Browse Items (2150 total)
Sort by:
-
Cloud-Based Cataract Recognition System Using Hybrid Classifier Model
One of the key challenges of ophthalmologists is to diagnose the various ranges of ophthalmological illnesses such as diabetic retinopathy, cataract, and glaucoma. Here, cataract disease is identified as the one of the leading and most common ophthalmological problems that occurs due to aging. A computer-assisted cataract detection and diagnosis support system is required by the ophthalmologists to overcome the error that occurs during manual screening process. So, a cloud-based cataract recognition system is proposed using the convolutional neural network with support vector machine classifier model to improve the prediction accuracy, sensitivity, specificity, precision, recall, F1-score, and Mathews correlation coefficient. Moreover, the four-layer convolutional neural network is finetuned with a rich set of features and trained with various network models such as Inception V3, MobileNet, VGG-16, VGG-19, and ResNet-101. Therefore, the proposed hybrid combination of ResNet-101 with support vector machine classifier makes better cataract detection and outperforms the existing classifier models in terms of above-mentioned performance evaluation metrics. Moreover, the proposed hybrid approach provides the better telemedical solution to remote people by providing accurate disease prediction and severity grading such as normal, mild, premature, and severe cataract. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024. -
Cloud-Based Diabetic Retinopathy Severity Recognition System Using Ensemble Deep Convolutional Neural Network Classifier Model
One of the key reasons for visual impairments is due to the ignorance of diabetic retinopathy disease. This research study focuses on the early recognition of diabetic retinopathy disease from the fundus images and identifies its severity stages to make successful treatments against blindness risk. Some traditional approaches explored the decision tree, kernel-based support vector machine, and Nae Bayes classifier models to extract the features from fundus images. Most of the researchers applied the modern approach of convolutional neural network model through transfer learning mechanism to extract relevant features from the fundus images. It helps in the diagnosis of diabetic retinopathy that may delay the prediction process and create inconsistency among the doctors. So, a deep learning-based approach is proposed in this research study to provide stage-wise prediction of diabetic retinopathy disease with a multi-task learning mechanism. As a result, the proposed deep convolutional neural network classifier with an ensemble model outperforms the existing classifier with EfficientNet-B4, EfficientNet-B5, SE-ResNeXt50 (380?380), and SE-ResNeXt50 (512?512) networking methods in the context of prediction correctness, sensitivity, specificity, macro F1, and quadratic weighted kappa (QWK) score metrics. Exploiting hyperparameter optimizations on the deep learning classifier model and multi-task regression learning approaches make significant improvements over the performance evaluation metrics. Finally, the proposed approaches make the effective recognition of diabetic retinopathy disease stages based on the human fundus image. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024. -
CloudML: Privacy-Assured Healthcare Machine Learning Model for Cloud Network
Cloud computing is the need of the twenty-first century with an exponential increase in the volume of data. Compared to any other technologies, the cloud has seen fastest adoption in the industry. The popularity of cloud is closely linked to the benefits it offers which ranges from a group of stakeholders to huge number of entrepreneurs. This enables some prominent features such as elasticity, scalability, high availability, and accessibility. So, the increase in popularity of the cloud is linked to the influx of data that involves big data with some specialized techniques and tools. Many data analysis applications use clustering techniques incorporated with machine learning to derive useful information by grouping similar data, especially in healthcare and medical department for predicting symptoms of diseases. However, the security of healthcare data with a machine learning model for classifying patients information and genetic data is a major concern. So, to solve such problems, this paper proposes a Cloud-Machine Learning (CloudML) Model for encrypted heart disease datasets by employing a privacy preservation scheme in it. This model is designed in such a way that it does not vary in accuracy while clustering the datasets. The performance analysis of the model shows that the proposed approach yields significant results in terms of Communication Overhead, Storage Overhead, Runtime, Scalability, and Encryption Cost. 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
Cloudsim exploration: A knowledge framework for cloud computing researchers
This paper aims to help find solutions for questions an early researcher may have to set up experiments in their development environment. Simultaneously, while identifying the steps required for experimenting, the authors narrowed on an experimenting toolkit for Cloud Computing as an area of their study. Because of such simulators, the cloud computing environment itself is available easily at the comfort of ones desktop resources instead of visiting an actual physical data center to collect trace and log files as data sets for real workloads. This paper acts as an experience sharing to naive researchers who are interested in how to go about to start cloud computing setups. A new framework called Cloud Computing Simulation Environment (CCSE) is presented with inspiration from Procure Apply Consider and Transform (PACT) model to ease the learning process. The literature survey in this paper shares the path taken by researchers for understanding the architecture, technology, and tools required to set up a resilient test environment. This path also depicts the introduced framework CCSE. The parameters found out of the experiments were Virtual Machines (VMs), Cloudlets, Host, and Cores. The appropriate combination of the values of the parameters would be horizontal scaling of VMs. Increasing VMs does not influence the average execution time after a specific limit on the number of VMs allocated. Nevertheless, in vertical scaling, appropriate combinations of the cores and hosts yield better execution times. Thereby maintaining the optimal number of hosts is an ultimate saving of resources in case of VM allocations. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021. -
Cluster analysis for european neonatal jaundice
The objective of this paper is to propose and analyze clustering techniques for neonatal jaundice which will help in grouping the babies of similar symptoms. A variety of methods have been introduced in the literature for neonatal jaundice classification and feature selection. As far as we know, clustering techniques are not used for neonatal jaundice data set. This paper studies and proposes clustering techniques such as K-Means, Genetic K-Means and Bat K-Means for jaundice disease. To find the number of clusters elbow method is used. The clusters are validated using RMSE, SI and HI. The experimental results carried out in this paper shows bat k-means clustering performs better than K-means and genetic K-means. 2018, Springer International Publishing AG. -
Clustering-based Optimal Resource Allocation Strategy in Title Insurance Underwriting
Production of insurance policies in all types of Insurance requires a thorough examination of the entity against which the Insurance is to be issued. In health insurance, it is the past medical data of the individuals. Vehicle insurance needs the examination of the vehicle and the owner's data. Likewise, in Title Insurance, it is the historical data of the property which needs scrutiny before the policy issuance. Underwriters perform the job of examining the property records. The scrutiny of the property records requires a high degree of the domain and legal expertise, and title insurance underwriters are often associated with legal professions. They do the final round of validation of the examination process. There are examination teams that take care of the initial set of regular examination tasks associated with each title insurance order. Some human experts assign the orders to the team associates. Not all the orders are of the same complexity in terms of examination. The allocation of the tasks happens based on the gut feeling of the supervisor, considering their experience with the team members. Our research creates clusters of the orders based on specific parameters associated with the orders. It builds a cost model of the past associates working on orders belonging to different clusters. Based on this cost matrix, we have built an optimal task allocation model that assigns the orders to the associates with the promise of optimal cost using a Linear programming solution used frequently in operations research. 2022 IEEE. -
CNN based Model for Severity Analysis of Diabetic Retinopathy to aid Medical Treatment with Ayurvedic Perspective
One among the major modern life-style diseases is Diabetes. Diabetic Retinopathy is a major cause for blindness even at an early age. Clinical assessments for eye disease are done using visual examinations and probing. Retinal vessel segmentation is an important technique which helps in detection of changes that happens in blood vessel as well as gives information regarding the location of vessels. The work presented in this paper tries to detect and analyze the changes occurred in the blood vessels of human retina caused by diabetic retinopathy. Using digital imaging techniques, the severity screening technique facilitates the diagnosis of diabetic retinopathy. The model works in such a way that it helps the Ayurvedic treatment methodology for Diabetic Retinopathy. Results are obtained to categorize the data elements according to the severity of the disease and different classifications. 2022 IEEE. -
CNN-Bidirectional LSTM based Approach for Financial Fraud Detection and Prevention System
Detecting fraudulent activity has become a pressing issue in the ever-expanding realm of financial services, which is vital to ensuring a positive ecosystem for everyone involved. Traditional approaches to fraud detection typically rely on rule-based algorithms or manually pick a subset of attributes to perform prediction. Yet, users have complex interactions and always display a wealth of information when using financial services. These data provide a sizable Multiview network that is underutilized by standard approaches. The proposed method solves this problem by first cleaning and normalizing the data, then using Kernel principal component analysis to extract features, and finally using these features to train a model with CNN-BiLS TM, a neural network architecture that combines the best parts of the Bidirectional Long Short-Term Memory (BiLS TM) network and the Convolution Neural Network (CNN). BiLSTM makes better use of how text fits into time by looking at both the historical context and the context of what came after. 2023 IEEE. -
CNN-RNN based Hybrid Machine Learning Model to Predict the Currency Exchange Rate: USD to INR
Foreign currency exchange plays an imperative part in the global business and in monetary market. It is also an opportunity for many traders as an investment option and the advance knowledge of fluctuation helps the investors making right decision on time. However, due to its volatile nature, prediction of foreign currency exchange is a challenging task. This paper implements two models based on machine learning, namely Recurrent Neural Networks (RNN) and a Hybrid model of Convolutional Neural Networks (CNN) with RNN known as CNN-RNN to assess the accuracy in predicting the conversion rate of US Dollar (USD) to Indian Rupees (INR). The data set used to verify and validate the models is the daily currency exchange rate (USD to INR) available in public domain. The experimental results show that the simple RNN model performs slightly better than the hybrid model in this particular case. Though the accuracy of the hybrid model is very high in terms of error calculation still the single RNN model is the better performer. This does not straight away reject the hybrid model rather needs more experimental analysis with changing architecture and data set. 2022 IEEE. -
Co-MoS2 nanoflower coated carbon fabric as a flexible electrode for supercapacitor
Cobalt doped MoS2 (Co-MoS2) nanoflowers have been successfully synthesized via a simple one-step hydrothermal method for supercapacitor applications. To identify the crystalline nature and morphology, the as-prepared material is characterized by XRD, SEM, and TEM measurements. The material exhibits a specific capacitance value of 86 F g-1 at a current density of 1 Ag-1 in symmetric two-electrode configuration with excellent cyclic stability of 98.5% even after 10,000 chargedischarge cycles. The results suggest the suitability of Co-MoS2 as an efficient electrode material for supercapacitors. 2021 Elsevier Ltd. All rights reserved. -
Cognitive Engagement Scale (CES) in an Online Environment: Construction and Validation
Researchers have demonstrated linkages between active engagement of students with learning material and greater learning gains. Cognitive engagement is a significant component of educational experience. Understanding the challenges associated with cognitive engagement and measuring cognitive engagement in a MOOC environment is challenging. It is the need of the hour with online learning being equivalent to classroom learning in todays dynamic academic environment. The present study aims to construct cognitive engagement scale (CES) to measure the cognitive engagement of learners who sign up for the massive open online courses (MOOC). The aim of this study is dual-fold: firstly, to conceptualize the cognitive dimension of learner engagement within MOOCs, and secondly, to construct a theoretically informed scale for assessing cognitive engagement in online environments. Study presents a detailed process of the scale development, which included item generation, item evaluation, pilot testing, testing psychometric properties of the scale, and scale refinement. The researchers crafted the initial questionnaire drawing from both existing literature and personal insights. Subject matter experts then validated the items within the questionnaire and ensured its reliability through a pilot study, where it was administered to a sample of 100 participants The final version of the scale captures the four dimensions of cognitive engagement: Passive receiving, active manipulating, constructive generating, and interactive dialoguing. The present study contributes to the growing literature on cognitive engagement and adds to the existing literature of MOOC engagement scale with focus on cognitive engagement exclusively. The Author(s), under exclusive license to Springer Nature Switzerland AG 2025. -
Cognitive outcomes prediction in children using machine learning and big data analytics
This study explores the potential of machine learning (ML) and big data analytics in predicting cognitive outcomes in children, aiming to enhance early identification and intervention strategies. Leveraging a diverse dataset comprising neurocognitive assessments, genetic markers, socio-economic factors, and environmental variables, our research employs advanced ML algorithms to develop predictive models. The interdisciplinary approach integrates neuroscience, psychology, and data science to discern patterns and correlations within the expansive dataset. The study emphasizes the importance of early cognitive assessment for optimal child development and academic success. By harnessing the power of big data, our models seek to uncover nuanced relationships that traditional methodologies may overlook. Preliminary results indicate promising accuracy in predicting cognitive outcomes, offering a valuable tool for educators, healthcare professionals, and policymakers. Additionally, the model's interpretability allows for a deeper understanding of the factors influencing cognitive development. Ethical considerations, privacy safeguards, and data governance are integral components of this research, ensuring responsible use of sensitive information. The implications of this study extend beyond academia, with the potential to inform educational policies, personalized learning strategies, and targeted interventions for at-risk populations. As technological advancements continue, the integration of ML and big data analytics in predicting cognitive outcomes heralds a new era in pediatric research, promoting proactive approaches to support children's cognitive well-being. 2024 IEEE. -
CoInMPro: Confidential Inference and Model Protection Using Secure Multi-Party Computation
In the twenty-first century, machine learning has revolutionized insight generation by using historical data across domains like health care, finance, and pharma. The effectiveness of machine learning solutions depends largely on the collaboration between data owners, model owners, and ML clients, without privacy concerns. The existing privacy-preserving solutions lack efficient and confidential ML inference. This paper addresses this inefficiency by presenting the Confidential Inference and Model Protection, also known as the CoInMPro, to solve the privacy issue faced by model owners and ML clients. The CoInMPro technique is suggested with an aim to boost the privacy of model parameters and client input during ML inference, without affecting the accuracy and by paying a marginal performance cost. Secure multi-party computation (SMPC) techniques were used to calculate inference results confidentially after sharing client input and model parameters privately from different model owners. The technique was implemented in Python language using the open-source SyMPC library to support the SMPC function. The Boston Housing Dataset was used, and the experiments were run on Azure data science VM using Ubuntu OS. The result suggests CoInMPros effectiveness in addressing privacy concerns of model owners and inference clients, with no sizable impact on accuracy and trade-off. A linear impact on performance was noted with an increase of secure nodes in the SMPC cluster. 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
COLPOUSIT: A Hybrid Model for Tourist Place Recommendation based on Machine Learning Algorithms
Tourism is an important sector for a country's economic growth. The travel recommendations should be made focused on better growth and attract more travelers. There is a huge amount of travel information and ideas available on the web that allows the users to make poor travel decisions. This paper focuses on building a hybrid travel recommender system by implementing collaborative-based, popularity-based, and nearby place weighted recommender system. The proposed system recommends the travel spots to the users based upon their interests and other criteria specified. In order to implement these methods, we applied a comparative study on different machine learning algorithms for collaborative-based approach and have performed weighted hybridization. These methods provide a personalized and customized list of similar places with respect to places of interest to the users. Thus, a hybrid system built using these methods provides a better recommendation of places with the advantages of these methods. The obtained results confirm that the hybrid method better than other recommender approaches when used separately. 2021 IEEE. -
Combatting Phishing Threats: An NLP-Based Programming Approach for Detection of Malicious Emails and Texts
Attackers are employing more advanced strategies to trick people into divulging private information or carrying out harmful deeds, and phishing is still a serious cybersecurity risk. We provide a new method in this study that combines algorithms based on AI-based expert systems and deep learning (ML) with the use of NLP-based programming (NLP) approaches to identify fraudulent emails and messages. Using a variety of datasets that include samples of both authentic and phishing messages, our approach preprocesses textual data, extracts relevant characteristics, and trains AI-based expert systems and deep learning models. Metrics including accuracy, precision, recall, and F1-score are used to assess the effectiveness of different AI-based expert systems and deep learning methods, such as logistic regression, random forests, decision trees, and neural networks, among others. To collect semantic information and increase detection accuracy, we also investigate the integration of sophisticated NLP-based techniques, such as word embeddings. The efficacy of our suggested strategy in reducing this common cybersecurity issue is highlighted by our results, which show promising performance in correctly recognizing phishing attempts. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024. -
Combining Text Information and Sentiment Dictionary for Sentiment Analysis on Twitter During Covid
Presence of heterogenous huge data leads towards the 'big data' era. Technique's proliferation is rapidly increasing data and making dynamic changes that results in 'big data' world. Progressive transition in technologies and adoption of social media in the society also stepped into the 'big data' epoch. Social media popularity is uprising attention in the community. This platform reduces the communication gap among people. Recently, tweeter use increased with unprecedented rate. Presence of social media like tweeter has broken the boundaries and touches the mountain in generating the unstructured data. It opened research gate with great opportunities for analyzing data and mining 'valuable information'. Sentiment analysis is the most demanding, versatile research to know user viewpoint. Society current trend can be easily observed through social network websites. These opportunities bring challenges that leads to proliferation of tools. This research works to analyze sentiments using tweeter data using Hadoop technology. This study explores the big data arduous tool called Hadoop. Further, it explains the need of Hadoop in present scenario and role of Hadoop in storing ample of data and analyzing it. Hadoop cluster, HDFS, and Hive are also discussed in detail. Researchers enthusiastic work is deeply studied and presented here. Dataset used in performing the experiment is explained briefly. Moreover, this research explains thoroughly the implementation work and provide workflow. Next session provides the experimental results and analyzes of result. Finally, last session concludes the paper, its purpose, and how it can be used in upcoming research. 2024 IEEE. -
Compact Dual-Band Millimeter Wave MIMO Antenna for Wireless Communication Systems
The article presents the compact dual-band MIMO antenna resonating at 27.5 and 32 GHz. The radiating structure is a rose-shape with elliptical slots and a horizontal slit to achieve the above resonances. The MIMO antenna dimension is 6.2 0 mm2, where an edge-to-edge distance of 1.82 mm separates radiating elements. The ground plane has simple slits to suppress the mutual coupling. The simulation results of the MIMO antenna is validated through measured and diversity parameter results. 2024 IEEE. -
Compact out-of-phase wideband substrate integrated waveguide based power divider loaded by slots for Ku and K band applications
In this paper a novel Substrate Integrated Waveguide (SIW) based single layer ground-loaded compact wideband out-of phase equal power divider is proposed . The wide-band and out-of-phase operation of the proposed power divider is obtained by creating defects in the ground plane with rectangular slots. The Defected Ground Structure (DGS) allows the power divider to exhibit a wide passband. The obtained passband is 11.5 GHz wide considering the return loss better than -10dB. Compactness in the proposed design is attributed to the dispersion characteristic of the slow-wave. The proposed design working in the passband from 14.5 GHz to 26 GHz is fabricated and tested. The size of the proposed design is 0.57 ?2g excluding feed lines. Here ?g is the guided wavelength at free space. The measured amplitude imbalance of (01) dB is obtained within the passband. The measured and simulated results are compared and found with in good agreement. 2019 IEEE. -
Compact substrate integrated waveguide power divider with slot-loaded ground plane for dual-band applications
In this paper, a novel design of compact substrate integrated waveguide (SIW) dual-band power divider is proposed. The dual-band operation of the power divider is obtained by exploiting the loading of slots on the ground plane. The electric-dipole nature of these slots allows the power divider to exhibit a passband below the cutoff frequency of the SIW. An in-depth description of the proposed power divider, supported by detailed parametric analysis over the operating frequency bands is reported. Design examples are illustrated to achieve different operating frequency bands. To validate the design studies, a prototype of the dual-band power divider operating at 4.7 GHz and 11.7 GHz is designed, fabricated and tested. The measurement results are found to be in good agreement with the simulation results. 2018 IEEE. -
Comparative Analysis and Development of Recommendations for the Use of Machine Learning Methods to Identify Network Traffic Anomalies in the Development of a Subsystem for User Behavioral Analysis
This article discusses various machine learning methods in order to conduct a more effective analysis of user network traffic using a subsystem for analyzing user behavior and detecting network anomalies, since there is a need to evaluate big data. The methods and techniques used to detect network anomalies are analyzed. In analyzing the methods and technologies used to detect network anomalies, a classification of anomaly detection methods is proposed. To solve these problems, different algorithms can be used, differing in specificity and, as a result, efficiency. The classification of machine learning methods for detecting network anomalies is considered separately, since machine learning algorithms will be the most effective for the task. Various criteria for evaluating the effectiveness of machine learning models in solving the problem of network traffic profiling are considered. In accordance with the specifics of the tasks of user recognition and network anomaly detection, the most appropriate criteria for evaluating the effectiveness of machine learning models have been selected: AUC ROC the area under the error curve. Four stages of the subsystem for analyzing user behavior and detecting network anomalies are highlighted. The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.