Intelligent Systems

The polarization of the light can be used as the core principle of fiber optic sensors. One of the physical quantities which can be detected or measured this way is the longitudinal tension of the fiber. A set of measurements leading to approval of the suitability of polarization for this purpose was performed. This paper analyzes the dependency of differential group delay of the signal in slow and fast axes of the birefringent optical fiber on the longitudinal tension.

The range of diseases that can affect grape leaves has made it vital to analyze them. High-end data analytics and predictive analysis are required for a number of diseases, including black rot esca black measles, blight isariopsis, and others, in order to predict disease occurrence. For the prediction of leaf diseases, convolution neural networks combined with data augmentation have increased the degree of verification. For illness predictive analytics, a proper confusion matrix for support vector machines driven by CNN was created. Along with k-mean clustering, fuzzy logic with accurate feature extraction, and color moment definition, we also compared our results with these techniques. The findings indicate a higher effectiveness of up to 95% in correctly predicting grapes leaf disease.

In this article improved fuzzy PI regulator is stated for frequency regulation of Automatic Control of distributed power systems. Originally, a two-region nonwarm framework is utilized. The advantage of the stated fuzzy PI regulator is shown with the help of contrasting the outputs. All real structure shows non-straight nature, subsequently, traditional regulators are not generally ready to give great and precise outcomes. So fuzzy-logic controller can be utilized to get more exact outcomes. The primacy of the stated hybrid particle swarm optimization & pattern search (hPSO-PS) approach adjusted fuzzy-PI selector over PS changed fuzzy PI selector, PSO changed fuzzy PI selector, hBFOA-PS changed PI selector, Differential Evolution (DE) changed PI selector and Bacteria Foraging optimization algorithm (BFOA) adjusted PI selector is demonstrated. It is seen that the Fuzzy PI regulator is more effective for controlling frequency relative to the PI regulator.

Worldwide, depression affects millions of individuals even without their knowledge and is a crippling affliction. Primary care physicians frequently discover that they must treat mental health problems like depression despite having little or no formal training in how to do so. There is proof that an integrated strategy, where doctors regularly screen patients for mental health issues and collaborate with psychologists and other mental health specialists to treat patients, results in lower costs and improved patient outcomes. In order to handle and study the heterogeneous data and understand the correlation between aspects of quality of life and depression, this paper uses machine learning techniques. Machine learning is used to predict people who might have depression based on data that is found in CDC National Health and Examination Survey (NHAES) website. These forecasts could be used to more quickly and easily connect patients with qualified mental health specialists.

In Patient satisfaction, the most important factor in assessing the quality is patient happiness. The happiness key factor impacts the health policy decisions. An individual’s specific health requirements, individualised treatment, and desired health results are of the utmost importance in the period of patient-centered care. Across the past decade, treatment delivery, management, and reimbursement practices have all been impacted by patient satisfaction as a clear insight and quality management of patient experiences. Using machine learning algorithms, the most relevant factors for patient satisfaction are founds.

An endocrine gland that is allocated in the front of the neck is called the thyroid, which produces thyroid hormones as its main job. Thyroid hormone may be produced insufficiently or excessively as a result of its potential malfunction. There are various thyroid types including Hyperthyroidism, Hypothyroidism, Thyroid Cancer Thyroiditis, swelling of the thyroid. A goiter is an enlarged thyroid gland. When your thyroid gland produces more thyroid hormones than your body requires, you have hyperthyroidism. When the thyroid gland in our body doesn’t provide enough thyroid hormones, then our body has hypothyroidism; when you have euthyroid sick, your thyroid function tests during critical illness taken in an inpatient or intensive care setting show alterations. Hypothyroid, hyperthyroid, and euthyroid conditions are expected from these thyroid conditions. The Three similarly used machine learning algorithms are: Support Vector Machine (SVM), Logistic Regression, and Random Forest methods, were evaluated from among the various machine learning techniques to forecast and evaluate their performance in terms of accuracy. Random forest can perform both regression and classification tasks. Logistic Regression is used to calculate or predict the probability of a binary (yes/no) event occurring. SVM classifiers offers great accuracy and work well with high dimensional space. A thyroid data set from Kaggle is used for this. This study has demonstrated the use of SVM, logistic regression, and random forest as classification tools, as well as the understanding of how to forecast thyroid disease.

The significance of traffic symbol recognition technologies, which have played a key role in street security, has been the subject of much interest to researchers. To accomplish their assessment, specialists employed Artificial Intelligence, deep learning, and image processing tools. Convolutional Neural Networks (CNN) are deep learning-based designs that have sparked a new and ongoing research into traffic symbol classifications and recognition frameworks. The objective of this paper is to establish a CNN model that is suitable for insertion purposes and has a high level of order exactness. For the series of street symbols, we used an upgraded LeNet-5 model. The German Traffic Sign Recognition Benchmark (GTSRB) information base will function as the framework for our model architecture, which outperformed existing models. GTSRB will have 99.84 percent accuracy. We decided to use a camera to verify the proposed model for an implanted application because of its softness and reduced number of boundaries (0.38 million) based on the improved LeNet-5 structure. The outcomes are advantageous, demonstrating the effectiveness of the discussed strategy.

Many nations including India are being very badly affected by the second wave of the COVID-19 infections. The critical situation prevails in some states and cities of India. The mortality rate varies state to state depending on the health care facilities, immunological response of the individuals & comorbidities and vaccination status of that particular state. The multiclass prediction model is developed based on the status of data available from the different states of India considering their level of population density, intensity economic activities, education level, vaccination status and timing of lockdown or shut down. Based on this prediction model we can develop an application to motivate the internet of health things (IoHT), which can monitor the state and help in governing. This paper uses a multi class prediction model using Deep Neural Network (DNN) and validates the data set up to the year 2022, with accuracy level 98%. In this architecture, we have used 4 hidden layers between input and output layer. We have collected data from JHU CSSE Covid-19 and also follow our own algorithm to create our own dataset. We have taken 80% of data for training purposes and 20% of the dataset as validation purposes.

This paper presents a machine learning and web-based application for the detection of depression. The system mainly serves two components: two machine-learning-based models to detect depression and a web-based application. The first machine learning model is implemented to classify the positive and negative text entered by the user/patient. The negative text states the use of words indicating depression, which can be termed as one factor in deciding a patient's mental health. The model is built using libraries such as Natural Language Toolkit (NLTK), and WordCloud. The second model predicts the presence of depression based on multiple health-related features such as the patient’s data related to various other disorders he/she is having, age, weight, BMI, blood-related features such as levels of calcium, CO2, phosphorus, iron, etc., and work-life related parameters. The prediction is carried out based on the classification result implemented using Logistic Regression. The model predicts the results with 91.85% test accuracy, 93% precision, 99% recall, and 96% f1 score. The above-mentioned models are deployed on the web application. The web application not only helps in predicting mental health but also suggests the proper treatment to cure the condition.

In India Japanese Encephalitis (JEV) has been a major public health problem. In endemic districts of country each year there is a large-scale outbreak occurring of JEV. Research says that Japanese Encephalitis is a flavivirus related to West Nile Virus, Yellow Fever and Dengue and it is escalated by mosquitoes. Japanese Encephalitis is although rare, but the fatality rate is around 30%. Till now there is no cure for JEV, the entire treatment is focused for supporting the patient to overcome disease and relieving severe clinical sign. Maximum number of JEV cases in India are of infants and the fatality rate is around 30% which is a great matter of concern. Here Force of Infection denotes the rate at which sensitive individuals acquire an infectious disease. In India, states which report major outbreak of Japanese Encephalitis are Uttar pradesh, Andhra Pradesh, West Bengal, Karnataka, Assam, Tamil Nadu, Bihar, Goa and Manipur. The impacting factors include Climate, Rice Distribution, Livestock Distribution, Population Density, Specific Age Group Density, Urban/Rural Category and Elevation. Impacting Factors may change with the location. Here we have used Machine learning algorithms like Ridge Regression, Lasso Regression, ElasticNet Regression and Multi-layer Perceptron for the prediction of Force of Infection of Japanese Encephalitis Virus. ElasticNet Regression Algorithm is also used for extracting the significant attribute from the JEV Dataset. The proposed model generated an optimum performance in context to the error rate and accuracy of prediction.

With the advancement and rapid development of the internet, the most convenient strategies for patients are mainly provided with digital healthcare systems that mainly includes the use of mobile health technology which is quite efficient. Moreover, this field is slightly shifting and also indicating interest towards the smart and intelligent models as there are quite a lot of benefits associated with it like cost decrement, easy to understand and also including the personal satisfaction of patients. The latest application of m-health medical treatment is now still on the process of the investigation because still users are facing challenges in the clinical environment. This m-health approach can be applied to accurately determine skin cancer symptoms in patients. In this paper, an impact of m-healthcare on disease diagnosis is demonstrated. A new m-health module for skin cancer diagnosis called ‘Smart Skin-Proto’ is developed. Then its usage in skin cancer assessment is also highlighted and upon implementation, the model records optimal performance which records an accuracy of 96.2% with 15 decision trees count. Also the overall latency of this application is less than other existing mobile apps.

Devices can now effortlessly and wirelessly share data with one another over the internet or other networked systems thanks to a relatively new technology called Internet of Things (IoT). Despite these advantages, IoT systems are now more vulnerable to hacker attacks, which could lead to unfavourable outcomes. This is because of the IoT ecosystem’s continual expansion. These incursions may cause potential financial and physical harm. The Internet of Things is the automatically configuring network. This network is susceptible to a variety of attacks, all of which can be started by rogue nodes. For instance, during a denial of service attack, a malicious node bombards a targeted node with a large number of packets. For the purpose of locating these malicious nodes in a network, a threshold-based procedure utilising cutting-edge machine learning techniques is launched. By checking the path latency and alerting on it if it exceeds a set threshold value, the suggested method can help identify an attacker node. The NS2 programme will be used to mimic the suggested method. We evaluate the suggested methodology and demonstrate that our system performs well in terms of a number of measures, such as throughput, latency, and packet loss.

Writing has always been a prominent way of communication. The way in which the letters are written has been varying with time. From the conventional pen and paper to touch pad and stylus, the way of writing has evolved. Air- Writing is another development in which the characters are written in free space without being limited to a specific tool. This method of writing makes the hand movement easier compared to the conventional methods. Therefore, the air writing and recognition model will be of great help for children who start learning a language. The trajectory of the air written characters is obtained by mapping the focal point using Optical flow in OpenCV. The obtained trajectory is then preprocessed and given to Dense Net 121 which is a type of CNN model widely used for pattern matching along with the dataset from HP labs which contains 3000 images for 11 Tamil vowels. The model which is trained obtained a maximum training and validation accuracy of 98.2% and 91.83% respectively with minimum training and validation loss of 6.35% and 21.04% respectively.

Internet of Things (IoT) is a way of connecting the physical world to the internet where various devices are capable of communicating with each other. They all need a secure environment but the problem is implementation of any security approach for an IoT node is very difficult as resources of IoT nodes are very limited. So, trust evaluation and trust assessment are very important for IoT nodes. The paper proposes a fuzzy logic based trust evaluation model for IoT that employs different trust factors like End-to-end Packet Forwarding Ratio (EPFR), Amount of Energy Conversion (AEC), Packet Delivery Ratio (PDR) and Security Grade (SG) in order to construct a Fuzzy Inference System (FIS) that can calculate the trust value of each IoT node. Based on the resultant trust value, the IoT nodes are classified into three categories: Not Trustworthy, Not Sure and Trustworthy. An IoT node which belongs to Trustworthy only gets the access to forward the data packets or communicate with other IoT nodes. The Not Trustworthy and Not Sure IoT nodes are set to sleep mode for power conservation.

There are many chronic diseases out of which Diabetes is one; that increases sugar level in the blood and is one of the most fatal that effect different organs in the human body. Diabetes can cause a variety of slow bad consequences if not detected and left without given medical care. The emergence of machine learning approaches, on the other hand, solves this crucial issue. The purpose and objectives of this work is to build a prototypical model that can properly forecast diabetes whether or not a person will suffer from it. To detect diabetes at an early stage, our work employs three classification algorithms based on supervised learning: Random Forest, Naïve Bayes Classifier and Multilayer Perceptron Network. The PIDD Database has been used in the experiments. The Precision, Accuracy, Recall, F-Measure, and ROC Area are all used to calculate the efficiency of the above three algorithms. The correctness and accuracy of a classification system is measured by the number of occurrences that are correctly classified and those that are mistakenly classified.

In this paper we present the prototype of smart home which is powered by solar. It has a smart MPPT, smart health care tracking system and a smart home automation system. The sensors are spread all across the entrance Gate, corridor, room and kitchen. This (IOT) design prototype has LCD transistor which keep on provides the information. We have also use Wi-Fi technology for online control and monitoring. we also have an LCD which keeps us providing the information regarding the data. We have also used Wi-Fi technology for the purpose of real time controlling and monitoring. The designed smart home utilizes the power from the solar panels through a maximum power point tracker and it has an LCD display which continuously gives us information regarding the solar input, charging efficiency and discharging rate etc. The internal infrastructure is so designed that it can work against an unexpected condition which may occur when the owner is not present in home and it also notify the owner about the problem that has occurred. All the power requirements of smart homes are met by a self-generated solar power.

In recent years, occupational accidents causes a huge loss of human life and the development of the economy of the country. Many techniques are evolved for automating the safety precautions for employees in the industrial sectors such as mining, metals, construction, chemical, and electrical sections. However, the automation cannot be accurate as the data analysis is based on real-life data. Since the real-life data are imbalanced and uncertain, it is necessary to identify better tools to overcome these issues. Thus the proposed model utilizes SMOTE (Synthetic Minority Over-sampling Technique) for data balancing, whereas a rough set is used for identifying the significant features that help to maintain data consistency. The consistent data is then applied to the Deep Neural Network (DNN) for the classification process. The performance of the proposed model is checked against the evaluation metrics and compared with the existing deep learning models to exhibit the efficiency of the proposed model. Thus the findings of the proposed model may improve the abilities of safety professionals in the industrial sector to develop safety intervention activities.

Out of all existing natural calamities, flood is one of the most dangerous among all. It occurs when an excessive amount of water is gathered in a given space. It frequently occurs as a result of severe rain. Floods are one of the worst affecting natural phenomena which cause heavy damage to property, infrastructure, and most importantly human life. To prevent such disasters, various predictive models are used to forecast the floods that can occur in the future. It’s hard to create a predictive model because of its complexity. In this system, the rainfall data is fed into different machine-learning models. Before this process, the data is cleaned and pre-processed, and the dataset for training is split into a train set and a test set in an 80:20 ratio. Then the accuracy of each model is compared and the confusion matrix parameters are taken to evaluate and analyze. In the end, the best model is chosen by comparing the accuracy.

Cycling is an integral part of daily life for many people. This project presents a vision-based cyclist travel lane and helmet detection system. This system can serve as surveillance to detect whether the cyclist is traveling in a devoted lane and wearing a helmet for safety measures. The model involves the application of scale-invariant feature extraction (SIFT) algorithm for feature description. The detection method is based on six machine-learning classification algorithms. The classifiers are evaluated based on testing accuracy, and the best classifier is selected for the final model creation. The random forest classifier provided highest training accuracy of 99% and testing accuracy of 85.44% for cyclist travel lane detection. The same classifier provided the highest training accuracy of 99.53% and testing accuracy of 87.83% for cyclists’ helmet detection. In future, this system can also serve as a small part of autonomous driver assistance systems by detecting the right lane.

Traditional design approaches involve determination of the design parameters that meet only one design condition at a time, which is inadequate for high consistency and reliability. To overcome this problem, an attempt is made to introduce two new design objectives i.e., the weight/volume of the profile modified spur gear drive and the contact or Hertzian stress at points of contacts in the multi-objective design fitness function. The main contributions of the study are multi-objective constraint based design optimization using Particle swarm optimization (PSO), manufacturing of non-standard gear sets using standard tooling, experimental investigation to explore the possibility of protection of the optimized spur gear set against wear, through monitoring of oil bath temperature, gear surface temperature, and frictional power loss.

Obtaining a chest x-ray image is one of the main clinical observations for screening novel coronavirus. Most patients with COVID-19 viral pneumonia have abnormalities on a chest x-ray, such as consolidation. Computer vision-based solutions are a viable option for improving COVID-19 detection accuracy. However, Other classification models are presently in use in the healthcare industry. One such model uses radiographs to identify pneumonia cases and has attained a high enough level of accuracy to be applied to actual patients. This research assesses the advantages of employing various feature extraction strategies in order to improve the classification performance of the COVID-19 detection. The objective is to create a COVID-19 classifier using several feature extraction techniques, such as Fractal Descriptor (FD), Histogram Oriented Gradient (HOG), and Local Binary Pattern (LBP), using the pneumonia dataset as a base. Combining these feature extraction methods, an accuracy of 95% was attained utilizing FD.

Cricket is globally a famous game in which different technologies are being used to help the umpire in correct decision management. The bowl delivery is whether a fair one or LBW which is generally a matter of concern. Sometimes the decision of third umpire also goes wrong. This kind of error in making the decision can change the dimension of the game so, it is essential to make the decision precisely. This paper proposes computer vision and image subtraction technique. At first, it must figure out that whether a delivery is no-ball or the fair one, for that image processing method is used on popping crease by detecting the pixel size. If the delivery is a fair one then, it will track the movement of the ball by using the colour segmentation method, video processing on the pitch. Then by the help of ultra-edge detection technique it can identify whether the ball has touched the edge of the bat or not.

Nowadays, Recommendation system plays a vital role in industries like e-commerce, music apps or newsgroup, retailers, etc. Broadly, recommender system techniques are categorized into collaborative filtering and content based. In contrast, most recommendation models adopt collaborative filtering techniques such as matrix factorization (MF) and cosine similarity. However, the above model only deploys single techniques, which leads to poor recommendations to the individuals. Therefore we propose a mixed collaborative filtering-based recommender system (RS), a novel approach to improve the performance of the RS and mitigate the drawback of a single technique-based collaborative model. The proposed model is composed of two techniques such as singular value decomposition and cosine similarity techniques. Further, we examined our model’s performance on real-world datasets and found that the proposed approach significantly outperformed the baseline models.

Over the past few decades, heart disease has seen significant growth among all ages and early prediction became necessary. Data mining and machine learning techniques are used to solve the prediction problem utilizing new approaches to supervised learning. The Internet of Medical Things (IoMT) emerged from the combination of multiple fields and machine learning. The goal of this research is to develop an adaptive model for predicting cardiac disease. We provide a ranking-based hybrid feature selection method for identifying essential characteristics. The model proposed in this paper employs a clustering method in conjunction with support vector machine (SVM) to save training time and eliminate classification errors, hence boosting the model’s performance and increasing its efficiency.

Rising demands in investment in cryptocurrencies are being discussed of late in recent times. The most established and well-known cryptocurrency is Bitcoin. An accurate prediction of the bitcoin price will always attract more investors. This paper aims to demonstrate the effectiveness and appropriateness of several deep learning models in time series forecasting. This experiment makes use of the CoinDesk Bitcoin Dataset. Our results demonstrate that the Gated Recurrent Unit (GRU) based model surpasses all other models in accurately predicting bitcoin prices. We experimented with different DL (Deep Learning) models, ranging from a simple model to a complicated model. Standard metrics, such as Mean Absolute Error and MSE, have been used to analyse each model. In order to make better decisions in the near future, this study will benefit the finance industry.

With the growth of population, there is an increase in demand in food, as well as growth in agribusiness and hence, an increase in contract farming. Contract farming provides the farmers with the benefits such as enhanced agricultural production and a rise in incomes of contract farmers, giving them an exposure to services and resources inculcated with opportunities to participate in markets. However, time and again contract farming is criticised for the uneven bargaining power that may lead to the exploitation of farmers by contractors or large agribusiness firms. The presence of middlemen or brokers in the supply chain further coerces the farmers to rely on them to sell their products. Again, there is another issue of differentiating among bids. So, the solution for all these problems lies in “BidBlock”, a bidding integrated e-commerce platform, having built over blockchain and machine learning. Blockchain, with smart contracts, helps in restoring the trust of the farmers with its decentralization and machine learning, by interpreting and simplifying bids, helps them gaining maximum profit. Results show that most buyers prefer to have deals directly with the farmers and with a ease of doing business (like easy contract and payment). The addition of delivery services will be better too, but in this paper, it provides emphasis on how to simplify bidding with smart contracts, both for farmers and farms. Hence, introduction to BidBlock.

A weed is a wild, undesired plant that grows naturally along with the desired crop. For their growth, they compete with the main crop for various resources like space, sunlight, irrigation, nutrients, etc. This leads to an enormous loss in the yield of the main crop and hence needs to be selectively controlled. Human intervention in the process of identification of weeds and subsequently its removal is extremely tedious and time-consuming too. Achieving the desired level of accuracy and preciseness in the manual weed identification process is impracticable. In recent years, researchers have proposed computer vision-based methods for automatic weed identification in precision agriculture. In this paper, we have used YOLOv7-X for automatically detecting weeds in cotton production systems. YOLOv7-X is a relatively new addition to the You Only Look Once (YOLO) family of fast and accurate algorithms. The benchmark dataset used for the purpose of validating our results is CottonWeedid-15. This dataset is customized with annotations suitable for YOLOv7-X by using the roboflow tool. The experimental study demonstrates that the YOLOv7-X model’s mean average precision (mAP@.5) can attain 96.6 %. The average precision and average recall of the model were 0.914 and 0.953 respectively. This model can also be used to classify several weeds in various crops.

Traditional health care system need to be smarter because of the increasing complicacy in health of human kind. The new technology like IoT and Artificial intelligence system has a big role in making it success. IoT, Cloud Computing, and AI have made the traditional healthcare system smarter. Improving medical care with IoT and AI. Combining IoT and AI provides healthcare new options. In this approach, AI and IoT can assist intelligent healthcare systems detect illness. This article uses AI and IoT convergence to discover heart disease and diabetes. The given model includes data collection, preprocessing, classification, and parameter adjustment. Wearables and sensors make it easier to collect IoT data, which AI can use to diagnose illness. The suggested technique for detecting illnesses uses CSO-CLSTM, based on Crowd Search Optimization. CSO fine-tunes the CLSTM model’s “weights” and “bias” to enhance medical data categorization. Outliers are removed using the isolation Forest (iForest) approach. CSO improves CLSTM’s diagnostics. Healthcare data proved CSO-validity. LSTM’s CNN2D now includes a new version of LSTM and a CSO features selection approach. Experiments utilising Heart and Diabetes data reveal that extension is correct.

The complexity of software is increasing day-by-day. For faster and convenient implementation the software developer used to write similar codes in several places of software and these code segments are known as code clone. During the maintenance, if we know the places of code clones then it will be very effective. In this paper, we have proposed an unique technique of code clone detection using System Dependence Graph (SDG). The SDG for a program is the pictorial representation of the control and data flow in the program. The SDGs are generated automatically by the ASM API. We have created the SDGs for two programs and by analysing the SDGs we have detected the code clones. With case studies we have explained and demonstrated the working of our proposed technique. As an advantage of this the maintenance and reliability of the software will enhance.

Agriculture is the largest engine of economic growth in Bangladesh, providing food for almost half of the population. Fungal diseases in agriculture have a strong influence on producers’ livelihoods. To address this major worry, simulated design of an agriculture robot is presented in this research paper. The primary goal of this initiative was to improve competency in manufacturing agricultural grains, which will eventually lessen food shortages in Bangladesh while saving human labor, time, and money. All structural designs in this study were created in Autodesk Fusion 360 and then implemented in the Webots simulator. Proteus 8.9 Professional was used to model electrical circuits and decision processes, while MATLAB Simulink was used to accomplish a portion of the simulation. The key decision and control system for the entire system is a microcontroller. Irrigation, seed planting, excavating, leveling, automated ploughing, harvesting operation, and obstacle distance assessment are all possible with this designed robot. The simulation data, findings, and limitations were all observed. Based on the results, it is feasible to conclude that this farm robot can effectively enhance efficiency when compared to human labor. As a result of this combined design, Using IoT infrastructure, the operator can carry out several farming activities quite conveniently. The study offered a comprehensive perspective of the prototype architecture that demonstrated how well the farm robot’s functions.

Customer segmentation has been deployed as a prudent marketing strategy by companies to ensure that their investments are less risky and more judicious. Segmenting customers helps the companies to divide the customers into groups that reflect similarity and maximize the value of each customer to the business. The main goal of this research is to use a machine learning clustering approach called K-means clustering to accomplish consumer segmentation. Besides, the research work is also focused on performing exploratory data analysis on the given dataset. To group the customers into clusters, a K-means clustering algorithm is performed on the customer dataset. To achieve optimization and validation of the clusters, popular heuristic, interpretation, and approximation methods have been included in this paper. Further, for analyzing and visualizing the important facets of the customer dataset and the operation of the K-means algorithm, the paper presents some colorful and informative representations. The implementation of this research work has been done in the R programming language. The outcome of this work includes visualizing the segments of the mall customers in the form of clusters based on their spending scores and annual incomes. Furthermore, a better customer segmentation could be achieved by taking product reviews and customer feedback into consideration. Nevertheless, customer segmentation remains a prospective topic for many researchers and companies due to dynamic customer behavior.

Recently, research in the spread of information is found to be a crucial domain in the field of social network analysis. Understanding information spreadability and controllability are the two aspects of the same study. One of the important network parameters, the Inverse Participation Ratio (IPR) of a network adjacency matrix can measure the state of information localization. Higher the value of IPR, the higher the state of localization. This paper proposes a new perturbation approach based on k-shell decomposition to meet the optimal IPR. The proposed Shell-based Perturbation (SP) approach is compared with one of the state-of-the-art approaches: Random Perturbation (RP). The result confirms the superior performance of the proposed SP approach over the existing RP approach.

Despite the fact that many character datasets for several languages are publicly available, there are only a very few standardized datasets for Tamil characters. This article presents a subset of the Mepco Tamil Character database, a Tamil font isolated character dataset representing the printed characters. This dataset includes 124 glyphs representing the 247 characters of Tamil language. This dataset is tested for its robustness using multiple experimentations using SVM classifier and is compared against UJTDchar, another dataset available for Tamil language. Also we have verified the robustness of DIGI-Net, CNN architecture for this Tamil character recognition problem using the UJTDchar dataset and the Mepco Tamil Character dataset. We report an accuracy of 90.59% and 97.66% while using SVM and DIGI-Net CNN on our newly created dataset.

Red meat is one of the most popular varieties of meat in the eastern part of India. Consumption of prolonged accumulated or spoiled meat results in many fatal diseases. Traditional detection methods, such as sensory testing, physical and chemical testing, microbiological testing, and instrument analysis, are all complex, time-consuming, destructive, and uneconomical. In this study, we have designed an efficient, and non-destructive procedure for the classification of red meat based on its freshness using a novel convolutional neural network algorithm called “HarNet”. Our “HarNet” model gives better results and outperforms many pre-trained models like VGG16, VGG19, ResNet50, and InceptionV3. After testing and performing statistical analysis, our proposed model has achieved an accuracy of 80%. The F1-score value for the spoiled class is 0.89 and the recall value of 0.96 is the highest attained by the HarNet model, followed by the fresh class with 0.78 as the F1-score value and the recall value of 0.82 after testing. The results given by our proposed model are better than many of state of the art deep learning methods.

The detection of microorganisms is an important task in the clinical microbiology field. It is equally important during the pandemic breakout. Pathogenic microbes’ orientational behavior helps in distinguishing them. However, it is not an easy task to classify them based on that behavior only. In this research work, image processing and CNN methods like Resnet50, DenseNet121, Inception-ResNetv2, and MobileNetv2 have been implemented to classify species of 33 different pathogenic microbes. The pathogenic microbes have stained with Gramm’s method to distinguish them as gram-positive and gram-negative bacteria. Lactobacillus, Staphylococcus, and Enterococcus are used for their intra-general classification. Further, Stochastic gradient descent and fine-tuning are used to tune the learning rate. The result shows that 90.62 accuracies have been obtained using ResNet architecture for discriminative fine-tuning (DFT) and 92.96 accuracies have been obtained using stochastic gradient descent with the warm restarts (SGDR) approach. Similarly, 91.41 accuracies have been obtained using DenseNet 121 architecture for DFT and an accuracy of 98.7 has been obtained using Stochastic gradient descent with warm restarts approach. Also, an accuracy of 96.88 has been obtained using MobileNet architecture for DFT and an accuracy of 99.2 using Stochastic gradient descent with the warm restarts method. Further, Inception-ResNetV2 architecture has obtained an accuracy of 99.15 using DFT and 99.53 for the SGDR approach.

An unexpected demise due to cardiac arrest is a significant physical anomaly and is responsible for several deaths. Death due to unexpected cardiac arrest has no significant symptoms, and cardiac arrest has no initial symptoms. In most cases, a person may suffer from cardiac arrest, despite having a normal electrocardiogram (ECG). In this work, we have used two concepts, i.e., rough set theory (RST) to find the significant symptoms of cardiac arrest and support vector machines (SVM) to predict cardiac arrest. Our work aims to predict unexpected cardiac arrest less than half an hour before its occurrence. We have validated our claim using statistical techniques.

Globally, liver disease is the leading cause of death for a huge number of people. Inflammation of the liver is caused by a number of factors. Diagnosing liver infection early is essential for more effective treatment. In the current scenario, sensors are employed to identify liver diseases. Precise classification methods are necessary for the automatic diagnosis of illness samples. The cost of diagnosing this illness is high and complicated. The purpose of this study is to decrease the high cost of chronic liver disease diagnosis through prediction. This paper reviews the emerging techniques of data pre-processing, feature extraction, and classification on liver MRI. The primary goal of the current work is to use clinical data to predict the presence or absence of liver disease from MRI by applying various Machine Learning methods. In this paper, we have performed feature extraction from liver MRI using the HOG method followed by the Random Forest algorithm for the classification of images. With our approach, the accuracy achieved is 91.67%.

Text classification has gained importance due to the quickly rising content volume. During the text categorization process, it is necessary to complete tasks including extracting relevant information from different viewpoints, reducing the high feature space, and improving efficiency. Many studies on feature selection have been conducted, but increasing efficiency by reducing features remains a challenge. To evaluate its effectiveness, the Amazon review dataset is used in this proposed work. A chi-square-based enhanced genetic algorithm (CSEGA) approach is used in this paper to achieve the purpose of the study. The execution of the task includes the preprocessing task, followed by the optimization process for the selection of the optimal features. The unique crossover and selection process have made this work superior to other algorithms. With 93.0% accuracy, 94.5% precision, 90.5% recall, and a 92.47% F-score, the proposed approach outperforms other state-of-the art algorithms with fewer features.

Mobile edge computing (MEC) represents the promising technology that targets at facilitating different resources for processing and storing near the edge of mobile devices. Nevertheless, limited availability of resources in MECs possesses a necessity for adopting an appropriate management for preventing wastage of resources. In specific, the scheduling of workflow indicates a process that attempts in the mapping of tasks to the most suitable set of resources available in MECs based on the satisfaction of the objectives. In this paper, Tuna Optimization Algorithm-based Data Placement and Scheduling (TOA-DPS) is proposed with improved convergence by preventing the problems of local optima to map tasks to most suitable resources. It adopted a method of task prioritization for determining the order in which the tasks in the scientific workflows are executed. Further, Tuna Optimization Algorithm (TOA) is utilized for attaining data-demanding workflow scheduling along with data placement using Dynamic Voltage and Frequency Scaling (DVFS) in MEC environments. The performance evaluation of proposed TOA-DPS-based scheduling mechanism is conducted using extensive simulations carried out over renowned scientific workflows of various sizes. The outcomes of the proposed TOA-DPS-based scheduling scheme confirmed better performance of data access by 21.38%, and minimized energy consumption by 19.84%, better than the baseline approaches used for investigation.

Analysis of power output vs load profile in a system with scattered generation resources connected to the existing traditional energy system is critical, because even a minor frequency shift might cause a total blackout. The load frequency management problem for a hybrid coal-based system incorporated with DG, consisting of fuel cells, diesel engine generators, wind turbine generators, aqua-electrolyzer, and battery energy storage system is explored in this work. Due to the significant output power variation of wind energy systems, integrating them into DG offers a challenge for the creation of an appropriate controller. The stochastic volatility of the load profile makes this issue more difficult. The study’s suggested control method relies on differential evolution (DE). For various disturbances, the efficiency of frequency stabilization is studied. The results demonstrate how the hybrid DG system’s PID controller was able to achieve the least amount of frequency variation.

Heart disease, commonly referred to as cardiovascular disease and heart failure, has been the leading cause of mortality globally. Many risk factors for heart disease are associated with prompt access to reliable, dependable, and practical early diagnosis and disease management procedures. Identifying heart disease through early-stage signs is challenging in today’s global climate. If not caught in time, this could result in death. When there are no heart specialist doctors in remote, semi-urban, or rural areas, precise risk prediction and analysis might be critical in the early-stage identification of heart disorders. Machine learning (ML) and Deep learning (DL) approaches were employed in this study to assess massive volumes of complex medical data, supporting specialists in predicting heart illness and mortality from heart failure. This study used two datasets: one to forecast heart disease and the other to analyze and forecast death due to heart failure. Predicting cardiac illnesses using Artificial Neural Networks is 91.52% accurate (ANN). The bagging ensemble predicted heart failure with 90% accuracy. The primary contribution of this research is an ensemble strategy with high performance that multiple measurements have demonstrated to predict heart failure and cardiac disorders using ANN.

As digital communication has grown in a variety of industries where data must be secure and confidential, secret image transmission has grown in popularity. It will be a severe issue if false data is produced in the medicine, military, or diplomacy sectors. The three actors in conventional secret image sharing are the dealer, the participants, and the combiner. There is a chance that the Combiner will fabricate information, a participant will alter the data and submit a manipulated share, or someone will pose as a dealer and provide false information. To prevent data tampering, all actors in the secret image-sharing process should be verified and authenticated during transmission. This issue introduces a new area of study on secret sharing known as verification of secret sharing scheme, which addresses cheating activities by implementing various verification which includes dealer verification, cheating detection, and combiner verification techniques. This paper implements a secret image sharing (SIS) scheme with the ability to detect the cheater. Bitwise OR secret sharing scheme is used to share the secrets, and for verification one-way hash function, image hashing, and XOR operations are used. Encryption of the secret is done using the entropy of the image as the key. The dealer assigns shadow shares and shadow ID’s to the participants, which are then reconstructed by the combiner to create the original image. Participants validate the combiner’s authenticity, and the combiner validates the integrity of the shares before reconstruction. After the construction, the participants once again verify if the secret constructed is true or false data.

Data storage on cloud servers has become a common practice across businesses. Data security and accessibility in cloud environments are jeopardized when stored on unreliable cloud servers, making it necessary to convert user information into encrypted text that is difficult to understand and analyze, even if it is compromised. The novel encryption method known as Ciphertext Policy Attribute Based Encryption (CPABE) can be used to provide fine-grained access control and privacy in a cloud environment. To avoid the costly bilinear pairings, several researchers suggested CPABE approaches based on the elliptic curve cryptosystem (ECC). But there is a key-escrow concern with these systems. Hence, in this paper, we propose a multi-authority lightweight CPABE methodology. In addition, time and geographical attributes are also taken into consideration to enhance fine-grained access control. Our proposed solution also solves the issue related to dynamic attribute revocation by relegating repeated and tedious tasks to proxy servers. The detailed performance analysis demonstrates that the suggested method outperforms current schemes.

Schizophrenia is a chronic illness that most frequently affects people between the ages of 16 and 30. There are many elements that lead to a patient receiving a diagnosis of the illness, but since the origin of the illness is unknown, fuzzy analysis can be an important tool in identifying the factors. The purpose of this study report is to inform readers of the factors that have the greatest influence on an individual. Fuzzy logic is a powerful tool for tackling a wide range of problems in research. It can be used to model complex systems, analyze decision-making processes, and develop systems that can recognize patterns in data. In addition, fuzzy logic can be used to model complex systems such as financial markets and natural phenomena. As fuzzy logic is able to deal with uncertainty and imprecise data, it is particularly well-suited to a wide range of research. Fuzzy TOPSIS (Technique for Order Preference by Similarity to Ideal Solution) is a method used to determine the best of the options under consideration with respect to the weights and influence of each of the attributes associated with every option. Fuzzy AHP (Analytic Hierarchy Process) is a multi-criteria decision-making method that uses fuzzy set theory to evaluate alternatives. This method seeks to identify the best option from a set of alternatives based on the relative importance of the criteria. The goal of this study is to emphasize a through and in-depth literature evaluation of multi criteria decision making challenges in addition to pinpointing the aspects that contribute most significantly. We have implemented MCDM techniques to propose an effective approach towards decision making of various factors at multiple levels in patients suffering from schizophrenia and observed that the highest ranked Person (P4) with 0.8097 CCi can be inferred to most likely be affected by Schizophrenia, while the person at the bottom of the rank hierarchy (P1) with 0.2813 CCi is least likely to be affected.

This paper describes the classification of sports activities using still images. Sports activities specifically include discus throw, shot put and javelin throw. This system is useful for athletes preparing for these sports and it will help them to understand and practice the right posture for these sports. Along with them, this model is also helpful to the coaches to train their players and increase their chances of winning. Dataset is created by collecting the different images of the following sports from different sources. Various machine learning algorithms such as random forest, KNN, SVM (Polynomial), voting classifier has been applied to the model out of which SVM (poly) performed better than the rest with an accuracy of 98.7%. Voting classifier and Random Forest has also performed well. BRISK feature descriptor method is found useful to get point of interest in the image dataset. Techniques like image augmentation used for creating artificial dataset to which is useful for training the machine learning model and Principal Component Analysis (PCA) used for dimensionality reduction is found helpful to increase the accuracy of the model.

The unpredicted coronavirus outbreak, termed COVID-19, has placed numerous governments worldwide in a difficult position. The scarcity of resources to tackle the outbreak, combined with the fear of overburdening the healthcare system, has forced most countries into a state of lockdown. Many governments have shown great interest in digital contact tracing applications that can help automate the demanding task of tracking newly infected individuals’ recent contacts. However, these apps have created a great deal of discussion, especially regarding their technology, architecture, and the adoption rate needed. This study aims to contribute to an increased understanding of the acceptance of this technology in the Norwegian population. Based on the unified theory of acceptance and use of technology (UTAUT) model, our research model incorporates the following five constructs: performance expectancy, effort expectancy, social influence, facilitating conditions, and privacy consideration. A survey was distributed amongst the Norwegian population, and the results were obtained from a sample of 258 respondents. The results from this study indicate that performance expectancy has the most significant impact on the intention to use a contact tracing application. Privacy considerations are also important, followed by effort expectancy and social influence. Facilitating conditions were found to be much less important.

The development of the digital communication system is increasing rapidly and all the manual data management process is massively converted to the digital processing system to make faster processing and reduce the valuable time of human users and increase productivity. Now the input and output system of the device is converted to a higher-level interface medium so that the device can communicate with the user and environment through audiovisual components and communicate like a human. Also in the present scenario devices are implemented by human society to deal with very secure and sensitive information like banking, the military, the research environment, the medical sector, etc. Hence there is a chance of misutilization of information through the device. Many Security using mathematical techniques like cryptography, digital signature, etc. are used to make secure information from the unauthentic person, but to protect the copywriting of the digital information watermark technology is used. Through the watermark, we can add the identification of the owner hence the other person can not claim or tamper with the information for its misutilization. In this proposed work we reviewed different techniques to create a watermark with the algorithm associated with it which helps the researcher extend the watermark technique in a better way.

Stress is vital in assessing the physical and mental state of the human body with significant psychological and physiological changes. A proper and timely diagnosis of stress may make one healthier, happier, and more productive. In the workplace, undergoing many changes leads to stress, trauma, and anxiety. At the same time, hormonal changes in the human body due to stress can be reflected in terms of psychological and physiological changes. This paper has identified three different activities (normal, tension, and exercise) with varied positions (laying, sitting, and standing). Airflow, Temperature, and Galvanic Skin Response (GSR) are different sensors that sense data. This work has emphasized GSR sensors and conceptually connected them with other sensors. GSR values differ regarding the contact surface area with the body. Different machine learning algorithms such as; Naive Bayes, Support Vector Machine, Decision Tree (J48), and Random Forest have been used to analyze sensed datasets. Random Forest Algorithm has been observed to perform better in the proposed work.

Data privacy and data security are the main concerns in the digital era. The 3.5% centralized increase in annual digital data and the use of machine learning and deep learning approaches in the centralized computing environment endanger data privacy and security. The evolution of various body sensors also increases the digital health parameter data, which also demands privacy and security, which are difficult to achieve in a centralized computing environment. In the article, a two-level VANET-based federated learning framework is proposed for the classification of health parameters, in which Road Side Unit (RSU) acts as the local server in the first level and cloud networks act as the remote server in the second level. Health parameters considered for the proposed work are body temperature, heart rate, and systolic and diastolic blood pressure. The proposed work classifies health parameters into four categories, such as normal, low-risk, medium-risk, and high-risk data. Accuracy, precision, recall, and loss are used to evaluate the proposed work. In addition, in terms of average false classification rate and multi-class classification accuracy, the proposed federated learning computing is compared to centrally managed computing.

In Wireless Sensor Network (WSN)-enabled Internet of Things (IoT) environment, potential resource utilization and efficient service delivery are of great interest. In specific, IoT-based networks completely depend on energy-efficient clustering architecture that is used for transferring data between heterogeneous devices, and optimal energy-aware deployment methods in WSN. Clustering-based energy-efficient routing and sensor node deployment are identified to extend network lifetime. Efficient selection of Zone Heads (ZHs) during the process of partitioning the network into different zones or clusters is essential for maximizing the reachability of nodes within clusters, and for achieving better communication with the Base Station (BS). In this paper, ELECTRE I-based Zone Head Selection (EZHS) scheme based on distinct factors such as number of times a node is selected as ZH, distance of the sensor node from the center, distance between adjacent nodes and level of energy is proposed, since they directly influence network lifetime and drain of node energy. It adopts the merits of ELECTRE I-based Multi-Criteria Decision Making (MCDM) model for determining the relative influence of impactful parameters during the process of ZH selection. The results of the proposed EZHS approach confirm an improvement in network lifetime by 21.98% and network stability by 22.94% in contrast to the benchmarked strategies.

Electrochemical pH sensors are progressively in demand for applications such as continuous monitoring of water quality and health monitoring. This is due to better stability over a wide pH range, ease of fabrication, mechanical robustness & miniaturization. In this work, a thick film-based pH sensor is fabricated using pH-sensitive oxides RuO2 & TiO2. An interdigitated conducting electrode’s surface was screen printed with the sensitive RuO2 - TiO2 paste for conductimetric pH monitoring. The constructed IDE pH sensor has the following main advantages: faster and less expensive fabrication. Performance analysis of fabricated sensors using impedance spectroscopy is presented and discussed. Optimized performance is obtained for thick film pH sensor when we mix two metal oxides RuO2 & TiO2 in the proportion 80:20 wt.%. Its pertinency for measurements of sweat pH is checked by measuring the pH value of an artificial sweat i.e., human sweat equivalent solution. The pH sensors were characterized using electrochemical impedance spectroscopy in the 10 Hz to 8 MHz frequency range.

Wireless sensor networks are equipped with energy constrained and limited communication range sensor nodes to transmit the sensed data. The constraints made prolonged network lifetime is one of the major concerns while dealing with the energy hole problem. In the presence of nodes whose behaviour is unpredictable as a relay node (unreliable nodes/URNodes) will raise the concern about reliable data delivery and makes enhancement of network lifetime difficult. In this paper, the energy hole problem is addressed by looking for an Alternate Path (AP) by availing the residual energy of other sensor nodes who involves in data transmission rarely. In the presence of URNodes with various percentage of unreliability, the AP approach reaches almost same network lifetime when compared with the network without URNodes. Experiments have been done with increased density of the network sensor nodes with varied number of unreliable node (URNodes) with various percentage of unreliability. The main concern of the work is reliable data delivery with extended network lifetime. So we also considered the adjustable communication range to ensure reliable data delivery. With an adjustable communication range, along with the enhanced network lifetime, the total number of dropped messages due to URNodes are also got reduced.