International Journal of Advanced Engineering Science and Information Technology 2022-07-08T00:49:27+00:00 Editor Open Journal Systems <p><em>NOTE: DEAR RESEARCHER, </em></p> <p><em>IJAESIT ( ISSN 2349-3216) was registered in the year <a href="">2014</a>. It was publishing articles in the domain IJAESIT.ORG, and recently we moved to this new domain. The old archives are being slowly moved to this domain. For any queries contact editor.<a href=""></a></em></p> <p><em>IJAESIT </em>is an inclusive journal community working together to advance science for the benefit of society, now and in the future. Founded with the aim of accelerating the pace of scientific advancement and demonstrating its value, we believe all rigorous science needs to be published and discoverable, widely disseminated and freely accessible to all.</p> <p>The research we publish is multidisciplinary and, often, interdisciplinary. IJAESIT<em> </em>accepts research in over two hundred subject areas across science, engineering, medicine, and the related social sciences and humanities. We evaluate submitted manuscripts on the basis of methodological rigor and high ethical standards, regardless of perceived novelty</p> TEA HARVESTER WITH ISOLATED CHARGING STATION 2022-06-29T13:10:55+00:00 Mr.Syril Deepak A , Mr.Umar Muktar S <p><strong><em>Solar energy is one of the main renewable energy sources that can be used to efficiently charge a battery. The purpose of this system was to design and develop a proof-of-concept Solar powered tea harvester with isolated charging station, which incorporates the fuzzy logic algorithm. A custom maximum power point tracker (MPPT) was designed to extract the maximum amount of power available from the solar panels. This MPPT sampled the voltage and current output of the solar panels and executed the Fuzzy algorithm to determine the maximum power point. In this system consists of Solar panel, Buck-Boost Converter, Isolated Transformer, rectifier, battery. This system is works by DC battery can be charged up via an isolated charging station, charging station of solar panel produce power from solar energy which is fed to buck boost converter to increase or decrease the voltage based on solar energy it generates Pulse based on Fuzzy logic Algorithm. Then the converter is converts to DC-to-DC power Which ensure appropriate voltage to charge the battery, based on MPPT fast charging of loaded battery can be done properly. This model is developed by using MATLAB Simulink program.</em></strong></p> 2022-06-28T00:00:00+00:00 Copyright (c) 2022 PREDICTING THE STUDENTS GPA USING MACHINE LEARNING TECHNIQUES 2022-06-29T13:15:25+00:00 Ms.T.Jayapratha Priya Mahalakshmi S Rani Isvarya N Shameli K Vijatha S <h2>---- Analysing degree marks have been a significant area of examination in schooling because of the longing to recognize the basic factors that impact scholarly execution. In light of restricted achievement in foreseeing the grade point normal (GPA), a large portion of the earlier examination has zeroed in on anticipating grades in a particular arrangement of classes in view of understudies' earlier exhibitions. The issues related with information driven models of GPA forecast are additionally intensified by a little example size and a moderately enormous dimensionality of perceptions in a trial. In this paper, we use the cutting-edge AI procedures to build and approve a prescient model of GPA exclusively founded on a bunch of self-administrative not entirely settled in a moderately little example analyze.It examines instructive ramifications of arising innovations on the manner in which understudies learn and foundations educate and advance. Late innovative progressions and the speeding up taking on new advancements in advanced education are investigated to anticipate the future idea of advanced education in this present reality where man-made consciousness is important for the texture of our colleges.</h2> 2022-06-29T00:00:00+00:00 Copyright (c) 2022 IOT BASED SMART ARTIFICIAL VENTILATOR FOR COVID-19 PATIENTS 2022-06-29T13:19:15+00:00 S.Gladson, B.Abarna,E.Elackya, K.Hemalatha, J.Jeyalakshmi, <p><strong><em>The crisis resulting from the COVID19 pandemic has generated an adverse situation in which thousands of people dies due to lack of artificial ventilation devices.The preliminary design of a simple, easy-to-use, and easy-to-build ventilator with an unique design that can be used for COVID-19 patients in emergencies and to prevent massive loss of life in resource-poor environments.The last but now the least is the setting to adjust the time duration for inhalation to exhalation ratio. The ventilator we here design and develop using Arduino encompasses all these requirements to develop a reliable yet affordable ventilator to help in times of pandemic. We here use a silicon ventilator bag coupled driven by servo motor with one side push mechanism to push the ventilator bag. Our system makes use of blood oxygen sensor along with sensitive heart Beat sensor to monitor the necessary vitals of the patient and display on a webpage using IOT. To adjust the time duration for inhalation the option command given in the IOT application to set. The entire system is driven by Arduino controller to achieve desired results and to assist patients in COVID pandemic and other emergency situations.</em></strong></p> 2022-06-28T00:00:00+00:00 Copyright (c) 2022 AUDIO AND BIT STREAM ENHANCED CRYPTOGRAPHIC DATA SECURITY 2022-06-29T13:23:10+00:00 Nithya.T,Nishanthini.K.G, Sasikala.M, Soorya.S, Suriyaa.V.K <p><strong><em>JPEG steganography schemes take the effects of embedding in the spatial domain tend to exhibit higher security and introduce less artifacts that can be captured by the prevalent steganalyzers ensuing the paradigm, this work proposes a new plan of the distortion measure for JPEG steganography by incorporating the statistics of both the spatial and DCT domains. The spatial statistics of the decompressed JPEG images are firstly well characterized with distortion measures of some efficient steganography schemes in the spatial domain and the resulting embedding entropies of spatial blocks in arrangement with DCT blocks be then transformed into the DCT domain to obtain the distortion measures for JPEG steganography. Investigational cost demonstrate that the normal system outflanks impressively other best in class JPEG steganography schemes and UERD, for the most effective feature set GFR at present, and rivals them used for other characteristic sets.</em></strong></p> 2022-06-29T00:00:00+00:00 Copyright (c) 2022 SHILLING ATTACKS DETECTION BASED ON SPAMMER BEHAVIOURAL METHODS USING MACHINE LEARNING 2022-06-29T13:27:51+00:00 K.Pavithra, K.Puvarshini Nanthithaa, M.Sanjith Gunal, S.Vasanthasena×Prema P× <p><strong><em>The online reviews can change decision of the customer and they can finalize a product by comparing with the brands of products, the customer can select the product and satisfy their requirement only if the reviews are not fake. On the contrary, if the reviews are fake then it misleads the customer. To solve this problem the identification of the fake opinions from the customer needs to be extracted. Users behaviors are extracted based the semantical analysis of his review content for the purpose of identifying the review as fake or not. The reviews are extracted from the web for a particular product, along with the reviews of several other information related to the reviewers also been extracted to identify the fake reviewers using k-means clustering classifier and Information Gain. Signification of the features on the decision is validated using information gain. Experiments are conducted on exhaustive set of reviews extracted from the web and demonstrated the efficiency of the proposed approach.</em></strong></p> 2022-06-28T00:00:00+00:00 Copyright (c) 2022 EFFICIENT CLASSIFICATION OF BRAIN TUMOR IMAGES USING NEURAL NETWORK TECHNIQUE 2022-06-29T13:34:34+00:00 Ms.T.Kokilavani, Dr. K. Ganesh Kumar, N.Kiruthika, G.Mathumitha, W.Merlin Sweatha, S.Monisha <p><strong>Using biopsy, the classification of brain tumors is performed, which isn't usually conducted before definitive surgical process. the development of technology and machine learning helps radiologists in diagnostics of tumor without invasive measures. convolutional neural network (CNN) could be a machine-learning algorithm that has achieved substantial leads to image segmentation and classification. a number of the first brain tumors are gliomas, meningiomas, and pituitary tumors. Gliomas are a general term for tumors that arise from brain tissues apart from nerve cells and blood vessels. On the opposite hand, meningiomas arise from the membranes that cover the brain and surround the central systemanervosum, whereas pituitary tumors are lumps that sit inside the skull. the foremost important difference between these three forms of tumors is that meningiomas are typically benign, and gliomas are most ordinarily malignant. This project presents a replacement CNN architecture for brain tumour classification of tumor types. With good generalization capability and good fastness, the new developed CNN architecture may well be used as an efficient decision-support tool for radiologists in medical diagnostics. Python 3.7 is employed to develop the project</strong></p> <p><strong><em>&nbsp;</em></strong></p> 2022-06-29T00:00:00+00:00 Copyright (c) 2022 IMPLEMENTATION OF BLOCKCHAIN FOR AGRICULTURAL PRODUCTS TRANSACTION USING SHA – 256 2022-06-29T13:37:52+00:00 Gopalakrishnan K, S.NithishKumar,S.Pranesh, S.Suvetha, S.Thanigai Singaravelan <p><strong><em>: Blockchain is an arising computerized innovation permitting pervasive monetary exchanges among conveyed untrusted parties, without the need of middle people like banks. This article analyzes the effect of blockchain innovation in farming and food production network, presents existing continuous tasks and drives, and examines generally suggestions, difficulties and potential, with a basic view over the development of these ventures. Our discoveries show that blockchain is a promising innovation towards a straightforward store network of food, with numerous continuous drives in different food items and food-related issues, yet numerous hindrances challenges actually exist, which frustrate its more extensive prevalence among ranchers and frameworks. These difficulties include specialized angles, training, arrangements and administrative systems.</em></strong></p> 2022-06-28T00:00:00+00:00 Copyright (c) 2022 IMPLEMENTATION OFIMPROVED PREDICTIVE MODEL FOR STUDENTS GRADE PREDICTION USING RANDOM FOREST ALGORITHM 2022-06-29T13:39:59+00:00 Dr. V. K. Manavalasundaram, C. P. Nithish Kumar, K. V. S. Prabakaran, K. Ranganathan, D. Santhosh <p><strong><em>Internet learning stages like Massive Open Online Course (MOOC), Virtual Learning Environments (VLEs), and Learning Management Systems (LMS) work with thousands or even huge number of understudies to get the hang of as indicated by their inclinations without spatial and fleeting imperatives. Other than many benefits, web based learning stages face a few difficulties like understudies' indifference, high dropouts, low commitment, understudies' self managed conduct, and convincing understudies to take more time for settings their own objectives. In this review, we propose a prescient model that examines the issues looked by in danger understudies, therefore, working with educators for ideal intercession to convince understudies to expand their review commitment and further develop their review execution. The prescient model is prepared and tried utilizing different AI (ML) and profound learning (DL) calculations to describe the learning conduct of understudies as indicated by their review factors. The exhibition of different ML calculations is looked at by utilizing exactness, accuracy, backing, and f-score. The ML calculation that gives the best outcome regarding exactness, accuracy, review, backing, and f-score metric is eventually chosen for making the prescient model at various rates obviously length. The prescient model can help teachers in distinguishing in danger understudies right off the bat in the course for opportune mediation in this manner staying away from understudy dropouts. Our outcomes showed that understudies' appraisal scores, commitment power for example clickstream information, and time-subordinate factors are significant variables in internet learning.</em></strong></p> 2022-06-28T00:00:00+00:00 Copyright (c) 2022 IMPROVED LOAN PREDICTION SYSTEM USING MACHINE LEARNING 2022-06-29T13:42:55+00:00 Dr. S. Rajalakshmi, . S. Nakulan, D. Prasath, S. SathiyaMoorthi, V.Roja <p><strong><em>The aim of this project is to create a ensemble machine learning model that can determine whether a user is eligible for a loan. It is calculated using the user's personal data from the dataset such as education, number of dependents, income, loan amount,marital status, credit history, and work history. A data set of 600 instances is employed, with 70% of them being approved and the rest being refused. This&nbsp;ensemble algorithm helps in determining the applicant's ability to repay his or her loan. The prediction model benefits both the applicant and the bank by lowering risk and lowering the number of defaulters. The study will begin with exploratory data analysis, followed by preprocessing and testing in a developed hybrid machine learning model.</em></strong></p> 2022-06-28T00:00:00+00:00 Copyright (c) 2022 IDENTIFICATION OF BEARING FAULT IN ROBOTIC PRECEPT USING SFFS 2022-06-29T13:45:45+00:00 Dr. R. Menaka, B.P. Naveenya, S. Snekha, M. Yamuna <p><strong><em>The ultimate goal of the project is to find out the faulty bearing earlier by using machine learning. The faulty bearing also affects the other parts of machine such as high contact friction leads to heat results in fire. The wear and torn of faulty bearing cause the machine to consume excessive power due to load and affects the product quality. Although the depreciation of bearing affects the timing mechanism in a machine. The data for training, testing development of model are taken from Kaggle. The dataset from the Kaggle is pre- processed to remove noisy and to fill missing data by mean, median of the columns. After the pre- processing the data is split for training and testing. Initially the prediction is done by supervised learning using Support Vector Machine and compared with unsupervised learning by using Artificial Neural Network to achieve more accuracy.</em></strong></p> 2022-06-28T00:00:00+00:00 Copyright (c) 2022 DRIVER DROWSINESS DETECTION USING CNN 2022-06-29T13:50:51+00:00 Dr.S.Viveka, Jeevitha S, Keerthana B, Lakshmi L, Monika M <p><strong><em>Drowsy driving is a prevalent and a serious public health issue that deserves attention. Recent studies estimate that around 20% of car crashes have been caused by drowsy drivers. A person when he or she does not have a proper rest especially a driver, tends to fall asleep causing a traffic accident. Nowadays, one of the main goals in the development of new advanced driver assistance systems is trustworthy drowsiness detection. It is why the present project wants to realize a system that can detect the drowsiness of the driver, in order to reduce traffic accidents. For that system, it will take the processing of images through a camera which will focus on the driver. In that, it is going to analyze the changes that happen in the face and then will be processed through a program in order to detect drowsiness to send an alert to the driver. These methods are used in a wide range of applications, the image classification being one of the fields in which these are employed with success. After the face is detected using CNN, the region containing the eyes and mouth has to be separated. An image which taken inside a vehicle includes the driver’s face. </em></strong></p> 2022-06-28T00:00:00+00:00 Copyright (c) 2022 FACE RECOGNITION ATTENENCE SYSTEM USING DEEP LEARNING 2022-06-29T13:53:58+00:00 1S. Vasuki, 2A. Ashok Kumar, B.Bogeshwar, M.Pranod, T.Raveendra <p><em>In state-of-the-art educational machine, to Maintain the student`s attendance record with each day activities is a tough paintings for faculty. The pick out of student called thru college this takes time taking and misplacement of attendance which ends up in proxy attendance. Marking attendance manually isn`t completely time ingesting but more over it ends in unsecure, unreliable and might moreover be manual attendance out of place due a range of reasons like losingrecordsandsoon.TodecorateManualattendancemachinewiththeresourceoftheusage of smart attendance tool plays widespread characteristic to gain manual tool disadvantages. In Now-a-days smart attendance gadget plays crucial role for taking attendance for maintain student`s attendance files in a university database which further utilized in assessment performance.Theeverydayattendanceofuniversitycollegestudentsisrecordedperiodsensiblehatisstoredalreadybywayoftheuniversityadministratorteamandadditionallycollegehaving a duplicate attendance records. The above scenario will takes area on the time corresponding subject`scollegearrivesandloggedintotheirtoolandroboticallystartsoffevolvedtakingsnaps the usage of their identified database to submit accurate and proper attendance will put up withinsidetheuniversityportal.Thedetectingmachineisadvancedthruthemixingofubiquitous factors to make portable machine for taking snaps of college students. It could be coping with and tracking the university college students attendance files using the technological understanding like Face Recognition that is designed in a shape of software program software for a hardware machine.</em></p> 2022-06-28T00:00:00+00:00 Copyright (c) 2022 A Secure Method of File Sharing in Multi cloud Environment 2022-07-07T14:05:19+00:00 S.Naveen , J P Sabeela, S R Suryaa, B Velavan Mr. A .Logeswaran <div class="page" title="Page 1"> <div class="layoutArea"> <div class="column"> <p>Provable Data Possession (PDP) empowers cloud clients to confirm the information honesty without recovering the whole record. all the current PDP plans depend on the Public Key Infrastructure (PKI). The conspire is effective, adaptable and upholds private confirmation, designated check and public verification.ID-DPDP is imperfect since it neglects to accomplish soundness. Fix the blemish by introducing a nonexclusive construction. A new ID-DPDP convention is gotten by stretching out the fundamental ID-PDP to multiple cloud environments. With information capacity and sharing administrations in the cloud, clients can without much of a stretch change and offer information collectively. To guarantee shared information respectability can be checked openly, clients in the gathering need to figure marks on every one of the parts in shared information. Various parts in shared information are by and large endorsed by various clients because of information adjustments performed by various clients. For the sake of security, when a client is disavowed from the gathering, the parts which were recently endorsed by this denied client should be re- marks on every one of the parts in shared information. Various parts in shared information are by and large endorsed by various clients because of information adjustments performed by various clients. For the sake of security, when a client is disavowed from the gathering, the parts which were recently endorsed by this denied client should be re-endorsed by a current client. The clear technique, which permits a current client to download the relating part of shared information and once again sign it during client denial, is wasteful because of the enormous size of shared information in the cloud. In this work, we propose an original public inspecting instrument for the respectability of imparted information to proficient client repudiation at the top of the priority list. By using the possibility of intermediary re-marks, we permit the cloud to re-sign parts for the benefit of existing clients during client renouncement, so that endorsed by the cloud. In addition, our system can uphold group evaluating by confirming numerous reviewing undertakings all the while. Exploratory outcomes demonstrate the way that our component can altogether work on the productivity of client disavowal.</p> </div> </div> </div> 2022-06-30T00:00:00+00:00 Copyright (c) 2022 ARMA BASED CROP YIELD PREDICTION USING TEMPERATURE AND RAINFALL PARAMETERS WITH GROUND WATER LEVEL CLASSIFICATION 2022-07-08T00:41:07+00:00 Dr.V.K.MANAVALASUNDARAM, Dr. K. Ganesh Kumar, M.AMRISH, V.BRINDHA, D.JAYAPRAKSH <div class="page" title="Page 1"> <div class="layoutArea"> <div class="column"> <p>Now a days, wireless telecommunication networks are promising alternative for rainfall measuring instruments that complement previous monitoring devices. Because of big dataset of the rainfall and therefore the telecommunication networks data, empirical computational methods mean less adequate of actual data. So, deep learning models are proposed for the analysis of massive data and provides more accurate presentation of real measurements. This project performrainfall monitoring results from experimental measurements. The most aim of this study is supply a technique for rainfall data classification supported neural network methods supported historical rainfall data production data. Classification based on the previous years of rainfall can help farmers take necessary steps to live crop production within the coming season. Understanding and assessing future crop production can help ensure food security and reduce impacts of global climate change. During this work, ARMA (Auto Regressive Moving Average) method is used for proposed work. Past 10 years of information(data) set is taken for rainfall and ground water level for our country. The proposed work classifies the bottom water level data set records using ARIMA model to estimate the model for future test record data sets. The new model will helpf for analyzing ground water levels in past and then on find the long run levels.</p> </div> </div> </div> 2022-06-30T00:00:00+00:00 Copyright (c) 2022 FINGERPRINT-BASED SECURE DATA ENCRYPTION FOR PATIENTS HEALTHCARE 2022-07-08T00:43:57+00:00 C.Saravanan, R.Praveen,S.Priya Dharshini,P.Ragul,R.Snekha <div class="page" title="Page 1"> <div class="layoutArea"> <div class="column"> <p>Lately, remote sensor networks have been broadly utilized in medical care applications, like emergency clinic and home patient observing. Remote clinical sensor networks are more defenseless against snooping, modification, pantomime and replaying assaults than the wired organizations. A great deal of work has been done to get remote clinical sensor organizations. The current arrangements can safeguard the patient information during transmission, yet can't stop within assault where the manager of the patient data set uncovers the touchy patient information. In this paper, we propose a useful way to deal with forestall within assault by utilizing different information servers to store patient information.</p> </div> </div> </div> 2022-06-30T00:00:00+00:00 Copyright (c) 2022 SENTIMENT ANALYSIS OF CUSTOMER REVIEWS USING MACHINE LEARNING TECHNIQUES 2022-07-08T00:46:31+00:00 Nithya T, Atchaya S, Dharshini M, Harinesha D, Monisha M <p><strong><em>Sentiment analysis is one of the most important jobs in natural language processing, and it involves extracting attitudes, thoughts, views, or judgments on a certain issue. The internet is an unstructured and rich source of information that contains a large number of text documents offering thoughts and reviews. Individual decisionmakers, businesses, and governments may all benefit from sentiment recognition. We offer a deep learning-based technique to sentiment analysis on Twitter product evaluations in this study. The proposed architecture combines CNN-LSTM architecture with TF-IDF weighted Glove word embedding. Weighted embedding layer, convolution layer (where 1-g, 2-g, and 3-g convolutions have been used), max-pooling layer, followed by LSTM, and dense layer make up the CNN-LSTM architecture. The predictive performance of various word embedding schemes (e.g., word2vec, fastText, GloVe, LDA2vec, and DOC2vec) with various weighting functions (e.g., inverse document frequency, TF-IDF, and smoothed inverse document frequency function) was evaluated in conjunction with conventional deep neural network architectures in the empirical analysis. The suggested deep learning architecture outperforms traditional deep learning approaches, according to the empirical data.</em></strong></p> 2022-06-30T00:00:00+00:00 Copyright (c) 2022 THE EXPERIMENTAL CASE OF CYBERBULLYING DETECTION IN THE SOCIAL NETWORK 2022-07-08T00:49:27+00:00 Ms. Saranya T, Nivetha A, Parthiban S, Vidhyakaran R, Vikashini B <p><strong><em>Cyberbullying is a ubiquitous social practice that can have a variety of severe psychological, behavioral, and health consequences for cyberbullying victims, according to this investigation. According to studies, cyberbullying happens all throughout the world. Understanding the characteristics and processes that lead to cyberbullying is crucial for treatments aimed at decreasing antisocial behavior online. Cyber bullying is a widespread problem on the internet that affects both teens and adults. It has resulted in misfortunes such as suicide and sadness. Content regulation on social media sites is becoming increasingly important. The following work combines data from two types of cyber bullying, hate speech tweets from Twitter and comments from Wikipedia forums focused on personal assaults, to create a model for detecting cyberbullying in text data using Natural Language Processing and Machine Learning. To determine the optimum technique, three feature extraction approaches and four classifiers are investigated.</em></strong></p> 2022-06-30T00:00:00+00:00 Copyright (c) 2022