Browse Items (2150 total)
Sort by:
-
Advancing Road Safety through Driver Drowsiness Detection Using Deep Learning Model
Driver drowsiness poses a significant threat to public safety, contributing to numerous road accidents and fatalities annually. Drowsy drivers exhibit characteristic changes in facial expressions and behaviors, including eye closure, head nodding, and yawning. These indicators can be detected through various techniques, including image processing, computer vision, and machine learning. This research investigates a promising approach: utilizing a ResNet-101 deep convolutional neural network (CNN) for driver drowsiness detection based on eye, head, and mouth states. The model was trained on a vast dataset of 2.2 million images, covering diverse driving conditions. Despite achieving a 69% accuracy, suggesting real-world potential, computational limitations restricted training to only a quarter of the data. This necessitates further research with larger datasets and increased resources to enhance accuracy and robustness. 2024 IEEE. -
Advancing the Evaluation of Oral Fluency in English for Specific Classrooms: Harnessing Natural Language Processing Tools for Enhanced Assessment
A crucial component of language learning and teaching is evaluating students' speaking abilities. Natural language processing (NLP) techniques have been employed recently in language assessment to automate the evaluation process and produce more impartial and reliable findings. In this study, we offer a speaking evaluation tool based on Natural Language Processing (NLP) that assesses a learner's speaking ability in real-time using cutting-edge algorithms. The instrument is altered to assess the fundamental facet of speaking skills - Fluency. As a result of the tool's immediate feedback, learners may pinpoint their areas of weakness and focus on honing their language abilities. The usefulness of the instrument was assessed through an intervention with a sample size of 30 students of the post-graduate students of a college in Pune, India. Python libraries, including random and re, were utilized to implement the algorithm. Data preprocessing involved accurate transcription of videos using an online tool and manual checking for corrections. Despite acknowledging limitations, such as potential biases in manually inserted hesitation markers, the study serves as a pivotal step toward automated fluency assessment, presenting exciting prospects for NLP and language education advancements. 2024 IEEE. -
AdvanDNN: Deep Neural Network Analysis of Neuroimaging for Identifying Vulnerable Brain Regions in Autism Spectrum Disorder
Exploring the neurological framework of autism spectrum disorder (ASD) presents a significant challenge due to its diverse manifestations and cognitive impacts. This study introduces an innovative deep learning approach, employing an advanced deep neural network (AdvanDNN) model to identify and analyze brain regions vulnerable to ASD. Utilizing the AAL116 brain atlas for anatomical standardization, our model processes a comprehensive set of neuroimaging data, including structural and functional MRI scans, to discern distinct neural patterns associated with ASD. The AdvanDNN model, with its robust deep learning architecture, was meticulously trained and validated, demonstrating a notable accuracy of 91.17% in distinguishing between ASD-affected individuals and controls. This marks an improvement over the state of the art, contributing a significant advance to the diagnostic processes. Notably, the model identified a pronounced anticorrelation in brain function between anterior and posterior regions, corroborating existing empirical evidence of disrupted connectivity within ASD neurology. The analysis further pinpointed critical regions, such as the prefrontal cortex, amygdala, and temporal lobes, that exhibit significant deviations from typical developmental patterns. These findings illustrate the potential of deep learning in enhancing early detection and providing pathways for intervention. The application of the AdvanDNN model offers a promising direction for personalized treatment strategies and underscores the value of precision medicine in addressing neurodevelopmental disorders. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024. -
AI and Machine Learning Applications in Predicting Energy Market Prices and Trends
The worldwide energy market is intricate and unstable, shaped by several aspects including geopolitical occurrences, supply-demand variations, and regulatory modifications. Precisely forecasting energy prices and trends is essential for stakeholders, such as energy producers, dealers, and policymakers. This study investigates the utilization of artificial intelligence (AI) and machine learning (ML) to improve energy price forecasting models. Conventional forecasting methods frequently fail to account for the dynamic and non-linear characteristics of energy markets; however, AI/ML techniques, including neural networks, decision trees, and reinforcement learning, provide enhanced prediction precision. By including external variables such as meteorological conditions and economic metrics, AI models can produce more accurate and useful insights. Case studies illustrate the effective implementation of AI in energy markets, showcasing its capacity to surpass traditional methods. This article addresses difficulties such as data quality and computing expenses while delineating potential developments in AI-driven energy market forecasts. The Authors, published by EDP Sciences. -
AI and Machine Learning Enabled Software Defined Networks
The telecommunications industry has not been exempt from the technology sectors massive artificial intelligence (AI) and machine learning (ML) boom in recent years. Artificial intelligence (AI) and machine learning (ML) provide advanced analytics and automation that are in line with modern networking concepts like software-defined networking (SDN) and software-defined wide-area networks (SD-WAN). Work is being done to determine how AI/ML can benefit SD-WAN and to demonstrate these benefits in a real SD-WAN network using a workable example. Modern ML techniques and algorithms are the extent of AI/ML. Todays Internet is under constant threat from DDoS (Distributed Denial of Service) attacks. As the volume of Internet traffic grows, its getting harder and harder to tell whats legitimate and whats malicious. The DDoS attack was detected using a machine learning approach that makes use of a Random Forest classifier. To better detect DDoS attacks, we tweak the Random Forest algorithm. The proposed machine learning approach outperforms, as demonstrated by our results. 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
AI Based Seamless Vehicle License Plate Recognition Using Raspberry Pi Technology
This research presents the implementation of an innovative Vehicle Management System designed specifically for the Christ University Project 'CampusWheels.' The system incorporates cutting-edge technologies, including YOLOv8 and Tesseract OCR, for robust license plate recognition. Addressing the unique challenges faced by Christ University in managing and securing vehicular movements within the campus, this project becomes crucial as the number of vehicles on campuses continues to grow. It not only provides an effective solution to these challenges but also introduces innovative methodologies, marking a significant departure from conventional campus management practices. The paramount importance of this project lies in its ability to enhance campus security through real-time vehicle monitoring and identification. The utilization of YOLOv8 for vehicle detection and Tesseract OCR for license plate recognition ensures a high level of accuracy in identifying and tracking vehicles entering and leaving the campus. This precision significantly contributes to the prevention of unauthorized vehicle access, a common security concern on educational campuses. Moreover, the system's ability to streamline traffic flow and improve efficiency in parking and access control addresses practical issues faced by campus administrators and security personnel. 2024 IEEE. -
AI Based Variable Step Size Block Least Mean Square Filter for Noise Cancellation System
Most of the Active Noise Cancellation (ANC) systems working properly in low-frequency noises only. To make it suitable for isolating high-frequency noise, it needs an additional circuit which consumes more energy. This problem is mitigated in this study by designing a Variable Step size Block Least Mean Square (VSBLMS) filter which is suitable for an effective noise cancellation system. VSBLMS filter is designed with RCA to make a design area efficient and it is designed with a novel adder to achieve high speed as well as less energy consumption. The proposed filter is designed and simulated using Xilinx ISE 13.2. The simulation results shows that the proposed VSBLMS filter design mitigates the unwanted noises in various frequency bands. The proposed VSBLMS reduces the energy consumption by 9.32%, 27.63%, 13.53%, 11.80%, 10.71 %, 13.14% and 9.26% when compared with state of the art methods. 2023 IEEE. -
AI Driven Finite Element Analysis on Spur Gear Assembly to Enhance the Fatigue Life and Minimized the Contact Pressure*
The major goal of the current research is to carry out mathematical and finite element analysis on spur gear assemblage to improve fatigue life as well as minimize contact pressure among contact teeth by modifying the face width of spur gear. AI automates FEA simulations and analyses, speeding up the design process. The investigation presented above was conducted using three separate 3d models of driving gear. The equivalent stress for the spur gear assembly of design-3 has decreased up to 13.45% in comparison to design-1, and the fatigue life has increased up to 81.59% at 600 N m, according to the results. Further AI models shall predict stress distribution, contact pressure, and other relevant factors in spur gear assemblies. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024. -
AI Healthcare Industry in Life Science Industry: A Perspective View
The main goal of this study is to look at how well the innovation system for AI healthcare technology is working in the life science business and find things that are getting in the way of progress. A lot of different types of research were used for this study. It combines both quantitative and qualitative data from tertiary studies, business-related written sources, and conversations with 21 experts and 25 life science management leaders to get new ideas. The results make it clear that innovation system performance is being held back by a lack of resources and poor communication from top healthcare experts about what they need to improve healthcare with AI technology innovations. The study says that to deal with these problems, policymakers need to make changes that increase the resources that are available and come up with clear goals and visions for how AI technology can improve healthcare. Using the socio-technical technological advancement System (TIS) approach in the healthcare setting, the study adds to our knowledge of how the innovation system works and how different parts of it affect each other. Overall, this study throws light on the complicated ways that innovation works in the life science field. It gives lawmakers, industry workers, and other interested parties useful information for pushing AI healthcare technology forward in a sociotechnical framework. 2024 IEEE. -
AI Sovereignty in Autonomous Driving: Exploring Needs and Possibilities for Overcoming Challenges
With the development of artificial intelligence, advancements in navigation systems for self-driving cars have become a new direction over the last decade. The inclusion of AI-driven actuators in autonomous vehicles has broken the barriers in terms of real-time high-quality data processing resources, accuracy of decisive actions and generalization of environment-action pairs. Upgradation from a car with no automation to a car with minimal to no human intervention has become a boon of AI, as it resolves most of the transportation problems on roads, including human error, lack of visibility in adverse weather conditions, tiredness of drivers in long journeys, etc. This study focuses on AI-enabled tasks, including object detection and identification, lane detection, notification for lane departure and reinforcement learning from the operational environment. However, there exist serious issues in deploying AI-empowered modules in autonomous cars, as the consumer rights to explain, trustworthiness, and reliability of the machine have not yet met the requirements. Our work explores the needs and prospects of AI sovereignty in autonomous driving by overcoming the aforementioned issues so that the healthy progress of technological society can take care of the future world. 2023 IEEE. -
AI-based Power Screening Solution for SARS-CoV2 Infection: A Sociodemographic Survey and COVID-19 Cough Detector
Globally, the confirmed coronavirus (SARS-CoV2) cases are being increasing day by day. Coronavirus (COVID-19) causes an acute infection in the respiratory tract that started spreading in late 2019. Huge datasets of SARS-CoV2 patients can be incorporated and analyzed by machine learning strategies for understanding the pattern of pathological spread and helps to analyze the accuracy and speed of novel therapeutic methodologies, also detect the susceptible people depends on their physiological and genetic aspects. To identify the possible cases faster and rapidly, we propose the Artificial Intelligence (AI) power screening solution for SARS- CoV2 infection that can be deployable through the mobile application. It collects the details of the travel history, symptoms, common signs, gender, age and diagnosis of the cough sound. To examine the sharpness of pathomorphological variations in respiratory tracts induced by SARS-CoV2, that compared to other respiratory illnesses to address this issue. To overcome the shortage of SARS-CoV2 datasets, we apply the transfer learning technique. Multipronged mediator for risk-averse Artificial Intelligence Architecture is induced for minimizing the false diagnosis of risk-stemming from the problem of complex dimensionality. This proposed application provides early detection and prior screening for SARS-CoV2 cases. Huge data points can be processed through AI framework that can examine the users and classify them into "Probably COVID", "Probably not COVID"and "Result indeterminate". 2021 The Authors. Published by Elsevier B.V. -
AI-Controlled Wind Turbine Systems: Integrating IoT and Machine Learning for Smart Grids
Advances in renewable energy technologies are pivotal in addressing the challenges posed by the depletion of traditional energy sources and their associated environmental impacts. Among these, wind energy stands out as a promising avenue, with wind turbine farms proliferating globally. However, the unpredictable nature of wind and intricate interplay between turbines necessitate innovative solutions for efficient operation and maintenance. This paper reviews advancements in intelligent control systems, notably those proposed by Smart Wind technologies. These systems leverage a network of sensors and IoT devices to gather real-time data, such as wind speed, temperature, and humidity, to optimize turbine performance. A significant focus is on turbines employing doubly-fed induction generators, which offer benefits like adjustable speed and consistent frequency operation. Their integration into smart grids introduces challenges concerning power system dynamics'security and reliability. This review delves into the dynamics, characteristics, and potential instabilities of such integrations, emphasizing the uncertainties in wind and nonlinear load predictions. A noteworthy finding is the rising prominence of artificial intelligence, particularly machine and deep learning, in predictive diagnostics. These methodologies offer costeffective, accurate, and efficient solutions, holding potential for enhancing power system stability and accuracy in the smart grid context. The Authors, published by EDP Sciences, 2024. -
AI-Driven Home Climate Optimization: The Role of ChatGPT in Enhancing AC Efficiency
The convergence of Internet of Things (IoT) and Artificial Intelligence (AI) has revolutionized home automation, yet traditional air-conditioning (AC) systems still struggle with energy inefficiency. Our research presents a novel solution, integrating AI, IoT, and user-centric design with ChatGPT, to optimize AC systems responsively to occupants' needs. Our methodology employs ChatGPT's capability to analyze historical data, discern patterns, and provide intelligent recommendations for AC operation. This transcends the functions of standard smart thermostats through AI-driven decision-making, optimizing every AC operational moment for both comfort and energy conservation. The system's foundation in data-driven decisions ensures alignment with external and internal conditions, enhancing energy efficiency and user comfort. 2024 IEEE. -
AI-enabled risk identification and traffic prediction in vehicular Ad hoc Networks
The proposed research presents a two-fold approach for advancing Vehicular Ad-Hoc Networks (VANETs). Firstly, it introduces a Residual Convolutional Neural Network (RCNN) architecture to extract real-time traffic data features, enabling accurate traffic flow prediction and hazard identification. The RCNN model, trained and tested on real- world data, outperforms existing models in both accuracy and efficiency, promising improved road safety and traffic management within VANETs. Secondly, the study introduces a Genetic Algorithm-enhanced Convolutional Neural Network (GACNN) routing algorithm, challenging traditional VANET routing methods with metaheuristic techniques. Experiments in various VANET network scenarios confirm GACNN's superior performance over existing routing protocols, marking a significant step toward more efficient and adaptive VANET traffic management. 2024 Author(s). -
AI-Enhanced IoT Data Analytics for Risk Management in Banking Operations
Using IoT data analytics in conjunction with artificial intelligence (AI) has the potential to improve banking operations' risk management. Sophisticated analytical methods are necessary for the detection and management of possible risks due to the increasing complexity and amount of data generated by the banking industry. This research proposes a novel method for analysing real-time data from IoT devices by employing artificial intelligence algorithms. The risks associated with financial transactions and operations can be better and more accurately assessed using this method. Through the integration of AI's pattern recognition, anomaly detection, and predictive modelling capabilities with the massive amounts of data generated by Internet of Things devices, this project aims to substantially enhance the efficacy and efficiency of risk management approaches in the banking sector. Research like this could lead to innovative solutions that make financial institutions more resistant to rising risks by enhancing decision-making, reducing operational weaknesses, and so on. 2024 IEEE. -
AI-Powered IoT Framework for Enhancing Building Safety through Stability Detection
The rapid urbanization and increasing structural complexities of modern buildings have heightened the need for advanced monitoring systems to ensure building safety. The research presents an AI-powered IoT framework that enhances building safety through advanced stability detection mechanisms. The proposed framework employs a novel algorithm, Ensemble Learning with IoT Sensor Data Aggregation (EnIoT-SDA), which integrates ensemble learning techniques with aggregated sensor data to provide accurate and real-time stability assessments of building structures. The effectiveness of EnIoT-SDA was evaluated through a comprehensive simulation analysis, comparing its performance against existing algorithms, including Support Vector Machine (SVM), Gradient Boosting Machines (GBM), and Fuzzy Logic Systems (FLS). Simulation metrics, such as accuracy, false positive rate, computational time, and detection latency, were used to assess and compare the algorithms' performance. The results demonstrated that EnIoT-SDA outperformed the existing methods in several key areas, offering improved accuracy and reduced detection latency, thus establishing its potential as a robust solution for building safety monitoring. The study underscores the significant advancements brought by integrating ensemble learning with IoT sensor data and highlights areas for future research and development in this domain. 2024 IEEE. -
Airline Twitter Sentiment Classification using Deep Learning Fusion
Since the advent of the Internet, the way people express their ideas and beliefs has undergone significant transformation. Blogs, online forums, product review websites and social media are increasingly the primary means of distributing information about new products. Twitter, in particular, is giving people a platform to air their views and opinions about a variety of events and products. In order to continually enhance the quantity and quality of their products and services, entrepreneurs constantly need input from their customers. Businesses are always looking for ways to increase the quality of their products and services. As a result, it's tough to understand the consumer's sentiments because of the large volume of data. In this research work, a Kaggle dataset of airline tweets for sentiment analysis was used. The dataset contains 11,540 reviews. We proposed an ensemble CNN, LSTM architecture for sentiment analysis. For comparison of the proposed system, LSTM alone also tested for similar dataset. LSTM was given an accuracy of 91% and the proposed ensemble framework with LSTM and CNN was given an accuracy of 93%. The experiments showed that the proposed model achieved better accuracy when compared to conventional techniques. 2022 IEEE. -
Algorithm trading and its application in stock broking services
Purpose: Algorithmic trading provides a more systematic approach to active trading than methods based on trader intuition or instinct. The aim of the study is to examine the level of awareness among the brokers when integrated with technology for the purpose of executing the trades. Design/Methodology: A self-administered and structured 350 questionnaires were designed and circulated to collect the preliminary information from the stock brokers operating in NSE and BSE within the geographical limits of Bangalore district using the Systematic Sampling method to obtain a sample size of 235. Awareness, Automated trading, Elimination of human error, portfolio management, tracking order, order placement were the critical variables observed to validate the hypothesis using Simple Percentage Analysis & Chi-Square Analysis using Statistical Analysis Software (SAS). Findings: It was found that there is robust association between the level of awareness of the mentioned technology in its application by the stock brokers of NSE and BSE operating in Bangalore. Portfolio management and automated trading are the highly associated application of Algorithmic trading among the stock brokerage services. Originality: Algorithmic trading makes use of complex formulas, combined with mathematical models and human oversight, to make decisions to buy or sell financial securities on an exchange. It can be used in a wide variety of situations including order execution, arbitrage, and trend trading strategies. Algorithmic traders often make use of high-frequency trading technology, which can enable a firm to make tens of thousands of trades per second. The Authors, published by EDP Sciences. -
Algorithmic Trading: Financial Markets Using Artificial Intelligence
This research study gives an in - depth view of the recent developments in the fields of Machine Learning (ML) and Reinforced Learning (RL) techniques as they are related to various models for forecasting and systems for financial trading. The practical usage of deep learning models, that incorporates Neural Networks such as Recurrent, Convolutional along with hybrid models integrating genetic algorithms with LSTM networks, for forecasting the stock market patterns as well as bank failures, and fluctuations in exchange rate which is addressed in this study in an in - depth review analysis of the latest literature. In addition to this it also investigates how trading algorithm performance as well as risk management can be enhanced by applying techniques of deep reinforcement learning. This study also demonstrates the enhanced, efficacy, precision and the profitability achieved by using these artificial intelligence methods as compared with conventional economic modelling and detailed technical study models by analysing a number of stock markets and different kinds of assets. 2024 IEEE. -
Alkali-Activated Materials - A Review for Sustainable Construction
New, sustainable low-Carbon Dioxide (CO2) construction materials must be developed for the global building sector to decrease its environmental impact. During the last several decades, Alkali-activated Materials (AAMs) is a Portland cement-free form, have been intensively researched as a potential alternative for ordinary Portland cement concrete (OPCC), with the objective of lowering CO2 emissions while repurposing a large volume of industrial waste by-products. The suitability of using AAMs made up of industrial waste by-products such as blast furnace slag (BFS), calcined clay (metakaolin), and fly ash (FA) was investigated in this study utilizing a performance-based approach that was unaffected by binder chemistry, history, or environmental effect, Binder paste microstructural assessment and influence on engineering effectiveness, including fresh and hardened characteristics of these materials, In the Viewpoints area, we analyze specific premature phase and long-phase performance of AAMs, as well as Upcoming scientific breakthroughs are also discussed in the Viewpoints section. 2022 American Institute of Physics Inc.. All rights reserved.