple of years, this field is believed to be still in its infancy. To advance research in this area, this paper thoroughly investigates current state of the art and could have a very positive impact on further research in popularity bias. Besides the mitigation techniques discussed in this paper, allied evaluation metrics that were used in measuring popularity bias have also been discussed.

This paper analyzes the performance of text classification using machine learning and deep learning models. The data is preprocessed, followed by text preprocessing. Natural language processing requires the conversion of text data to numerical vectors before they are passed to machine learning or deep learning models. Bag of words and Term Frequency and Inverse Document Frequency (TFIDF) techniques are used for converting text to a numeric vector. The machine learning models demonstrated are Naïve Bayes Classifier and Logistic Regression. Here, the Bag of words is used with Naïve Bayes Classifier, and TFIDF is used with the logistic regression model. The deep learning models demonstrated here are Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU). The data set used consists of 1786 instances. Train test ratio used is 80:20. The performance of text classification models depends upon the dataset and how well its features are preprocessed as well as the selection of train test data. In this paper, four different algorithms are applied to the same dataset to analyze the performance of text classification.

Recent advances in Recommender Systems have shifted the attention of researchers towards fair and unbiased recommendations. A growing number of users use Recommender Systems for decision making and information retrieval and in turn have a significant impact on public opinion. Therefore, it is of pivotal importance to address the unfairness issue in recommender systems for fair inclusion of disadvantaged groups. The study of fairness in recommender systems is a relatively new field with a vast scope for further research and improvement. This study presents a thorough investigation of existing metrics in fairness evaluation from different contexts like user fairness, item fairness, group fairness, individual fairness, multi-sided fairness, etc.

In data mining, association rule mining algorithms executing over massive datasets, generate vast amounts of rules. These rules pose a gruesome strain of knowledge post-processing on the user. This compels the user to dig through the rules to find relevant knowledge for decision making. For simplification of this practice, we propose the concept of association rule chains. Contrary to conventional association rules, the association rule chains map the values from dataset according to a user defined hierarchy of features, irrespective of the value constraints of interestingness measures. This mapping of values into associations rule chains relies on feature-wise hierarchical clustering of data items with the feature at the root of hierarchy termed as profile. The discovered association rule chains allow the user to assemble the domain relevant information from limited profile based subsets of the search space into a comprehensive feature specific knowledge base. In this paper, we demonstrate our approach by implementing it on a section of global terrorism database (GTD) of terrorist attacks for 4 regions. Our approach facilitates modeling of event profiles for specific classes of terrorist attack events by generating the relevant association rule chains for preemptive event analysis. Finally, we evaluate our approach against occurrences of terrorist attacks from the test dataset.

Due to the suspended particles in the atmosphere formed by hazy or foggy weather conditions produces a degraded image which causes change of color, reduced contrast and poor visibility. Practically haze cannot always be uniform in nature sometime non uniform haze can be formed. On the basis of haze formation single image dehazing can be categorized into two type’s i.e. homogeneous and non homogeneous haze. In this paper we have discuss different methods of single image dehazing used for homogeneous and non homogeneous haze. As qualitative parameter such as Peak Signal to Noise ratio (PSNR) and Structure Similarity Index Measures (SSIM) is used to check the quality of output dehazed image. On the basis of result obtained we have compared different techniques of single image dehazing. Average PSNR and SSIM value also calculated for 30 set of samples for both homogeneous and non homogeneous hazy images. Result is also analysis with plotted graph of average PSNR and SSIM value.

It is essential to store speech and audio data in various speech processing and detection applications. Sometimes this data is too much bulkier and requires ample storage. So, there is a simple method to compress the signal and then again decompress. But this method is too much old and requires a higher sampling rate as per Nyquist's theorem. Proper reconstruction can also be done with the least values of samples, and the Compressive Sensing Theory proves it. This work will elaborate on how efficiently this technique can reconstruct speech and audio signals using very few measurements. We will also represent the effect of the male and female voices on the compression process. The effectiveness of the reconstruction is measured by using MSE & PSNR values. The time-frequency response for the actual and reconstructed signal is also represented in this paper by using that we can easily understand the compressive sensing phenomena.

The potential of distributed generation to deliver cost-effective, ecologically sustainable, rising, and more reliable solutions has resulted in an increase in its use for electricity generation globally. DGs and capacitors are chosen over centralized power generation to bring electricity demand closer to load centres. The proper positioning and capability of DG is vital in solving common power system concerns such as power network loss reduction, stability, and voltage profile enhancement. An Analytical method is used in this paper to determine optimal size and site for placement of DG units. Type-I and type-II DGs are integrated in the system to achieve the objective. Analytical method is tested on IEEE 69-bus system at varying penetration levels (30–70%). The results are compared to those obtained by other latest approaches and found superior.

Just In Time Defect Prediction abbreviated as JITDP refers to a software that helps to detect whether any change made to that software leads to defect or not and that too immediately in no time. That is why the name of the model is just in time defect prediction. It helps to develop a defect free software. Various researchers have proposed their models to showcase their accuracy, efficiency and also the scenario in which they perform the best. A systematic review is performed on the proposed approaches which have the following conclusions:1. Random forest algorithm proves to be the most efficient algorithm and outperforms all the other proposed models. 2. Random Forest is observed to be a generalized algorithm which can be used from various perspectives. 3. Apart from random forest, deep learning fusion technique, AB+LMT and DeepJIT are also efficient algorithms with an average accuracy of 0.80. 4. CBS+ is the most efficient algorithm in case of long term approach followed by random forest.

In today’s world, images are used to extract information about objects in a variety of industries. To retrieve images, many traditional methods have been used. It determines the user’s query interactively by asking the user whether the image is relevant or not. Graphics have become an important part of information processing in this day and age. The image is used in image registration processing to extract information about an item in a variety of fields such as tourism, medical and geological, and weather systems calling. There are numerous approaches that people use to recover images. It determines an individual’s query interactively by asking users whether the image is relevant (similar) or not. In a content-based image retrieval (CBIR) system, effective management of this image database is used to improve the procedure’s performance. The study of the content-based image retrieval (CBIR) technique has grown in importance. As individuals, we have studied and investigated various features in this manner or in combinations. We discovered that image Registration Processing (IRP) is a critical area in the aforementioned industries. Several research papers examining color feature and texture feature extraction concluded that point cloud data structure is best for image registration using the Iterative Closest Point (ICP) algorithm.

Data privacy and security is the need of the hour. The importance elevates further when we submit the data over a network. It is very easy to extract personnel’s history like illness, vulnerability etc. through the information posted. Through this research, a technique for data compression and abstraction, particularly in columnar databases is proposed. It provides domain compression at the attribute level to both row and columnar databases using Attribute Domain Compression (ADC). Large datasets may be stored more efficiently by using this approach to minimize their size. Because we want the second process to be as flexible as possible, we give it the value (n) so that it can find all $${\text {n}}-1$$ n - 1 more tuples.

COVID-19 has caused havoc on educational systems around the globe, impacting over 2 billion learners in over 200 countries. University, college, and other institutional facility cutbacks have impacted more than 94% of the world’s largest population of students. As a result, enormous changes have occurred in every aspect of human life. Social distance and mobility restrictions have significantly altered conventional educational procedures. Because numerous new standards and procedures have been adopted, restarting classrooms once the restrictions have been withdrawn seems to be another issue. Many scientists have published their findings on teaching and learning in various ways in the aftermath of the COVID-19 epidemic. Face-to-face instruction has been phased out in a number of schools, colleges, and universities. There is concern that the 2021 academic year, or perhaps more in the future, will be lost. Innovation and implementation of alternative educational systems and evaluation techniques are urgently needed. The COVID-19 epidemic has given us the chance to lay the groundwork for digital learning. The goal of this article is to provide a complete assessment of the COVID-19 pandemic’s impact on e-learning and learning multiple papers, as well as to suggest a course of action. This study also emphasises the importance of digital transformation in e-learning as well as the significant challenges it confronts, such as technological accessibility, poor internet connectivity, and challenging study environments.

A quantitative data analytic tradition is the Exploratory Data Analysis (EDA) based on the original work of John Tukey. In Statistics, EDA is the process of cleaning data and using appropriate visualizations to extract insights and summarize the data in the given dataset. The computational and core conceptual tools of EDA cover the use of interactive data display and graphics, diagnosing, evaluating, an emphasis on model building, and addressing the issues of the fundamental measurements that are consorted with several distributions and also undertaking some of the procedures that are resistant to mislead or flawed results because of the unpredictable change of real world data. EDA can be further classified as Graphical or non-graphical and Univariate or multivariate data. EDA can be helpful for expressing what does the data refers before the modeling task. It is not easy to see at a large dataset or a whole spreadsheet and to decide important characteristics of the data. It may be typical to obtain insights by seeing at plain numbers. So, EDA techniques have been evolved as an assist in this type of situation. In this article we are going to take a nutritional dataset and visualize it using EDA. This visualization contains Bar chart, Histogram, Boxplot, Scatterplot, Barcode and Violin plot.

We describe how to integrate data-driven predictive maintenance (PdM) in machine decision-making and data collection and processing. A brief overview of maintenance methods and procedures is provided. In this article is a solution for a real-world machining issue. The answer requires many stages, which are clearly described. The outcomes demonstrate that Preventive Maintenance (PM) might be a PdM method in an actual machining process. To offer a Visual examination of the cutting tool’s RUL. This study proves shown in one procedure but reproducible for most of the series piece productions.

Dimensionality Reduction is a technique that performs feature extraction to enhance the evaluation of the model which uses data mining techniques. When the number of features in the model increases, the number of samples taken for tested also increases proportionally. The samples should also be selected to include all possible combination of the features. Web Mining plays a significant role in understanding the user’s browsing behavioral pattern on their web usage data. Web Usage mining records the user’s visited web page in their website. Each user’s visited umpteen number of pages as well as each website also contains more number of web pages. These pages help to build the model which finds the user’s behavioral pattern leads to a complex model which may have lot of inconsistent or redundant features in the data there by increase the computation time. When a greater number of features are included, there is also a high possibility of overfitting. Thus, to address the issue of overfitting and increase in computation time dimensionality reduction technique is introduced. Feature extraction technique helps to find the smaller set of new pages with the combination of the given independent web pages. Comparing three dimensionality reduction techniques such as PCA, Isomap and KPCA in which non-linear reduction technique of KPCA is used. Furthermore, unsupervised clustering technique of fuzzy Clustering technique is applied to identify the similar user behavioral pattern. Empirical study proved that the clustering accuracy is improved with the non-linear dimensionality reduction technique of KPCA is used with CTI dataset of DePaul University and MSNBC dataset is taken from the Server logs for msnbc.com

The study shows that numerous academic researchers are utilizing machine learning and artificial intelligence approaches to regulate, administer, and run networks, as a result of the recent explosion in interest in these fields. In contrast to the scattered and hardware-centric traditional network, Software Defined Networks (SDN) are a linked and adaptive network that offers a full solution for controlling the network quickly and productively. The SDN-provided network-wide information may be used to improve the efficiency of traffic routing in a network environment. Using machine learning techniques to identify the fewest overloaded path for routing traffic in an SDN-enabled network, we investigate and demonstrate their application in this study. These years have seen an increase in the number of researchers working on traffic congestion prediction, particularly in the field of machine learning and artificial intelligence (AI). This study topic has grown significantly in recent years on account of the introduction of large amounts of information from stationary sensors or probing traffic information, as well as the creation of new artificial intelligence models. It is possible to anticipate traffic congestion, and particularly short-term traffic congestion, by analyzing a number of various traffic parameter values. When it comes to anticipating traffic congestion, the majority of the studies rely on historical information. Only a few publications, on the other hand, predicted real-time congestion in traffic. This study presents a comprehensive summary of the current research that has been undertaken using a variety of artificial intelligence approaches, most importantly distinct machine learning methods.

In practical scenarios, an entity presence can depend on existence probability instead of binary situations of present or absent. This is certainly relevant for information taken in an experimental setting or with instruments, devices, and faulty methods. High-utility patterns mining (HUPM) is a collection of approaches for detecting patterns in transaction records that take into account both object count and profitability. HUPM algorithms, on the other hand, can only handle accurate data, despite the fact that extensive data obtained in real-world applications via experimental observations or sensors are frequently uncertain. To uncover interesting patterns in an inherent uncertain collection, potential high-utility pattern mining (PHUPM) is developed. This paper proposes a Spark-based potential interesting pattern mining solution to work with large amounts of uncertain data. The suggested technique effectively discovers patterns using the probability-utility-list structure. One of our highest priorities is to improve execution time while increasing parallelization and distribution of all workloads. In-depth test findings on both real and simulated databases reveal that the proposed method performs well in a Spark framework with large data collections.

Limited node energy has been always remained key challenge for the designers of clustering techniques in Wireless Sensor Networks (WSNs). From homogeneous to heterogeneous WSNs, there is no best fit clustering technique; therefore plenty of different techniques have been proposed by researchers from time to time. This paper aims to develop a generalized deterministic framework, Penetration Testing Framework (PTF), for researchers to validate their proposed techniques on a common platform giving best fit techniques for homogeneous as well as heterogeneous WSNs. The PTF consists of a set of network scenarios having different network parameters and performance parameters mixes. Some of the most relevant as well as popular clustering techniques designed for homogeneous and/or heterogeneous WSNs have been tested under PTF to validate its applicability and generality. A rigorous in-depth analysis has been made through development of different network scenarios and performance parameters mixes. The results attained are quite interesting, having useful directions for the designers of clustering techniques for modern WSNs.

Containers have existed for many years, Dockers have not invented anything in this sense or almost nothing, but it should not be taken away from them. Every timeless monolithic application is developed and based on modules, allowing a more agile, fast, and portable development. Well-known companies like Netflix, Spotify, Google, and countless startups use microservices-based architectures in many of the services they offer. Limit and control the resources that the application accesses in the container, they generally use their file system like UnionFS or variants like aufs, btrfs, vfs, overlayfs, or device mapper which are layered file systems. The way to control the resources and capabilities it inherits from the host is through Linux namespaces and cgroups. Those Linux options aren’t new at all, but Docker makes it easy, and the ecosystem around it has made it so widely used. In this paper, different Docker security levels are applied and tested over the different platforms and measure the security attribute to access the legitimate resource. Additionally, the flexibility, comfort, and resource savings of a container provided by a virtual machine, or a physical server are also analyzed.

Low power and lossy networks (LLN) are flourishing as an integral part of communication infrastructure, particularly for growing Internet of Things (IoT) applications. RPL-based LLNs are vulnerable and unprotected against Denial of Service (DOS) attacks. The attacks in the network intervene the communications due to the inherent routing protocol’s physical protection, security requirements, and resource limitations. This paper presents the performance analysis of the Hatchetman attack on the RPL based 6LoWPAN networks. In a Hatchetman attack, an illegitimate node alters the received packet’s header and sends invalid packets to legitimate nodes, i.e. with an incorrect route. The authorised nodes forcefully drop packets and then reply with many error messages to the root of DODAG from all other nodes. As a result, many packets are lost by authorised nodes, and an excessive volume of error messages exhausts node energy and communication bandwidth. Simulation results show the effect of Hatchetman attack on RPL based IoT networks using various performance metrics.

In India, small scale farmers face challenges acquiring crop loans and insurance. The government of India has many schemes for both. However, due to lapses in processing, these commercial bank loans are difficult to avail. Thus farmers approach Microfinance Institutions (MFI) or moneylenders, who charge high interest rates. To eliminate this difficulty, a novel blockchain based module is proposed in this paper, incorporating farmers’ credibility, verifying insurance claims, and assured guaranteed repayment of loans to investors. The inherent property of blockchain ensures trust and transparency. The proposed blockchain module has four participating entities, the farmer, investors, customers, and the government agent, governed by the smart contract rules. The concept of trusted oracles is used to gather weather data, and the smart contract decides the payout and the subsidized loan repayment. This module is implemented and tested in the Ethereum live test network. Performance evaluation shows that the proposed module has an average latency of 30 s and minimal fees.

In this generation we know how much the internet matters to us. Every basic work can be done with the help of the internet. People have nowadays become very dependent on internet so it is very much needed to provide the users with enough security so that it becomes a reliable platform. This security can be provided by different cryptographic techniques. This makes it difficult for the hackers and the cryptanalyst from stealing crucial informations. This can be done efficiently with the help of asymmetric key and symmetric key cryptographic techniques. We can make a system more safe if we can combine both. So here we have we have used two layers of each encryption and decryption thus providing two layers of security. Here we have worked with different images so firstly, the image has been encrypted using Rabin Cryptography followed by ECC Cryptography. Similarly while decrypting, the decryption was first done using ECC and then Rabin Cryptography. We have tested this approach using different datasets and have received great accuracy. With this approach it can provide images with a double layer of security efficiently while sharing images.

Biometric authentication refers to the process of confirming the identity of an individual based on his physiological or behavioral features. This authentication method has outperformed the traditional authentication method of using passwords, one time pin etc. and now emerged as an overwhelming method of authentication. The biometric-based authentication is categorized into Unimodal and Multimodal authentication out of which Multimodal authentication provides better results in terms of accuracy, intrusion avoidance and user acceptance. This paper discusses some state-of-the-art Multimodal authentication systems along with their advantages as well as disadvantages. Also, some future research directions have been provided based on these disadvantages. Lastly, the utilization of deep learning methods in current Multimodal systems has been discussed followed by the analysis of some state-of-the-art Multimodal systems based on fingerprint and face traits.

The study aims to identify modes of fraudulent payments and create awareness of such incidences to avoid decisive virtual activities. Disruptive developments such as contactless payments, mobile payments, and e-cloud payments witnessed a large-scale data breach. As a result, new fraud pathways have emerged and made traditional detection technologies less effective. Thus, there is a need to understand the contemporary fraud induced technology and corrective remedies—a relevant assessment of published approaches for detecting fraudulent card payment is critical to study. The researchers tried to identify fundamental issues using AI-ML-DL to detect fraud. An intellectual computing strategy is offered as a capable research avenue while boosting commercial data patronage. The paper discusses a standard method used by cybercriminals to mislead individuals and businesses. The study also focused on the methods utilized by cybercriminals and the economics of each confronted gadget. The paper talked about how systems can detect and block cybercriminals in three domains: card payment fraud, mobile payment fraud, and telephonic fraud.

Healthcare system is one of the most prime concerns for mankind. The pandemic compelled to search new alternative solutions for present healthcare system and results in telehealth. The paper proposed a model for the “Telehealth and Medicine System” based on one of the most powerful transparent, decentralized, and secured technology i.e., “Blockchain”. The basic aim of the proposed system is to provides the best remote healthcare facilities to people speedily and securely. The model also supports for managing the healthcare resources information for controlling the huge load of patients in hospitals. Present centralized system fails to maintain security, confidentiality, functional transparency, health-records immutableness, and traceability of medical apparatus and medicines. The present centralized system is also unable to detect scams correlated to patient’s transaction prerogatives and doctor authorized prescription data. In this paper we explore the possible techniques and opportunities of health care system using blockchain technology in telehealth and medicine area. This paper provides the illustration of blockchain technology which gives essential data safety and confidentiality to this model. Moreover, the paper deals with secure functioning-transparency, health-records immutableness, and traceability of medical apparatus and medicines information. Smart Contract is one of the most popular and secure application of ‘Blockchain’. It is based on public-blockchain framework. Blockchain technology can improve telehealth and medicine system by offering distant healthcare facilities on a decentralized platform along with tamper-proof transactions, transparency of data, flexible traceability of records, reliability to patients, and high security. This model identifies whole scenario of medicine and apparatus production from initial to final stage of delivery to the patient. This model also leads health specialists and doctors for treating patients correctly and share the prescriptions securely over the network. At the last, paper discuss numerous opportunities and techniques needs to be resolute the present system by acceptance of power decentralized technology ‘Blockchain’ in telehealth and health-care systems.