Browse Items (2150 total)
Sort by:
-
PE-v-SVR based Architecture to Predict and Prevent Low and Slow-Rate DDoS Attacks using Machine Learning
Distributed Denial of Service (DDoS) attacks continue to emerge; low and slow attacks pose a serious threat. These small-scale attacks often evade traditional security protections and increase the risk of long-term outages and loss of service. Our research aims to develop effective predictive models and strategic defences to detect and mitigate slow DDoS attacks. The proposed model combines Power Spectral entropy and V-Support Vector Regression. More importantly, the version achieves the first-class error price in the variety of zero to at least one, demonstrating its effectiveness in detecting and predicting DDoS attacks. Research results show the effectiveness of the proposed design using PSD (power spectral density) entropy and V-SVR. The best mean square error obtained further confirms the ability of the model in this context. V-SVR in low and sluggish DDoS assaults. 2024 Bharati Vidyapeeth, New Delhi. -
Regression Analysis on Macroeconomic Factors and Dividend Yield on Bank Nifty Index Returns
The study has examined an impact of macroeconomic variables and dividend yield on Bank NIFTY Index. It analyses the relationship amongst macroeconomic variables and dividend yield. The study used quarterly data from 1 January 2010 to 31 December 2019. It employed statistical measures like regression analysis to analyse the impact of independent variables (macroeconomic factors and dividend yield) on the dependent variable (Bank NIFTY returns) and multicollinearity tests to understand the relationship amongst the independent variables. The observations concluded that GDP, government bond yield and dividend yield have a significant impact on Bank NIFTY returns but CPI does not have a significant impact on Bank NIFTY returns. We can also conclude that all the independent variables are not correlated to each other. The study suggested to policy makers, in India, that they should maintain economic stability through policies of growth that will eventually boost the banking sector and the economy. 2021, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
Managing with Machines: A Comprehensive Assessment on the Use of Artificial Intelligence in Organizational Perspectives
This complete study, delves into the multifaceted impacts of artificial Intelligence (AI) inside organizational settings, highlighting its ability and demanding situations. The investigation spans numerous aspects along with AI-driven customer relationship management (CRM), employee productivity, and overall performance enhancement thru AI. By analyzing distinct AI applications and methodologies across different organizational functions, this studies presents insights into how AI can transform industries, decorate CRM, improve employee productiveness, and foster sustainable development. Despite the promising programs, the study also addresses the pitfalls and enormous hesitancy in AI adoption due to disasters in some high-profile AI projects. The paper underscores the significance of strategic AI integration, context-consciousness, and the want for organizational readiness to leverage AI's full capability whilst aligning with the Sustainable improvement goals (SDGs). 2024 IEEE. -
Beyond Automation: Understanding the Transformational Capabilities of AI in Management
The investigation explores at the various ways that artificial intelligence (AI) is affecting management techniques. The study highlights the dichotomy between automation and augmentation, highlighting how artificial intelligence (AI) can replace human work through automation, but its ultimate use in augmenting human capabilities (augmentation) leads to better organisational performance. This analysis reveals how AI-driven tactics enhance operational efficiency, decision-making, and productivity by synthesising research findings from a variety of domains, including manufacturing, banking, municipal sectors, and remote work environments. It also looks at how AI may change management through big data and data analytics, recommending a shift to an integrated strategy that combines automation and human understanding to promote creativity and long-term growth. 2024 IEEE. -
Deep Learning-Based Optimised CNN Model for Early Detection and Classification of Potato Leaf Disease
After rice and wheat, potatoes are the third-largest crop grown for human use worldwide. The different illnesses that can harm a potato plant and lower the quality and quantity of the yield cause potato growers to suffer significant financial losses every year. While determining the presence of illnesses in potato plants, consider the state of the leaves. Early blight and late blight are two prevalent illnesses. A certain fungus causes early blight, while a specific bacterium causes late blight. Farmers can avoid waste and financial loss if they can identify these diseases early and treat them successfully. Three different types of data were used in this study's identification technique: healthy leaves, early blight, and late blight. In this study, I created a convolutional neural network (CNN) architecture-based system that employs deep learning to categorise the two illnesses in potato plants based on leaf conditions. The results of this experiment demonstrate that CNN outperforms every task currently being performed in the potato processing facility, which needed 32 batch sizes and 50 epochs to obtain an accuracy of about 98%. 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
Post Covid Scenario Effective E-Mentoring System in Higher Education
During Covid-19 pandemic many people and institutions preferred online coaching instead of in person education. The problem with online is that it will be difficult to carry on interconnections between students and professors in that environment. The main constraint for conducting online session is that the people in remote areas may find a difficulty to connect to online sessions having network issues. Electronic mentoring (e-mentoring) is implemented like a website in which the mentor and mentee can communicate with each other. With the help of this mentoring the project can provide a best solution for both the mentor and mentee. They can communicate with each other with the help of online platform and even with the help of emails.This proposed method will help them to keep the track of their academic progress and achievements of students. This article mainly focus on the mentoring through physical and virtual environment in which the mentee will be interacting with the mentor to know the progress of their academics. This article discusses about the website which is developed to fulfill the needs of the student and it discusses about the various stages of development that helped in building the website. Students can share their difficulties and their achievements with the mentor who are assigned for them particularly. In future planning to implement artificial intelligence technique to online mentoring process, this is for the betterment of student's growth. 2023 IEEE. -
Drought PredictionA Comparative Analysis of Supervised Machine Learning Techniques
Drought is a natural phenomenon that puts many lives at risk. Over the last decades, the suicide rate of farmers in the agriculture sector has increased due to drought. Water shortage affects 40% of the world's population and is not to be taken lightly. Therefore, prediction of drought places a significant role in saving millions of lives on this planet. In this research work, six different supervised machine learning (SML) models namely support vector machine (SVM), K-nearest neighbor (KNN), decision tree (DT), convolutional neural networks (CNNs), long short-term memory (LSTM), and recurrent neural networks (RNNs) are compared and analyzed. Three dimensionality reduction techniques principal component analysis (PCA), linear discriminant analysis (LDA), and random forest (RF) are applied to enhance the performance of the SML models. During the experimental process, it is observed that RNN model yielded better accuracy of 88.97% with 11.26% performance enhancement using RF dimensionality reduction technique. The dataset has been modeled using RNN in such a way that each pattern is reliant on the preceding ones. Despite the greater dataset, the RNN model size did not expand, and the weights are observed to be shared between time steps. RNN also employed its internal memory to process the arbitrary series of inputs, which helped it outperform other SML models. 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
Decoding Big Data: The Essential Elements Shaping Business Intelligence
In today's Business Intelligence (BI) world, Big Data Analytics integration has become critical, transforming company strategy and decision-making processes. This study investigates the complex influence of Big Data on business intelligence, focusing on important drivers of this transition. It investigates how Big Data's improved data processing capabilities, integration of advanced analytics techniques such as machine learning, and real-time data insights enable businesses to make more informed decisions and achieve a competitive advantage. Furthermore, the paper emphasizes the importance of personalized consumer insights, operational savings, and strategic benefits obtained from predictive analytics when adopting Big Data for BI. 2024 IEEE. -
Multi-Agent Deep Reinforcement Learning for Hybrid Motion Planning in Dynamic Environments
This research presents a novel approach to addressing the challenges of gesture forecasting in impenetrable and dynamic atmospheres by integrating a hybrid algorithm within a multi-agent system framework. Traditional methods such as Force-based motion planning (FMP) & deep reinforcement learning (RL) often struggle to handle complex scenarios involving multiple autonomous agents due to their inherent limitations. To overcome these challenges, we propose a hybrid algorithm that seamlessly combines the strengths of RL and FMP while leveraging the coordination capabilities of a multi-agent system. By integrating this hybrid algorithm into a multi-agent framework, we demonstrate its effectiveness in enabling multiple agents to navigate densely populated environments with dynamic obstacles. Through extensive simulation studies, we illustrate the superior performance of our approach compared to traditional methods, achieving higher success rates and improved efficiency in scenarios involving simultaneous motion planning for multiple agents. A hybrid motion planning algorithm is also introduced in this very research. Performance Comparison of Hybrid Algorithm, Deep RL, and FMP are also discussed in the result section. This research paves the way for the development of robust and scalable solutions for motion planning in real-world applications such as collaborative robotics, autonomous vehicle fleets, and intelligent transportation systems. 2024 IEEE. -
Effect of pH on the structural and optical properties of cobalt oxide nanoparticles synthesized by hydrothermal method
The paper focuses on the synthesis and characterization of cobalt oxide nanoparticles synthesized under different alkaline pH of the precursor solution by hydrothermal method. Cubic spinel Co3O4 crystallites were observed by X-ray diffraction pattern (XRD) and Raman spectrum. The crystallite size decreases as the pH value increases. The absorption spectrum exhibited two broad bands which are in good agreement with the cobalt oxide band structure. The change in bandgap was observed with pH of the precursor solution in agreement with size effects. Photoluminescence (PL) spectra consist of a broad emission with different peaks which are due to point defects. 2022 -
Multi-Layer Ensemble Deep Reinforcement Learning based DDoS Attack Detection and Mitigation in Cloud-SDN Environment
Cloud computing (CC) remains as a promising environment which offers scalable and cost effectual computing facilities. The combination of the SDN technique with the CC platform simplifies the complexities of cloud networking and considerably enhances the scalability, manageability, programmability, and dynamism of the cloud. This study introduces a novel Multi-Layer Ensemble Deep Reinforcement Learning based DDoS Attack Detection and Mitigation (MEDR-DDoSAD) technique in Cloud-SDN Environment. The major aim of the presented technique lies in the recognition of DDoS attacks from the cloud-SDN platform. The MEDR-DDoSAD technique transforms the input data into images and the features are derived via deep convolutional neural network based Xception model. 2022 IEEE. -
Comparative Analysis of Predictive Models for Customer Churn Prediction in the Telecommunication Industry
To determine the best model for churn prediction in the telecom industry, this paper compares 11 machine learning algorithms namely Logistic Regression, Support Vector Machine, Random Forest, Decision Tree, XGBoost, LightGBM, Cat Boost, AdaBoost, Extra Trees, Deep Neural Network, and Hybrid Model (MLPClassifier). It also aims to pinpoint the top three factors that lead to customer churn and conducts customer segmentation to identify vulnerable groups. The results indicate that the Logistic Regression model performs the best, with an F1 score of 0.6215, 81.76% accuracy, 68.95% precision, and 56.57% recall. The top three attributes that cause churn are found to be tenure, Internet Service Fiber optic, and Internet Service DSL; conversely, the top three models in this article that perform the best are Logistic Regression, Deep Neural Network, and AdaBoost. The K means algorithm is applied to establish and analyze four different customer clusters. This study has effectively identified customers that are at risk of churn and may be utilized to develop and execute strategies that lower customer attrition. 2024 IEEE. -
Impact of Urban Environmental Quality, Residential Satisfaction, and Personality on Quality of Life among Residents of Delhi/NCR
Environmental quality and Sustainability seek to preserve, enhance and protect our environmental resources that directly aim at providing an amicable quality of life and sustainable development for the upcoming generations. Considering the hazardous environmental urban quality in Delhi NCR, air pollution is the topmost factor deteriorating health of the population in general. The urban air database by WHO reports Delhi exceeding the maximum PM10 limit by almost 10-times at 292 ?g/m3. Noticing that an individual's surroundings have an enormous value in human lives, the study aimed at understanding the impact of urban environmental quality, residential satisfaction, and personality on the quality of life among residents of Delhi NCR. In addition, we also track the environmental worldviews to attitudes on pro-environmental behavior in understanding sustainability. The results from the SEM model indicated that one index rise in RESS lead to a fall in quality of life by 0.029-point value whereas one index rise in personality could enhance the quality of life by 0.15-point value. Pro-Environmental Behaviors and Urban Environmental factors did not showcase any significant impact on the quality of life. The Electrochemical Society. -
A review on feature selection algorithms
A large number of data are increasing in multiple fields such as social media, bioinformatics and health care. These data contain redundant, irrelevant or noisy data which causes high dimensionality. Feature selection is generally used in data mining to define the tools and techniques available for reducing inputs to a controllable size for processing and analysis. Feature selection is also used for dimension reduction, machine learning and other data mining applications. A survey of different feature selection methods are presented in this paper for obtaining relevant features. It also introduces feature selection algorithm called genetic algorithm for detection and diagnosis of biological problems. Genetic algorithm is mainly focused in the field of medicines which can be beneficial for physicians to solve complex problems. Finally, this paper concludes with various challenges and applications in feature selection. Springer Nature Singapore Pte Ltd 2019. -
Application of LSTM Model for Western Music Composition
Music is one of the innate creative expressions of human beings. Music composition approaches have always been a focal point of music-based research and there has been an increasing interest in Artificial Intelligence (AI) based music composition methods in recent times. Developing an accurate algorithm and neural network architecture is imperative to the success of an AI-based approach to music composition. The present work explores the composition of western music through neural network using a Long Short-Term Memory (LSTM) algorithm. Compositions from seminal western composers such as J.S. Bach, W.A. Mozart, L.V. Beethoven, and F. Chopin were used as the dataset to train the neural network. Seven compositions were generated by the LSTM model and these outputs were presented to a group of thirty volunteers between 18-24 years of age. They were surveyed to identify the music piece as composed by a human or AI and how interesting they found the melodies of each piece. It was found that the LSTM model generated compositions that were thought to be made by a human and create melodies of interest from the perception of the volunteers. It is expected that through this study, more AI-based composition approaches can be developed which encompass more and more of the musical phenomenon. 2022 IEEE. -
Seismic Activity-based Human Intrusion Detection using Deep Neural Networks
Human intrusion detection systems have found their applications in many sectors including the surveillance of critical infrastructures. Generally, these systems make use of cameras mounted on strategic locations for surveillance purposes. Cameras based detection systems are limited by line-of-sight, need regular maintenance and dependence of electricity for operations. These are all detrimental to the efficiency of these detection systems, especially in remote locations. To overcome these challenges, intrusion detection systems based on seismic activities have been in use. The seismic activities collected through geophones from the human footfalls can act as the input for these detection systems. This also poses a challenge as the data generated by the geophones for the seismic activities produced from footsteps are not always identical and hence not accurate. In this proposed work, a Deep Neural Network based approach has been used on the dataset collected from the geophones to effectively predict the presence of humans. The results gave a success rate with 94.86% accuracy with testing data and 92.00% accuracy with real-time data with the geophones deployed on an area covered with grass. 2022 IEEE. -
Pneumonia Detection using Ensemble Transfer Learning
Pneumonia is among the most common illnesses and causes to death among the young children worldwide. It is more serious in under-developed countries as it is hard to diagnose due to the absence of specialists. Chest X-ray images have essentially been utilized in the diagnosis of this disease. Examining chest X-rays is a difficult task, even for an experienced radiologist. Information Technology, especially Artificial Intelligence, have started contributing to accurate diagnosis of pneumonia from chest X-ray images. In this work, we used deep learning, transfer learning, and ensemble voting to increase the accuracy of pneumonia detection. The models utilized are VGG16, MobileNetV2, and InceptionV3, all pre-trained on ImageNet, and used the Kaggle RSNA CXR image dataset. The results from these models are ensembled using the weighted average ensemble approach to achieve better accuracy and obtained 98.63% test accuracy. The results are promising, and the proposed model can assist doctors in detecting pneumonia quickly and accurately from Chest X-Ray. 2022 IEEE. -
Comparison of Full Training and Transfer Learning in Deep Learning for Image Classification
The deep learning algorithms on a small dataset are often not efficient for image classification problems. Make use of the features learned by a model trained on large similar dataset and saved for future reference is a method to solve this problem. In this work, we present a comparison of full training and transfer learning for image classification using Deep Learning. Three different deep learning architectures namely MobileNetV2, InceptionV3 and VGG16 were used for this experiment. Transfer learning showed higher accuracy and less loss than full-training. According to transfer learning results, MobileNetV2 model achieved 98.96%, InceptionV3 model achieved 98.44% and VGG16 model achieved 97.405 as highest test accuracies. The full-trained models did not achieve as much accuracy as that of transfer learning models on the same dataset. The accuracies achieved by full-training for MobileNetV2, InceptionV3 and VGG16 are 79.08%, 73.44% and 75.62% respectively. 2021, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
Heart Disease Prediction Using Ensemble Voting Methods in Machine Learning
Heart disease is the leading cause of mortality globally according to the World Health Organization. Every year, it results in millions of mortalities and thus billions of dollars in economic damage throughout the world. Many lives can be saved if the disease is detected early and accurately. The typical methods to predict or diagnosis heart diseases require medical expertise. Such facilities and experts are relatively expensive and not very commonly available in under developed and developing countries. Recent times, much research is done on leveraging technology for the prediction as well as diagnosis of heart diseases. Machine Learning techniques have been extensively deployed as quick, inexpensive, and noninvasive ways for heart disease identification. In this work, we present a machine learning approach in detecting heart disease using a dataset that contains vital body parameters. We used seven different models and combined them with Soft-Voting and Hard-Voting ensemble approaches to improve accuracy in 7-model and various 5-model combinations. The ensemble combinations of 5 models achieved the highest test accuracy score of 94.2%. 2022 IEEE. -
Pothole Detection and Powertrain Control for Vehicular Safety
A new era of automotive technology has begun with the rapid advancement of electric vehicles (EVs), which promise efficiency and sustainability. With electric vehicles (EVs) becoming an integrated part of the traction systems, there is a growing need for novel safety and performance-enhancing features. The development of an Adaptive Cruise Control (ACC) system for autonomous powertrain control and pothole detection in electric vehicles is examined in this paper. The paper focuses on integrating an intelligent system that can detect potholes and autonomously regulate the powertrain to improve both the driving experience and safety of electric vehicles. The system makes use of Jetson Nano as the processing unit for regulation of the EV powertrain. This board enables quick and accurate reactions to changing road conditions by facilitating real-time data analysis and decision-making. The powertrain regulation will be performed by controlling the acceleration and braking signal provided to the powertrain. 2024 IEEE.