Ebook: Advances in Parallel Computing Technologies and Applications

This book presents the proceedings of the Virtual International Conference on Advances in Parallel Computing Technologies and Applications (ICAPTA 2021), held on the 15th and 16th of April 2021 in Chennai, India, at Justice Basheer Ahmed Sayeed College for Women, jointly with KAS Innovative India.

The aim of the conference is to provide a forum for sharing knowledge in various aspects of parallel computing in communications systems and networking, including cloud and virtualisation solutions, management technologies, and vertical application areas. The recent developments of parallel computing in various fields of application are used to handle the huge volume of data to provide solutions in a faster manner. Furthermore, the newer and innovative ideas are well groomed with adequate technical support in computer communication to enhance the decisions in intellectual aspects. It also provides a premier platform for scientists, researchers, practitioners and academics to present and discuss about the most recent innovations, trends and concerns as well as practical challenges encountered in this field.

ICAPTA 2021, has recorded more than 100 submissions of which 52 full length papers were accepted based on the reviews and comments from subject experts. The topics include parallel computing in communication, machine learning intelligence for parallel computing and parallel computing for software services in theoretical and practical aspects. The main program of the two-day conference included a chief guest address, two invited talks, two keynote talks and four technical sessions for paper presentations.

We hope that all participants have taken the opportunity not only to exchange their knowledge, experiences and ideas but also to make contacts for their ongoing research in new directions.

Finally, we would like to thank all the authors as well as all the committee members, reviewer sand session chairs for their support, enthusiasm and time, which helped to make ICAPTA 2021 a successful conference during this time of the pandemic. We express our sincere thanks to the directors of KAS Innovative India for co-organizing this event. We also wish to thank IOS Press for their support and tireless efforts in preparation for the publication of the conference proceedings.

The Editors

Networks have an important role in our modern life. In the network, Cyber security plays a crucial role in Internet security. An Intrusion Detection System (IDS) acts as a cyber security system which monitors and detects any security threats for software and hardware running on the network. There we have many existing IDS but still we face challenges in improving accuracy in detecting security vulnerabilities, not enough methods to reduce the level of alertness and detecting intrusion attacks. Many researchers have tried to solve the above problems by focusing on developing IDSs by machine learning methods. Machine learning methods can detect datas from past experience and differentiate normal and abnormal data. In our work, the Convolutional Neural Network(CNN) deep learning method was developed in solving the problem of identifying intrusion in a network. Using the UNSW NB15 public dataset we trained the CNN algorithm. The Dataset contains binary types of ‘0’ and ‘1’ in general for normal and attack datas. The experimental results showed that the proposed model achieves maximum accuracy in detection and we also performed evaluation metrics to analyze the performance of the CNN algorithm.

There is an increasing emphasis on enhancing the efficiency traffic management systems. Information is exchanged between the vehicular nodes to efficiently monitor and control huge volumes of vehicle. All existing applications in this area have focused on reliable data exchange and authentication process of vehicular nodes to forward messages. This study proposes a new entity centric trust framework using decision tree classification and artificial neural networks. Decision tree classification is used to derive rules for trust calculation and artificial neural networks are used to self-train the vehicular nodes, when expected value is not met. This model uses multifaceted role and distance based metrics like Euclidean distance to estimate the trust. The proposed entity centric trust model, uses a versatile new direct and recommended trust evaluation strategy to compute trust values. The suggested model is simple, reliable and efficient in comparison to the other popular entity centric trust models.

Tagging is a machine learning technique that provides tags to the information that the user can easily identify the related information. Manual tagging is widely used for constructing question banks; but, this approach is time-consuming and it would create pathway to consistency issues. Semi-manual tagging which is time-consuming and people must be experts in that domain have the ability to identify the question and tag them it is not possible in real-time and high in cost. The proposed associate degree automatic tagging exploitation information processing that mechanically tags automatic question tagging. In this paper, step up with a keywords-based model to automatically tag questions with information units. With regard to multiple-choice questions, the proposed models use mechanisms to capture helpful information from keywords to boost tagging performance. Automatic tagging method using NLP automatically tag questions where users can get information regarding to his search which overcomes earlier methods. In our experiments, the result shows that the model is credible and outperforms various existing models.

The impact of Artificial Intelligence in the domain of Healthcare has been growing, day by day. These applications bring a drastic change in the healthcare system and affects our lives based in the change it brings to the Patientcare system, transforming the traditional way of handling sicknesses and diseases. Machine Learning algorithms that use data, have a big role in the AI based applications that are used in the Healthcare. Hence the Data source and the nature of Data holds an important role in developing effective AI based solutions for many health issues in the society. Data is available in all the hospitals and medical care facilities for many years now. However, without transforming them into a format where Machine Learning algorithms work, it is impossible to use them to develop an AI based application. In this research paper, we briefly discuss the process of developing an AI based application to predict Malaria, which is one of the most common vector borne diseases in the coastal districts of Karnataka. This pioneer work was done over the data collected from the clinical notes of a 1500 bed hospital situated in Mangalore. Few machine learning algorithms like Logistic regression, Support vector machine XGB Booster classifier, CAT Booster Classifier and Random forest classifier were used over the dataset. Our experimental study revealed that, Random Forest classifier works efficiently for this data set, compared with the other algorithms that we used. It gave the best accuracy of 90.92.

Diabetic retinopathy is one of the leading reason for preventable blindness in the world. 10–18 % of diabetic people having diabetic retinopathy. The feature selection and classification is a vital task to find the seriousness of the diabetic retinopathy. The different researchers introduced different techniques to extract the features and classification of diabetic retinopathy images. The deep learning is one of the essential methods to extract the features. Most of the previous techniques are extracted information’s with the help of texture and extracted the whole image feature data. Some feature missed and thereby the accuracy is significantly less. Hence a proposed new technique called FRCNN (Fast Region-based Convolution Network) and Nearest Neighbour (NN) algorithm used to extract the features and classifications. The proposed method yields better accuracy (96%), sensitivity (98%) and specificity (97%) compared to the previous methods. The implementations Messidor Dataset is used for training and testing

The main aim of our project is to implement the door lock system using the fingerprint sensor. There are many modern locks that replace the regular locks due to the high security, while we use this we are able to both lock (or) unlock the door lock using the fingerprint sensor. You are not supposed to carry the bundle of keys wherever you go, so that we can implement this idea and for the high security too. The possibilities of threats occurring in regular locks is less compared to the wireless biometric lock (i.e.) modern lock. Only a particular person can open the door lock (or) unlock it. It is not like the regular lock means if X person opening the lock using a key means the lock will release on its own. In modern lock there is no possibility that only the developer will open it, if someone wants to access using smartphone we should get verification from the user by getting the OTP and login the app.

The power is generated by human motion while walking on the piezoelectric sensor, which is pressed and produces kinetic energy, which is then converted into electrical energy. The generated energy is stored in the battery. The energy in the battery is used to turn on the street lights using the LDR Sensor when the sun’s beam becomes dull, and to pass water to the grass using the motor with the help of the soil moisture Sensor when the soil becomes moisture. And also used for charge the mobile phones using the charging port which is installed in the park and to be used for other purposes in the park. All the data is get tracked and stored in the IOT for continuously monitoring and for future purpose.

The Internet of Things (IoT) refers to a network of physical objects that are embedded with sensors are able to collect and transfer information over a wireless network with human conciliation. In confined Home Patient Observation system light-weight wearable/embedded sensing element devices connected to record the patient health conditions. Adopting looming technologies like 5G and Mobile Edge Computing (MEC) a multi-staged journey builds ultra-low latency. The health care sector is drowning in information. An overflow of clinical records, patient history, complicated billing information, medical analysis and lot builds it very troublesome to manage such information at intervals the health care sector in a systematized approach. Escorted by IoT, since data mobility occurs, hackers gain access to clinical IoT device, acquire management and revamp data which oblige health partitioners take actions which will injury the health of their client. i) To overthrow security issues introduces Optimized blockchain with hyperledger fabric in edge-cloud computing is proposed. Since data stored in blockchain are immutable, the permission granted patients can only access their records using hash value. Thus, blockchains will provide an immutable audit trail of health facts. ii) Software Defined Network (SDN) deliver Quality of Service (QoS) steer further from network congestion.

Now a day’s recommendation system has changed the fashion of looking the items of our interest. OTT Movie Application Recommendation for mobile users is crucial. It performs a complete aggregation of user preferences, reviews and emotions to help you make suitable movies. It needs every precision and timeliness, however,this can be info filtering approach that’s accustomed predict the preference of that user. Recommender System may be a system that seeks to predict or filter preferences in keeping with the user’s selections. The very common purpose where recommender system is applied are OTT platforms, search engines, articles, music, videos etc During this work we tend to propose a Collaborative approach-based Movie Recommendation system. it is supported collaborative filtering approach that creates use of the knowledge provided by users, analyzes them so recommends the flicks that’s best suited to the user at that point. The suggested motion picture list is sorted in keeping with the ratings given to those movies by previous users. It conjointly helps users to search out of their selections supported the movie expertise of alternative users in economical and effective manner while not wasting a lot of time in useless browsing [1]. Therefore, we tend to offer the item-oriented methodology of the analysis of social network as the steering force of this method to further improve accuracy within the recommendation system. We tend to propose economic healthcare associates during this paper The algorithmic rule of the Film Recommendation supported improved KNN strategy that measures simpler advisory system accuracy. However, to evaluate performance, the k closest victimized neighbors, the maximum inner circles, as well as the basic inner strategies are used [2]. The exception to this is the projected results, which use algorithms to check for (supposedly) involvement.The performance results show that the projected strategies improve additional accuracy of the Movie recommendation system than the other strategies employed in this experiment.

A system and method for facilitating a visually impaired person for identifying a person. The method includes the step of storing a plurality of instructions for facilitating the visually impaired person identify the person in front of them by their face and/or voice characteristics by updating our project with Mask Detection using OpenCv and keras. It includes the step of receiving voice signals from the person present in surrounding of the visually impaired person and includes the step of capturing the pictures of a particular person and their surroundings of the visually impaired person and storing the processed data into the database or any storage devices. The data will be processed to AWS server or any local storage for processing and determining the person with the help of the database we already have. After processing and identifying the person with the help of face and voice recognition modules the name is sent to the visually impaired user’s phone in the form of a text message which will be read aloud by his phone’s virtual assistant.

IEEE 802 is used in LAN networks that expose or provide sensitive data to complex applications or services. These are protocols for accessing, managing and controlling access to network-based services and applications in general. Port-controlled network access controls network access and prevents the transmission and reception of nameless or unauthorized persons, leading to network interruption, service theft and data loss. This paper introduces a new approach to investigate whether a data packets in wired networks transferred to a management device is authenticated packet. The data packets are sent to the SDN from RAR and share the information associated with each packet with a limited rate for the access management and are received by the RFC. Here it detects whether the data packet arrived is accepted or restricted. The speed at the authentication start packet is restricted to manage the number of terminals that enter later authentication, and it avoids avalanche impact of wireless authentication which may cause faults to lots of terminals which enter later authentication at the same time.

The world has seen a substantial growth in the number of senior citizens who often endure from heart disease in recent times and has seen a major spread of the viral disease (COVID-19) in recent months, resulting in thousands of deaths, particularly among the elderly and people suffering from heart disease. Among the most promising health care that seek to relieve the suffering of patients, particularly the elderly, by eliminating the trouble of going to hospital centers for treatment and enabling them to obtain medical attention in their homes. The IoT has been depended on and its use has seen broad adoption in health care, where it is commonly used remotely to track patient health. Fog computing expands the computational capabilities of the cloud to the edge devices of the IoT, enabling many smart devices in healthcare to provide services such as storing and retrieving information to their users. However in the traditional cloud-based system, there is a timing delay in providing a reliable and secure heart monitoring system. The main objective of this work is to develop a fog-enabled cloud-based (FECB) heart rate monitoring unit that allows better network use and energy consumption. In terms of performance metrics, such as latency and delay, the proposed system also outperforms the existing system.

Worldwide human health and economic has been affected due to the ongoing pandemic of corona virus (COVID-19). The major COVID-19 challenges are prevention, monitoring and FDA approved vaccines. IOT and cloud computing play vital role in epidemic prevention and blocking COVID-19 transmission. Mostly lungs and hearts are affected. Other than lungs many parts are affected which are not considered as prominent conversational cue. In this paper, we have proposed smart system that is effective through detection of pancreas, kidney and intestine. It detects acute pancreatitis, protein leak, microscopic blood leak, post infectious dysmotility and gastrointestinal bleeding. The data from the edge devices are collected and mapped into the cloud layer. The cloud consists of COVID-19 patients medical records which compare the user data with the existing patient records. Once the data matches it sends warning message to the user regarding the result of affected parts. Based on the result from KPI system, it analyzes with all data and using deep Convolutional Neural Network (CNN) it classifies whether the pancreas, kidney and intestine are affected or not due to COVID-19.

Vehicular ad-hoc networks (VANET) are listed as an extension of mobile ad-hoc networks (MANETs) which can improve road safety and provide the Intelligent Transportation Systems (ITS). In addition to its advantages, VANET faces many obstacles ranging from high-energy consumption to instability induced by high changes in topology. The main goal of designing an optimum route algorithm is to decrease the probability of contact failure and reduce the energy consumption of nodes within the network. Clustering is thus a method to combine nodes and make the network more robust. With no node consciousness, it is often shorter on resources, which causes network execution problems and changes in topology. At that point, a primary energy issue emerges in the AODV routing protocol that aims to improve the energy efficiency of the V2V communication in nodes lifetime and to connection lifetime problems in the network. This article proposed a clustering-based optimization technique called Energy Efficient Clustering Technique (EECT) with the AODV protocol’s K-Medoids clustering algorithm in order to cluster vehicle nodes and find nodes that are convincing to interact in a defined secured and reliable path, which detailed in previous works. Efficient nodes are recognized from each cluster with the goal of energy-efficient communication, to optimize the parameter as minimum energy consumption in VANET.

A packet has to be delivered within a distinct time limit as data delivery is a critical issue in a wireless network. In emergency situations, real-time data distribution of multimedia file is addressed using wireless network. The real-time data examined here are image and audio files and these files are broken into sequence of packets and forwarded to the destination peer node based on a Priority algorithm. Prioritized data dissemination processes the sequence of packets to be forwarded based on permanent priority scheduling. The packets are encrypted based on Trust-based Public key management using public key generated in key generation phase and decrypted at the receiver end. The simulation results prove that the proposed technique has the enhanced and secured data transmission. The design of the network requirements and detailed experimental results are presented.

The handwritten Multilanguage phase is the preprocessing phase that improves the image quality for better identification in the system. The main goals of preprocessing are diodes, noise suppression and line cancellation. After word processing, various attribute extraction techniques are used to process attribute properties for the identification process. Smoothing plays an important role in character recognition. The partitioning process in the word distribution strategy can be divided into global and local texts. The writer does not use this header line to write the text which creates a problem for skew correction, classification and recognition. The dataset used are HWSC and TST1. The tensor flow method is used to estimate the consistency of confusion matrix for the enhancement of the text recognition .The accuracy of the proposed method is 98%.

The impact of Covid-19 outbreak has become a matter of grave concern in India. The scarcity of resources to endure the epidemic outbreak combined with the fear of overburdened health care systems had forced many Indian states to go for a partial or complete lockdown. Stimulated by the need for employing Artificial Intelligence (AI) in battling Covid-19 crisis, this paper highlights the importance of AI in responding to the Covid-19 outbreak and curtail the severity of the disease. Artificial Intelligence (AI) aim is to open the door for human in the prediction of the novel Corona Virus disease. Researchers are in process of using AI in finding new drugs and medicines for Covid cure and focusing on detection and analysis of infectious patients, through AI based Machine learning (ML) Algorithms and Expert Systems. Owing to huge number of cases, Covid-19 remains a global health problem, Al could take individual conditions into account, produce suitable decisions and promise to make great strides in managing Corona Virus. We begin with an outline of AI, identify the various applications aimed at fighting the disease, then highlight the challenges and issues associated with the cutting edge solutions finally arriving on the recommendations for the communication to effectively control the Covid-19 crisis.

Android is a widely distributed mobile operating system developed especially for mobile devices with touch screens. It is an open source, Google-distributed Linux-based mobile operating system. Since Android is open source, it enables Android devices to be targeted effectively by malware developers. Third-party markets do not search for malicious applications in their databases, so installing Android Application Packages (APKs) from these uncontrolled market places is often risky. Without user’s notice, these malware infected applications gain access to private user data, send text messages that costs the user, or hide malware apk file inside another application. The total number of new samples of Android malware amounted to 482,579 per month as of March 2020. In this paper deep learning approach that focuses on malware detection in android apps to protect data on user devices. We use different static features that are present in an Android application for the implementation of the proposed system. The system extracts various static features and gives them to the classifier for deep learning and shows the results. This proposed system will assist users in checking applications that are not downloaded from the official market.

The rapid rate of innovations and dynamics of technology has made humans life more dependent on them. In today’s synopsis Microblogging and Social networking sites like Twitter, Facebook are a part of our lives that cannot be detached from anyone. Through these social media each one of them carry their emotions and fix their opinions based on a particular situations or circumstances. This paper presents a brief comparison about Detection and Classification of Emotions on Social Media using SVM and Näıve Bayesian classifier. Twitter messages has been used as input dataset because they contain a broad, varied, and freely accessible set of emotions. The approach uses hash-tags as labels to train supervised classifiers to detect multiple classes of emotion on potentially large data sets without the need for manual intervention. We look into the usefulness of a number of features for detecting emotions, including unigrams, unigram symbol, negations and punctuations using SVM and Näıve Bayesian Classifiers.

Lyme disease appears by various means one of the causes is infection from a bite of a black-legged tick which leads to the formation of a rash called Erythema Migrans (EM). This Lyme disease is the direct root of skin cancer and the primary phase is known as Basal Cell Carcinoma (BCC). Skin cancer causes pathological situations it integument the body’s surfaces, including skin, hair, nails, and associated glands. Identifying EM and BCC in biomedical representation is a common platform known as biopsies and many works are done in the traditional methodology. But early detection of EM and BCC in the field of medical imaging using computer-aided diagnosis provides more accuracy and rapidity. This survey is classified under two categories; the first is EM detection and the second is the segmentation of BCC. The circumstances of this review are under the base of different algorithms, various methods used by the non-medicals appliance are discussed and compared by the several measures taken by increasing the parameters to improve the accuracy levels.

Sign language is a terminology that encloses a motion of hand gestures which is an environment for the auditory impairment, individual (deaf or dumb) to deal with others. Nevertheless, so as to impart with the hearing impaired individual, the communicator obtains to acquire acquaintance in sign language. As follows is frequent to make undoubted that the message provided by the hearing impaired person acknowledged. This implemented system propounds an implementation of real time American Sign Language perception in Convolutional Neural Network (CNN) with the support of You Only Look Once version (YOLO) algorithm. The algorithm initially executes data acquisition, subsequently the pre-processing of gestures and are conducted to trace hand movement utilize a combinational algorithm.

Performance analysis of learning outcomes is a system that will aim for excellence at all levels and dimensions of the student’s field of interest. This paper suggests a comprehensive EDM framework in the form of a rule-based recommender system that not only analyses and predicts student achievement, but also demonstrates the reasons for it. The suggested framework examines students’ demographic information, study-related characteristics, and psychological factors to gather as much information as possible from classmates, teachers, and parents. School reports and queries used to collect the world’s latest data (e.g., student marks, population, social and school-related factors. Using a set of potent data mining methods, aiming for the greatest possible precision in academic performance prediction. The framework is effective in identifying the student’s weaknesses and making relevant recommendations. In contrast to current frameworks, the proposed framework outperforms them in a practical case study involving 200 individuals.

The most demanded advanced technology throughout the world is cloud computing. It is one of the most significant topics whose application is being researched in today’s time. Cloud storage is one of the eminent services offered in cloud computing. Data is stored on multiple third-party servers, rather than on the dedicated server used in traditional networked data storage in the cloud storage. All data stored on multiple third-party servers is not bothered by the user and no one knows where exactly data saved. It is minded by the cloud storage provider that claims that they can protect the data but no one believes them. Data stored over the cloud and flowing through the network in the plain text format is a security threat. This paper proposes a method that allows users to store and access the data securely from cloud storage. This method ensures the security and privacy of data stored on the cloud. A further advantage of this method is we will be using encryption techniques to encrypt.

The revolution in communication and embedded systems Electronic Toll Collection, the new era of intelligent transportation systems enables the automatic collection of Toll fees from the prepaid account through RFID. Yet there is lack of security in the existing system, number plate detection and recognition can be made. In the existing system detection is made with condition random field algorithm. Recognition method is implemented by optical character recognition. A system can be designed to extract the number plate from vehicle automatically using image processing technique, match with database automatically and generate the One Time Password(OTP) and bill without any delay and identify the theft vehicles. This is done by using Image processing and motion capturing technology