Ebook: Recent Trends in Intensive Computing
In a world where computer science is now an essential element in all of our lives, a new opportunity to disseminate the latest research and trends is always welcome.
This book presents the proceedings of the first International Conference on Recent Trends in Computing (ICRTC 2021), which was held as a virtual event on 21 – 22 May 2021 at Sanjivani College of Engineering, Kopargaon, India due to the restrictions of the COVID-19 pandemic. This online conference, aimed at facilitating academic exchange among researchers, enabled experts and scholars around from around the globe to gather for the discussion of the latest advanced research in the field despite the extensive travel restrictions still in place. The book contains 134 papers selected from 329 submitted papers after a rigorous peer-review process, and topics covered include advanced computing, networking, informatics, security and privacy, and other related fields.
The book will be of interest to all those eager to find the latest trends and most recent developments in computer science.
We are glad to introduce you to the 2021 1St International Conference on Recent Trends in Computing (ICRTC 2021), which was successfully held on May 21–22. Different from the previous four times, ICRTC 2021 was carried out in the form of a virtual conference due to the impact of COVID-19. Because there is a worldwide travel restriction, we held this flexible online conference to gather experts and scholars around the globe with the aim to continue disseminating the latest advanced research in the field of Recent Trends in Computing and developing the academic exchange among researchers.
The committee of ICRTC 2021 consists of about hundred and thirty-four experts in the area of Computing Science from home and abroad. With its professional and powerful influence, we are honored to have invited five renowned experts as our keynote speakers. Prof. Fadi Al-Turjman, is a Professor at Near East University Applied Artificial Intelligence Research Centre, Turkey., held a speech on recent computing trends in artificial intelligence. In addition, Mr.Nimit Shishodia, CEO, Ecode Networks, UK., performed his speech entitled A 5G and Research Opportunities. Dr. Parikshit Mahalle, BOS Member, SPPU Professor & Head, Computer Engg. Dept, SKN COE, Pune., performed his speech entitled Future of Data Science. Mr. P. Vinoth Kumar, Software Quality Analyst, Glencore Information Services Pvt Ltd, UK., performed his speech entitled Emerging Trends in the area of Extended Reality. In addition, Mr. R. Sairam, CFO, Cybertheronai Pvt. Ltd, Singapore., performed his speech entitled Security using ML and ELK. The keynote speeches were the first session of the conference, which was each delivered in about 30 minutes via Google meet.
We are glad to share with you that we still received lots of submissions from the conference during this special period. Hence, we selected a bunch of high-quality papers and compiled them into the proceedings after rigorously reviewing them. These papers feature the following topics but are not limited to Advanced Computing, Networking, Informatics, Security and Privacy, and other related fields. All the papers have been through rigorous review and process to meet the requirements of international publication standards.
We would like to express our sincere gratitude to the Shri Amit Kolhe, Managing Trustee, Dr. A. G. Thakur, Director, and Dr. D. B. Kshirsagar, Professor and Head, Department of Computer Engineering Sanjivani College of Engineering, Kopargaon, India, the distinguished keynote speakers, as well as all the participants. We also want to thank the publisher for publishing the proceedings. May readers gain some valuable knowledge from the book. We are expecting more and more experts and scholars from all over the world to join this international event next year.
The Book Editors
Prof. Dr. M. Rajesh, and Prof. Dr. K.Vengatesan
Department of Computer Science and Engineering,
Sanjivani College of Engineering, Kopargaon, Maharashtra. India
Prof. Dr. J. M. Gnanasekar
Department of Computer science and engineering,
Sri Venkateswara College of engineering Chennai, Tamil Nadu, India
Dr. Sitharthan R
School of electrical engineering,
Vellore Institute of Technology, Vellore, Tamilnadu, India
Dr. A. B. Pawar, Dr. P. N. Kalvadekar, and Dr. P. Saiprasad
Department of Computer Science and Engineering,
Sanjivani College of Engineering, Kopargaon, Maharashtra. India
In this digital era, most of the hospitals and medical labs are storing and sharing their medical data using third party cloud platforms for saving maintenance cost and storage and also to access data from anywhere. The cloud platform is not entirely a trusted party as the data is under the control of cloud service providers, which results in privacy leaks so that the data is to be encrypted while uploading into the cloud. The data can be used for diagnosis and analysis, for that the similar images to be retrieved as per the need of the doctor. In this paper, we propose an algorithm that uses discrete cosine transformation frequency and logistic sine map to encrypt an image, and the feature vector is computed on the encrypted image. The encrypted images are transferred to the cloud picture database, and feature vectors are uploaded to the feature database. Pearson’s Correlation Coefficient is calculated on the feature vector and is used as a measure to retrieve similar images. From the investigation outcomes, we can get an inference that this algorithm can resist against predictable attacks and geometric attacks with strong robustness.
Indian Sign Language is one of the most important and widely used forms of communication for people with speaking and hearing impairments. Many people or communities have attempted to create systems that read the sign language symbols and convert the same to text, but text or audio to sign language is still infrequent. This project mainly focuses on developing a translating system consisting of many modules that take English audio and convert the input to English text, which is further parsed to structure grammar representation on which grammar rules of Indian Sign Language are applied. Stop words are removed from the reordered sentence. Since the Indian Sign Language does not support conjugation in words, stemming and lemmatization will transform the provided word into its root or original word. Then all the individual words are checked in a dictionary holding videos of each word. If the system does not find words in the dictionary, then the most suitable synonym replaces them. The system proposed by us is inventive as the current systems are bound to direct conversion of words into Indian Sign Language on-the-other-hand our system aims to convert the sentences in Indian Sign Language grammar and effectively display it to the user.
Now a day’s network security becomes more important to organizations, government offices. With the fast advancement of the innovation, assaults throughout the years have turned out to be both progressively various and modern. Ransomware attack becomes one of the most popular weapons for network attackers that ransomware attack is increased rapidly year by year. The study shows that the ransomware attack is one of the top attacks that are most attacked malware by the attackers. In this paper, we focus on mitigation techniques that can be used to recover and mitigate the ransomware attack. The mitigation or recovery approach is very difficult as ransomware is depending upon cryptographic algorithms which are very difficult to crack.
Online financial transactions play a crucial role in today’s economy. It becomes an unavoidable part of the business and global activities. Transaction fraud executes thoughtful intimidations to e-commerce spending. Now-a-days, the online contract or business is fetching additional sound by knowing the types of online transaction frauds associated with, these are raising which disturbs the currency accompanying business. It has the capability to confine and encumber the contract accomplished by the intruder from an honest consumer’s credit card information. In order to avoid such a problem, the proposed system is established transaction limit for the customers. Efficient data is only considered for detecting fraudulent user action and it happens only at the time of registration. Transaction which is happening for any individual is not at all known to any FDS (Fraud Detection System) consecutively at the bank which mainly issues credit cards to customers. To speak out this problem, BLA (Behaviour and Location Analysis) is executed. The FDS tracks at a credit card provided by bank. All the inbound business is directed to the FDS aimed at confirmation, authentication and verification. FDS catches the card particulars and matter to confirm that the operation is fake or genuine. The pick-up merchandises are unknown to Fraud Detection System. If the transaction is assumed to be fraud, then the corresponding bank declines it. In order to verify the individuality, uniqueness or originality, it uses spending patterns and geographical area. In case, if any suspicious pattern is identified or detected, the FDS system needs verification. The information which is already registered by the user, the system identifies infrequent outlines in the disbursement method. After three invalid attempts, the system will hinder the user. In this proposed system, most of the algorithms are checked and investigated for online financial fraud detection techniques.
Sentiment Analysis includes methods and techniques for businesses to understand and analyze customer reviews, feedback and opinion on a particular product or service. Sentiment Analysis uses Natural Language Processing (NLP) tools to analyze feelings or emotions, attitudes, opinions, thoughts, etc. behind the words. Sentiments such as positive, negative and neutral are associated with a particular product. Sentiment analysis is applicable in multi-domains such as customer feedback for a particular product, movie reviews, social and political comments. This survey basically focuses on different aspect-based word embedding models and aspect-based sentiment classification techniques, where the goal is to extract key features from the sentences and classify sentiment on entities at document level. Aspect Based Sentiment Analysis (ABSA) is a technique that concentrates not only the entire sentence but analyses key terms explicitly to predict the polarity as a whole. ABSA model accepts aspect categories and its corresponding aspect terms to generate sentiment corresponding to each aspect from the text corpus. This article provides a comprehensive survey on different word embedding models under CNN framework for aspect extraction and different machine learning techniques applicable for sentiment classification purpose.
In today’s world, technology is constantly evolving; various instruments and techniques are available in the agricultural field. And within the agrarian division, the IoT preferences are Knowledge processing. With the help of introduced sensors, all information can be gathered. The reduction of risks, the mechanization of industry, the enhancement of production, the inspection of livestock, the monitoring of environment conditions, the roboticization of greenhouses, and crop monitoring Nearly every sector, like smart agriculture, has been modified by Internet-of-Things (IoT)-based technology, which has shifted the industry from factual to quantitative approaches. The ideas help to link real devices that are equipped with sensors, actuators, and computing power, allowing them to collaborate on a task while staying connected to the Internet, dubbed the “Internet of Things” (IoT). According to the World Telecommunication Union’s Worldwide Guidelines Operation, the Internet of Things (IoT) is a set of sensors, computers, software, and other devices that are connected to the Internet. The paper is highly susceptible to the consequences of its smart agriculture breakthrough.
The Absolute time monitoring, detecting and Alerting System for vehicles and children is required to trace and transmit the collected information at regular intervals to ensure safety and security of children. The illustration of the Realtime detecting and warning System consists of two units: Tracing Unit that traces the location information, transfers to the monitoring area, records the data in the database and takes the help of these data to locate the exact point of area of the vehicle with Google/other maps. The second unit is Alerting Unit that tracks the students using active Radio Frequency Identification Devices (RFID)which will be placed on student ID card. radio-wave trans-receiver transmits a common radio wave which is received by the RFID in the ID card. This radio-wave is modified by the RFID’s coil and resent to the receive RFID tags are also used for attendance which is updated directly to the database and displays the other student information.
Safeguarding and checking of air quality has becoming a most important practice in various modern and metropolitan zones now a days. The behaviour of air is unfavourably influenced by various types of contamination that we get by the transportation in production of power, natural resources utilizes and so for the emission of different poisonous gases is making a major danger for the personal life in various urban communities. The contamination of air is getting expanded day by day. Our main task is to provide a effective air contamination prediction observing techniques by which we can gather the data about infectious or poisonous gases present in every zone and gives us an idea of air contamination in each zone. Therefore, air quality measuring has become amongst important technique. The shape of air is influenced by multi-dimensional elements contains of area, time and unsure factors. As of now, various specialists get started to use the very large information investigation methodology because of various ways in large information applications and using the resources of nature detecting organisations and information sensed by sensor. The examination researches different various information and air quality evaluation methods using artificial intelligence methods. Moreover, it identifies and predict the future explanation needs.
In this era everything is digitalized we can find a large amount of digital data for different purposes on the internet and relatively it’s very hard to summarize this data manually. Automatic Text Summarization (ATS) is the subsequent big one that could simply summarize the source data and give us a short version that could preserve the content and the overall meaning. While the concept of ATS is started long back in 1950’s, this field is still struggling to give the best and efficient summaries. ATS proceeds towards 2 methods, Extractive and Abstractive Summarization. The Extractive and Abstractive methods had a process to improve text summarization technique. Text Summarization is implemented with NLP due to packages and methods in Python. Different approaches are present for summarizing the text and having few algorithms with which we can implement it. Text Rank is what to extractive text summarization and it is an unsupervised learning. Text Rank algorithm also uses undirected graphs, weighted graphs. keyword extraction, sentence extraction. So, in this paper, a model is made to get better result in text summarization with Genism library in NLP. This method improves the overall meaning of the phrase and the person reading it can understand in a better way.
High dimensional data analytics is emerging research field in this digital world. The gene expression microarray data, remote sensor data, medical data, image, video data are some of the examples of high dimensional data. Feature subset selection is challenging task for such data. To achieve diversity and accuracy with high dimensional data is important aspect of this research. To reduce time complexity parallel stepwise feature subset selection approach is adopted for feature subset selection in this paper. Our aim is to reduce time complexity and enhancing the classification accuracy with minimum number of selected feature subset. With this approach 88.18% average accuracy is achieved.
This examination work is focused around planning and simulating another kind of inset feed Disc Shaped Microstrip Patch Antenna (DSMPA) with Inset feed and Defected ground plane (DGP). By presenting a round space at the focal point of the ground plane, improved attributes of Microstrip patch antenna can be accomplished. The proposed Disc Shaped Microstrip patch antenna is reverberating at 5 GHz. Simulation has been finished by utilizing reenactment programming HFSS version15. From recreation results, it discovers that our examined Disc Shaped Microstrip patch antenna yields better return loss of - 25.1 dB & VSWR estimation of 0.96 dB. The examined DSMPA is yielding a higher radiation efficiency of 77.20 %. The minimized size and higher radiation efficiency contrasted with rectangular Microstrip patch antenna makes it all the more generally helpful for satellite communications.
Recommender systems play a vital role in e-commerce. It is a big source of a market that brings people from all over the world to a single place. It has become easy to access and reach the market while sitting anywhere. Recommender systems do a major role in the commerce mobility go smoothly easily as it is a software tool that helps in showing or recommending items based on user’s preferences by analyzing their taste. In this paper, we make a recommender system that would be specifically for music applications. Different people listen to different types of music, so we make note of their taste in music and suggest to them the next song based on their previous choice. This is achieved by using a popularity algorithm, classification, and collaborative filtering. Finally, we make a comparison of the built system for its effectiveness with different evaluation metrics.
To prevent the public from pandemic Covid’19 the government of India has started the vaccination from mid of January 2021. The government has approved the two vaccines, Covishield from the university of Oxford and Covaxin from Bharat Biotech.The vaccination started with frontline workers and is further extended to common public prioritizing the elders of above 60 years and people aged 45 years above with co morbidities. Though many people have got benefitted from it there is still a group of people not convinced with the vaccination. We have carried out this work to analyze those Indian people sentiments on the vaccines through the hash tags of tweets. The results show that though majority of the community has a positive belief on the vaccines but some of them still express negative emotions.
The income of the famers has decreased drastically over the past years as they do not have the proper channel for marketing their produce. This has also proved to be the factor that favors the landlords and money lenders to gain possession over their agricultural products at a very low cost and obtain a large profit from it. This also reflects the inability of farmers to obtain the righteous profit from their produce. The main aim of our project is to organizing and uniting the farmers under one umbrella, to reduce the unbalanced accumulation of the profit from perishable farm produces for the traders and the sellers and help maximise the income level of the farmers by self-marketing. This system has been implemented by considering the entire supply-demand eco system and it also helps avoid product wastage.
The security of any business plays a vital role. All enterprises expect high security because of the increase in robbery. It is challenging to manage security with traditional ways of protection. This paper emphasizes the sensor-based security system to protect against any unwanted entry in the business area. This system is developed using IoT-based sensors, and electronic materials develop the security system. The present scenario ensures protection and security have become inevitably necessary. There is regressive progress in the protection sector as the influence of new technology is hitting its height. It’s well-known as a modern home when there is a current home with minimal human effort. This technology aims to automate industrial area security and partially replace the security individual, enabling us to monitor unsuspecting activities and be warned during critical situations. Since wireless and emerging technology is taking place, an automated intelligent protection system is being introduced.
The internet is faced with many problems daily, one of them is decrement in network bandwidth because of Distributed Denial of Service (DDoS) attack on host server, which deplete host resources. Researchers has been invented many protection mechanisms such as detection, trace back, prevention, reaction, and characterization are in case of DDoS attacks, which will control the number of malicious packets received by the victim. But it does not provide efficient detection technique with high rate in real time network infrastructure. Thus, modern technologies are prepared on Mininet network simulators, which give more impact to simulate the real network. The architecture of Software Defined Networks (SDN) and OpenFlow architecture is used to demonstrate a programmable network model and centralized management of real network. In this research work, we provide design of software defined network (SDN) using mininet simulator and security issues related to the Software Defined Network.
In today’s growing cloud world, where users are continuously demanding a large number of services or resources at the same time, cloud providers aim to meet their needs while maintaining service quality, an ideal QoS-based resource provisioning is required. In the consideration of the quality-of-service parameters, it is essential to place a greater emphasis on the scalability attribute, which aids in the design of complex resource provisioning frameworks. This study aims to determine how much work is done in light of scalability as the most important QoS attribute. We first conducted a detailed survey on similar QoS-based resource provisioning proposed frameworks/techniques in this article, which discusses QoS parameters with increasingly growing cloud usage expectations. Second, this paper focuses on scalability as the main QOS characteristic, with types, issues, review questions and research gaps discussed in detail, revealing that less work has been performed thus far. We will try to address scalability and resource provisioning problems with our proposed advance scalable QoS-based resource provisioning framework by integrating new modules resource scheduler, load balancer, resource tracker, and cloud user budget tracker in the resource provisioning process. Cloud providers can easily achieve scalability of resources while performing resource provisioning by integrating the working specialty of these sub modules.
Paradigm shift towards cloud computing offers plethora of advantages both for cloud users and Cloud Service Provider (CSP). For cloud users, it offers saving of cost, scaling of resources, pay per use, elastic and on-demand services. On the other hand, it offers centralized resource management and provisioning of operations, safety and security for CSP. By holding multiple virtual IT resources (CPUs, storage servers, network components and software) over the internet, Infrastructure-as-a-Service (IaaS) serves as fundamental layer for all other delivery models. Along with benefits of IaaS, there exists several security and privacy issues and threats to confidentiality, integrity, authentication, access control and availability. In this paper, detailed study of IaaS components, associated security and privacy issues are explored and counter measures for the same are determined. Furthermore, as a result of the study, Model for IaaS Security and Privacy (MISP) is proposed. The model presents a cubical structure and adds more features than the existing models to enhance the security and privacy of data and operations and guide security assessment for safer adoption by enterprises.
The anti-counterfeiting traceability system based on blockchain technology can ensure the accuracy and consistency of the data stored by each participating node, protect the legitimacy of the data, ensure the product quality, improve the credibility of enterprises, and enhance consumers’ confidence in products. However, due to the low system throughput, high energy consumption and poor data availability, the combination of blockchain and traditional anti-counterfeiting traceability mode has many challenges, such as low efficiency. This paper aims to find an improved consensus mechanism based on contribution proof to improve the mining efficiency of honest miners. And plan to introduce a credit system, give priority to the high credit value of the mining block to package, improve the overall packaging efficiency of the system, to solve the problem of low feedback efficiency of the blockchain anti-counterfeiting system.
The modem data is collected by using IoT, stored in distributed cloud storage, and issued for data mining or training artificial intelligence. These new digital technologies integrate into the data middle platform have facilitated the progress of industry, promoted the fourth industrial revolution. And it also has caused challenges in security and privacy-preventing. The privacy data breach can happen in any phase of the Big-Data life cycle, and the Data Middle Platform also faces similar situations. How to make the privacy avoid leakage is exigency. The traditional privacy-preventing model is not enough, we need the help of Machine-Learning and the Blockchain. In this research, the researcher reviews the security and privacy-preventing in Big-Data, Machine Learning, Blockchain, and other related works at first. And then finding some gaps between the theory and the actual work. Based on these gaps, trying to create a suitable framework to guide the industry to protect their privacy when the organization contribute and operate their data middle platform. No only academicians, but also industry practitioners especially SMEs will get the benefit from this research.
In today’s world, the security of every individual has become an important aspect. There is a need for constant monitoring in public places. A Manual operating camera system is an unreliable and very basic and poor method for this purpose. Intelligent Video Surveillance is an approach where multiple CCTVs constantly record the scenes and proper algorithms are deployed in order to detect and monitor activities. Deep Learning frameworks and algorithms like Kera’s, YOLO, Convolutional Neural Networks or backbones for image detection like VGG16, Mobile net, Resnet101 have been used for human and weapon detection. The paper focuses on deep learning techniques and threading to collectively develop a Parallel Deep Learning Framework for Video Surveillance that aims at striking the right balance between accuracy and system performance or stability. Threading is used in terms of implementation of a uniquely proposed Dynamic Selection Algorithm that uses two backbones for object detection and switches between them based on the queue status for achieving system stability. A uniquely designed logistic regression filter is also implemented that boosts the system performance.
Gender classification is amongst the significant problems in the area of signal processing; previously, the problem was handled using different image classification methods, which mainly involve data extraction from a collection of images. Nevertheless, researchers over the globe have recently shown interest in gender classification using voiced features. The classification of gender goes beyond just the frequency and pitch of a human voice, according to a critical study of some of the human vocal attributes. Feature selection, which is from a technical point of view termed dimensionality reduction, is amongst the difficult problems encountered in machine learning. A similar obstacle is encountered when choosing gender particular features—which presents an analytical purpose in analyzing a human’s gender. This work will examine the effectiveness and importance of classification algorithms to the classification of gender via voice problems. Audial data, for example, pitch, frequency, etc., help in determining gender. Machine learning offers encouraging outcomes for classification problems in all domains. An area’s algorithms can be evaluated using performance metrics. This paper evaluates five different classification Algorithms of machine learning based on the classification of gender from audial data. The plan is to recognize gender using five different algorithms: Gradient Boosting, Decision Trees, Random Forest, Neural network, and Support Vector Machine. The major parameter in assessing any algorithm must be performance. Misclassifying rate ratio should not be more in classifying problems. In business markets, the location and gender of people are essentially related to AdSense. This research aims at comparing various machine learning algorithms in order to find the most suitable fitting for gender identification in audial data.
Pervasive computing has made life easy with communication devices. Today devise collaboration has enhanced everywhere in this environment. It has made computing devices invisible and the services. This pervasive framework provides applications with interactions, numerous cooperation and accessibility, and integration. The proposed work enumerates the applications, pervasive security challenges. It provides security predicaments by assigning certificate credentials, access controls, trust management, and some security techniques to overcome the security paradigms in these distributed networks with IoT and the pervasive computing framework. The work also encounters security perplexities in handling the security threats and user interaction issues. Nevertheless, security techniques are listed for various pervasive applications in distinct domains such as healthcare, industries, and transforming sensitive information. The smart applications with smart environments perhaps force towards the new technologies in the pervasive computing outlook. The work also embedded with middleware with the context-based situation in these pervasive applications
The Pre-owned cars or so-called used cars have capacious markets across the globe. Before acquiring a used car, the buyer should be able to decide whether the price affixed for the car is genuine. Several facets including mileage, year, model, make, run and many more are needed to be considered before getting a hold of any pre-owned car. Both the seller and the buyer should have a fair deal. This paper presents a system that has been implemented to predict a fair price for any pre-owned car. The system works well to anticipate the price of used cars for the Mumbai region. Ensemble techniques in machine learning namely Random Forest Algorithm, eXtreme Gradient Boost are deployed to develop models that can predict an appropriate price for the used cars. The techniques are compared so as to determine an optimal one. Both the methods provided comparable performance wherein eXtreme Boost outperformed the random forest algorithm. Root Mean Squared Error of random forest recorded 3.44 whereas eXtreme Boost displayed 0.53.