Browse Items (2150 total)
Sort by:
-
Applications of Classification and Recommendation Techniques to Analyze Soil Data and Water Using IOT
As we are moving to a computerized and scientific world, data becomes an intrinsic part of our life. Agriculture sector is still unorganized with regard to automation and data analytics. This task is accomplished through sensors, data mining and analysis. In this paper, we propose real-time sensors to detect the soil features and predict the suitable crop cultivation using trained dataset. This would help the farmers to predict the type of cultivation to be done depending on the soil features. Today, the farmer can understand what type of cultivation will be prepared in the soil. Also, people of the upcoming generation will be using that sensor, different plant can be make. The cost of cultivation can be improved. Water level of the soil can be easily predicted. Which type of plant will be produced in the different soil can be predicted. So, this new type of cultivation followed by the next generation also. This paper has presented an improved by the pH sensor, water level sensor. 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
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. -
A Deep Learning Method for Classification in Brain-Computer Interface
Neural activity is the controlling signal used in enabling BCI to have direct communication with a computer. An array of EEG signals aid in the selection of the neural signal. The feature extractors and classifiers have a specific pattern of EEG control for a given BCI protocol, which is tailor-made and limited to that specific signal. Although a single protocol is applied in the deep neural networks used in EEG-based brain-computer interfaces, which are being used in the feature extraction and classification of speech recognition and computer vision, it is unclear how these architectures find themselves generalized in other area and prototypes. The deep learning approach used in transferring knowledge acquired from the source tasks to the target tasks is called transfer learning. Conventional machine learning algorithms have been surpassed by deep neural networks while solving problems concerning the real world. However, the best deep neural networks were identified by considering the knowledge of the problem domain. A significant amount of time and computational resources have to be spent to validate this approach. This work presents a deep learning neural network architecture based on Visual Geometry Group Network (VGGNet), Residual Network (ResNet), and inception network methods. Experimental results show that the proposed method achieves better performance than other methods. 2023 IEEE. -
Machine Learning Techniques for Resource-Constrained Devices in IoT Applications with CP-ABE Scheme
Ciphertext-policy attribute-based encryption (CP-ABE) is one of the promising schemes which provides security and fine-grain access control for outsourced data. The emergence of cloud computing allows many organizations to store their data, even sensitive data, in cloud storage. This raises the concern of security and access control of stored data in a third-party service provider. To solve this problem, CP-ABE can be used. CP-ABE cannot only be used in cloud computing but can also be used in other areas such as machine learning (ML) and the Internet of things (IoT). In this paper, the main focus is discussing the use of the CP-ABE scheme in different areas mainly ML and IoT. In ML, data sets are trained, and they can be used for decision-making in the CP-ABE scheme in several scenarios. IoT devices are mostly resource-constrained and has to process huge amounts of data so these kinds of resource-constrained devices cannot use the CP-ABE scheme. So, some solutions for these problems are discussed in this paper. Two security schemes used in resource-constrained devices are discussed. 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
A Hybrid Machine Learning Model (NB-SVM) for Cardiovascular Disease Prediction
One of the leading causes of death is heart disease. The prediction of cardiovascular disease remains as a significant challenge in the clinical data analysis domain. Although predicting cardiac disease with a high degree of accuracy is highly challenging, it is possible with Machine Learning (ML) approaches. The implementation of an effective ML system can minimize the need for additional medical testing, minimize human intervention, and predict cardiovascular diseases with high accuracy. This type of assessment can reduce the disease's severity and mortality rate. Only a few studies show how machine learning techniques might forecast cardiac disease. This study presents a method for improving cardiovascular disease prediction accuracy using Machine Learning (ML) technologies. Various feature combinations and many known classification techniques are used to develop various cardio vascular disease prediction models. The proposed hybrid Machine Learning (ML) prediction model for heart disease leverages a higher degree of performance and accuracy. 2023 IEEE. -
Towards Computation Offloading Approaches in IoT-Fog-Cloud Environment: Survey on Concepts, Architectures, Tools and Methodologies
The Internet of Things (IoT) provides communication and processing power to different entities connected to it, thereby redefining the way objects interact with one another. IoT has evolved as a promising platform within short duration of time due to its less complexity and wide applicability. IoT applications generally rely on cloud for extended storage, processing and analytics. Cloud computing increased the acceptance of IoT applications due to enhanced storage and processing. However, the integration does not offer support for latency-sensitive IoT applications. The latency-sensitive IoT applications had greatly benefited with the introduction of fog/edge layer to the existing IoT-Cloud architecture. The fog layer lies close to the edge of the network making the response time better and reducing the delay considerably. The three-tier architecture is still in its earlier phase and needs to be researched further. This paper addresses the offloading issues in IoT-Fog-Cloud architecture which helps to evenly distribute the incoming workload to available fog nodes. Offloading algorithms have to be carefully chosen to improve the performance of application. The different algorithms available in literature, the methodologies and simulation environments used for the implementation, the benefits of each approach and future research trends for offloading are surveyed in this paper. The survey shows that the offloading algorithms are an active research area where more explorations have to be done. 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
Identification of Driver Drowsiness Detection using a Regularized Extreme Learning Machine
In the field of accident avoidance systems, figuring out how to keep drivers from getting sleepy is a major challenge. The only way to prevent dozing off behind the wheel is to have a system in place that can accurately detect when a driver's attention has drifted and then alert and revive them. This paper presents a method for detection that makes use of image processing software to examine video camera stills of the driver's face. Driver inattention is measured by how much the eyes are open or closed. This paper introduces Regularized Extreme Learning Machine, a novel approach based on the structural risk reduction principle and weighted least squares, which is applied following preprocessing, binarization, and noise removal. Generalization performance was significantly improved in most cases using the proposed algorithm without requiring additional training time. This approach outperforms both the CNN and ELM models, with an accuracy of around 99% being achieved. 2023 IEEE. -
Systematic Review on Decentralised Artificial Intelligence and Its Applications
Initially, Artificial Intelligence (AI) models were centralized. This resulted in various challenges. To overcome this challenge, the decentralized or distributed frameworks were developed. Recent advancements in blockchain technology and cryptography have accelerated the decentralization process. Decentralized Artificial Intelligence (DAI) is gaining a significant research attention in recent times. This study reviews various DAI techniques such as Decentralized machine learning frameworks, Federated Learning and Distributed AI marketplaces. In particular, this study focuses on reviewing the recent developments in DAI by analyzing its potential advantages and challenges. 2023 IEEE. -
ByWalk: Unriddling Blind Overtake Scenario with Frugal Safety System
Safety is crucial, and the truth is ineluctable with its practicality. We strive forward to rev up the safety protocols even more in the field of Road Safety in particular. Countries like India face around 5,00,000 accidents, which lead to 1,80,000 demises each year. The two-lane one-way roads present a risk of the overtaking vehicle crashing onto an incoming car (from the opposite direction) that the overtaking vehicle is unaware of. We seek to achieve two equivocal milestones with our idea in the blind overtake issue, namely, technological aid and economic feasibility. This makes our concept equally impactful in all situations. The technological precision and advancement will help anyone with enough resources to use them tangibly, and economic feasibility ensures a threshold of safety levels that must be put into action. In fact, we are slightly inclined toward the frugality of the architecture paradigm of our idea because safety is everyones right. On the economic side, we propose an LED board-based solution that presents enough information about the incoming vehicle with which a blind overtake condition can be avoided. Besides, we put forward the idea of vehicle-to-vehicle communication for streaming the video content to the trailing cars with smarter selection and added ease to the drivers. 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
A Comprehensive Study on E-learning Environments for Deaf or Hard of Hearing Learners
Quality education is the fundamental right of every individual regardless of the disabilities they have. For the Deaf or Hard of Hearing (d/DHH) people, e-learning is the most promising way to access the educational materials referred to as digital learning objects (LO) at any time and space which increase their autonomous learning skills. This form of instruction delivery was widely accepted during the outbreak of Covid-19. Hence a background study has been conducted to investigate the challenges in teaching the d/DHH learners during the pandemic. This research work aims at providing a personalized e-learning environment to the d/DHH student community belonging to St. Clare Oral Higher Secondary School for The Deaf, situated in Kerala. To build personalized systems, the primary step is to review the existing e-learning solutions available in the literature and the adaptation techniques implemented by them to offer personalization in line with the components of traditional adaptive e-learning systems. The study carried out in this paper illuminates the need of personalized e-learning platforms that adapt the basic needs, abilities and disabilities of deaf learners which will find the 'best learning solutions' in the form of learning objects. 2023 IEEE. -
Advancements in Cyber Security and Information Systems in Healthcare from 2004 to 2022: A Bibliometric Analysis
The main goals of the multifaceted healthcare system were to prevent, identify, and treat illnesses or conditions that affect human health. As the usage of IT in healthcare increased, the complexities in managing the IT infrastructure also increase, emphasizing the need of robust cyber security systems. The study aims to emphasize the advancements made in cyber security and information systems in healthcare, based on bibliometric analysis. 5,487 document's metadata was obtained from Scopus and data was analyzed using Vos Viewer. Ranking of articles was done with average yearly citations of the publications. Bibliometric analysis was performed based on 'bibliographic coupling of countries', 'co-occurrence of all keywords', 'author-based co-authorship', and 'term co-occurrence based on text data'. It was found that United States had the maximum publications (1337). 'Department of Information Systems and Cyber Security, The University of Texas at San Antonio, United States' is the most influential organization with 159 publications. IEEE Access is the most preferred platform for publication related to cyber security and information systems in healthcare (231 publications). 167 publications have received more than 100 citations. Choo K. K.R. is the most influential author with 185 publications. 2023 IEEE. -
Identifying a Range of Important Issues to Improve Crop Production
Crop yield production value update has a beneficial practical impact on directing agricultural production and informing farmers of changes in crop market prices. The main objective of the suggested method is to put the crop selection technique into practise so that it may be used to address a variety of issues facing farmers and the agricultural industry. As a result, the yield rate of crop production is maximised, which benefits our Indian economy. land conditions of several kinds. So, using a ranking system, the quality of the crops are determined. This procedure also alerts farmers to the rate of crops of low and high quality. Due to the use of multiple classifiers, using an ensemble of classifiers paves the way for better prediction decisions. The decision-making process for selecting the output of the classifiers also incorporates a rating system. The price of a crop that will produce more is predicted using this method. 2023 IEEE. -
Computational Modelling of Complex Systems for Democratizing Higher Education: A Tutorial on SAR Simulation
Engineering systems like Synthetic Aperture Radar (SAR) are complex systems and require multi-domain knowledge to understand. Teaching and learning SAR processing is intensive in terms of time and resources. It also requires software tools and computational power for preprocessing and image analysis. Extensive literature exists on computational models of SAR in MATLAB and other commercial platforms. Availability of computational models in open-source reproducible platforms like Python kernel in Jupyter notebooks running on Google Colaboratory democratizes such difficult topics and facilitates student learning. The model, discussed here, generates SAR data for a point scatterer using SAR geometry, antenna pattern, and range equation and processes the data in range and azimuth with an aim to generate SAR image. The model demonstrates the generation of synthetic aperture and the echo signal qualities as also how the pulse-to-pulse fluctuating range of a target requires resampling to align the energy with a regular grid. The model allows for changing parameters to alter for resolution, squint, geometry, radar elements such as antenna dimensions, and other factors. A successful learning outcome would be to understand where parameters need to be changed, to affect the model in a specific way. Factors affecting Range Doppler processing are demonstrated. Use of the discussed model nullifies use of commercial software and democratizes SAR topic in higher education. 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
Exploring Machine Learning Models to Predict the Diamond Price: A Data Mining Utility Using Weka
In contrast to gold and platinum, whose values may be fairly determined, determining a diamond's worth involves a far more complex set of considerations. The appropriate rate is based on many factors, not just one of the stones. Diamonds are graded based on their appearance, carat weight, cut quality, and how well they have presented dimensions like a table's surface, depth, and breadth. In order to accurately forecast diamond prices, this study seeks to develop the most effective approaches possible. Different machine learning classifiers are trained on the diamond dataset to forecast diamond prices based on the features. This article shows how to analyze diamond prices using WEKA's data mining software. Diamond data have been utilized for this study. These methods include M5P, Random Forest, Multilayer perceptron, Decision Stump, REP Trees, and M5Rules. For the purpose of estimating the cost of a diamond, different Machine Learning classifiers are compared and contrasted. Performance measures and analysis showed that Random Forest was the best-performing classifier. Experimental findings show, as shown by the coefficient of correlation that Random Forest is better than other classification methods. 2023 IEEE. -
Comparative Analysis of CPI prediction for India using Statistical methods and Neural Networks
Inflation is one of the main issues affecting the world economy right now, necessitating the accurate inflation prediction for the development of tools and policies by the monetary authorities to prevent extreme price volatility. Expectations of inflation influence many financial and economic actions, and this dependence motivates economists to develop techniques for precise inflation forecasting. Nearly everyone in the economy is impacted by inflation, including lending institutions, stock brokers, and corporate financial officials. In many cases, inflation determines whether a firm will accept a particular project or if banks will make a particular loan. These different economic actors can modify their financial portfolios, strategic goals, and upcoming investments if they are able to forecast changes in inflation rates. The multiple interaction economic components that depend on inflation will be better understood by economic agents operating in a business context if inflation forecasting accuracy is improved. There are numerous techniques to forecast inflation ranging from basic statistical methods to complex neural network methods. Therefore, this paper employs LSTM model to train and analyze the Consumer Price Index (CPI) indicators to obtain inflation-related prediction results. The experimental results on historical data show that the statistical model has good performance in predicting India's inflation rate compared to deep learning methods in case of smaller dataset. 2023 IEEE. -
Implementation of Supervised Pre-Training Methods for Univariate Time Series Forecasting
There has been a recent deep learning revolution in Computer Vision and Natural Language Processing. One of the biggest reasons for this has been the availability of large-scale datasets to pre-train on. One can argue that the Time Series domain has been left out of the aforementioned revolution. The lack of large scale pretrained models could be one of the reasons for this.While there have been prior experiments using pre-trained models for time series forecasting, the scale of the dataset has been relatively small. One of the few time series problems with large scale data available for pre-training is the financial domain. Therefore, this paper takes advantage of this and pretrains a ID CNN using a dataset of 728 US Stock Daily Closing Price Data in total, 2,533,901 rows. Then, we fine-tune and evaluate a dataset of the NIFTY 200 stocks' Closing Prices, in total 166,379 rows. Our results show a 32% improvement in RMSE and a 36% improvement in convergence speed when compared to a baseline non pre trained model. 2023 IEEE. -
Adopting Metaverse as a Pedagogy in Problem-Based Learning
Pedagogical practices vary from time to time based on the requirement of various academic disciplines. Course instructors are constantly searching for inclusive and innovative pedagogies to enhance learning experiences. The introduction of Metaverse can be observed as an opportunity to enable the course instructors to combine virtual reality with augmented reality to enable immersive learning. The scope of immersive learning experience with Metaverse attracted many major universities in the world to try Metaverse as a pedagogy in fields such as management studies, medical education, and architecture. Adopting Metaverse as a pedagogy for problem-based learning enables the course instructors to create an active learning space that tackles the physical barriers of traditional pedagogical practices of case-based learning facilitating collaborative learning. Metaverse, as an established virtual learning platform, is provided by Meta Inc., providing the company a monopoly over the VR-based pedagogy. Entry of other tech firms into similar or collaborative ventures would open up a wide array of virtual reality-based platforms, eliminating the monopoly and subsequent dependency on a singular platform. The findings of the study indicate that, currently, the engagements on Metaverse are limited to tier 1 educational institutions worldwide due to the initial investment requirements. The wide adoption of the Metaverse platform in future depends on the ability of the platform providers to bridge the digital gap and facilitate curricula development. 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG. -
A Novel Steganographic Approach for Image Encryption Using Watermarking
Steganography is a technique for obfuscating secret information by enclosing it in a regular, non-secret file or communication; the information is subsequently extracted at the intended location. Steganography can be used in addition to encryption to further conceal or safeguard data. Watermarking is one such technique practiced in the area of steganography. Watermarking can be practiced via multiple algorithmic techniques like Discrete Wavelength Transform (DWT), Discrete Cosine Transform (DCT), Singular Value Decomposition (SVD), Discrete Fourier Transform (DFT). In this study, a combination of such approaches along with AES encrypted watermarked images has been implemented. Validation of these techniques has been achieved by evaluating the Peak Signal to Noise Ratio (PSNR). 2023 IEEE. -
Dynamic Load Scheduling Using Clustering for Increasing Efficiency of Warehouse Order Fulfillment Done Through Pick and Place Bots
The domain of warehouse automation has been picking up due to the vast developments in e-commerce owing to growing demand and the need to improve customer satisfaction. The one crucial component that needs to be integrated into large warehouses is automated pick and place of orders from the storage facility using automated vehicles integrated with a forklift (Pick and Place bots). Even with automation being employed, there is a lot of room for improvement with the current technology being used as the loading of the bots is inefficient and not dynamic. This paper discusses a method to dynamically allocate load between the Pick and Place BOTs in a warehouse during order fulfillment. This dynamic allocation is done using clustering,an unsupervised Machine Learning algorithm. This paper discusses using fuzzy C-means clustering to improve the efficiency of warehouse automation. The discussed algorithm improves the efficiency of order fulfillment significantly and is demonstrated in this paper using multiple simulations to see around 35% reduction in order fulfillment time and around 55% increase in efficiency. 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
Systematic Review on Humanizing Machine Intelligence and Artificial Intelligence
In this era, Machine Learning is transforming human lives in a very different way. The need to give machines the power to make decisions or giving the moral compass is a big dilemma when humanity is more divided than it has ever been. There are two main ways in which law and AI interact. AI may be subject to legal restrictions and be employed in courtroom procedures. The world around us is being significantly and swiftly changed by AI in all of its manifestations. Public law includes important facets such as nondiscrimination law and labor law. In a manner similar to this when artificial intelligence (AI) is applied to tangible technology like robots. In certain cases, artificial intelligence (AI) might be hardly noticeable to customers but evident to those who built and are using it. The behavior research offers suggestions for how to build enduring and beneficial interactions between intelligent robots and people. The human improvement is main obstacles in the development and implementation of artificial intelligence. Best practices in this area are not governed by any one strategy that is generally acknowledged. Machine learning is about to revolutionize society as it is know it. It is crucial to give intelligent computers a moral compass now more than ever before because of how divided mankind is. Although machine learning has limitless potential, inappropriate usage might have detrimental long-term implications. It will think about how, for instance, earlier cultures built trust and improved social interactions via creative answers to many of the ethical issues that machine learning is posing now. 2023 IEEE.