Deep Learning for Hand Gesture Recognition in Virtual Museum Using Wearable Vision Sensors

IEEE Sensors Journal, Journal Year: 2024, Volume and Issue: 24(6), P. 8857 - 8869

Published: Jan. 23, 2024

Hand gestures facilitate user interaction and immersion in virtual museum applications. These allow users to navigate exhibitions, interact with artifacts, control environments naturally intuitively. This study introduces a deep learning-driven approach for hand gesture recognition using wearable vision sensors designed interactive environments. The proposed employs an image-based feature extraction strategy that focuses on capturing five partial occupancy areas of the hand. Notably, learning bidirectional Long Short-Term Memory (Bi-LSTM) model is adopted construct effective identification. bi-directionality Bi-LSTM enables it capture dependencies both forward backward directions, providing more comprehensive understanding temporal relationships data. nature allows better dynamics complexities motions, leading improved accuracy robustness. performance evaluation includes experiments publicly available datasets, considering real scenarios. results highlight Bi-LSTM-based approach’s superiority by accurately distinguishing various gestures. experimental findings demonstrate combining area ratios classification robust diverse effectively discriminates between similar actions, such as slide left right classes. Additionally, shows promising detection compared conventional machine models state-of-the-art methods. presented enhancing experiences.

Language: Английский

---

BankNet: Real-Time Big Data Analytics for Secure Internet Banking

Big Data and Cognitive Computing, Journal Year: 2025, Volume and Issue: 9(2), P. 24 - 24

Published: Jan. 26, 2025

The rapid growth of Internet banking has necessitated advanced systems for secure, real-time decision making. This paper introduces BankNet, a predictive analytics framework integrating big data tools and BiLSTM neural network to deliver high-accuracy transaction analysis. BankNet achieves exceptional performance, with Root Mean Squared Error 0.0159 fraud detection accuracy 98.5%, while efficiently handling rates up 1000 Mbps minimal latency. By addressing critical challenges in operational efficiency, establishes itself as robust support system modern banking. Its scalability precision make it transformative tool enhancing security trust financial services.

Language: Английский

---

DPMS: Data-Driven Promotional Management System of Universities Using Deep Learning on Social Media

Applied Sciences, Journal Year: 2023, Volume and Issue: 13(22), P. 12300 - 12300

Published: Nov. 14, 2023

SocialMedia Marketing (SMM) has become a mainstream promotional scheme. Almost every business promotes itself through social media, and an educational institution is no different. The users’ responses to media posts are crucial successful campaign. An adverse reaction leaves long-term negative impact on the audience, conversion rate falls. This why selecting content share one of most effective decisions behind success paper proposes Data-Driven Promotional Management System (DPMS) for universities guide selection appropriate promote which more likely obtain positive user reactions. main objective DPMS make Social Media (SMM). novel uses well-engineered optimized BiLSTM network, classifying sentiments about different university divisions, with stunning accuracy 98.66%. average precision, recall, specificity, F1-score 98.12%, 98.24%, 99.39%, 98.18%, respectively. innovative (PMS) increases impression by 68.75%, reduces 31.25%, 18%. In nutshell, proposed first management system universities. It demonstrates significant potential improving brand value increasing intake rate.

Language: Английский

---

Deep-Hill: An Innovative Cloud Resource Optimization Algorithm by Predicting SaaS Instance Configuration Using Deep Learning

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 92573 - 92584

Published: Jan. 1, 2024

The integration of Artificial Intelligence (AI) services within the framework Software-as-a-Service (SaaS) cloud architecture has significantly permeated our everyday routines. These AI diverge from traditional applications by offering a more personalized user experience. That is why predefined instance configuration not an optimal approach for these applications. challenge further compounded unpredictable nature demand, making resource allocation to instances complex task. This paper introduces innovative algorithm, termed Deep-Hill, designed enhance through precise prediction SaaS configurations. It combination 5-layer Deep Neural Network (DNN) and Hill-Climbing algorithm. unique classifies in one five classes with 96.33% accuracy, 90.83% precision, 90.96% recall, 90.86% F1-score. On average, it reduces number active hosts four, contributing 13.33% less power consumption. remarkable performance Deep-Hill algorithm underscores its potential set new benchmark optimization resources. paves way cost-effective applications, marking significant step forward evolution computing.

Language: Английский

---

RAP-Optimizer: Resource-Aware Predictive Model for Cost Optimization of Cloud AIaaS Applications

Electronics, Journal Year: 2024, Volume and Issue: 13(22), P. 4462 - 4462

Published: Nov. 14, 2024

Artificial Intelligence (AI) applications are rapidly growing, and more joining the market competition. As a result, AI-as-a-service (AIaaS) model is experiencing rapid growth. Many of these AIaaS-based not properly optimized initially. Once they start large volume traffic, different challenges revealing themselves. One maintaining profit margin for sustainability AIaaS application-based business model, which depends on proper utilization computing resources. This paper introduces resource award predictive (RAP) cost optimization called RAP-Optimizer. It developed by combining deep neural network (DNN) with simulated annealing algorithm. designed to reduce underutilization minimize number active hosts in cloud environments. dynamically allocates resources handles API requests efficiently. The RAP-Optimizer reduces physical an average 5 per day, leading 45% decrease server costs. impact was observed over 12-month period. observational data show significant improvement utilization. effectively operational costs from USD 2600 1250 month. Furthermore, increases 179%, 600 1675 inclusion dynamic dropout control (DDC) algorithm DNN training process mitigates overfitting, achieving 97.48% validation accuracy loss 2.82%. These results indicate that enhances management cost-efficiency applications, making it valuable solution modern

Language: Английский

---

PersoNet: A Novel Framework for Personality Classification-Based Apt Customer Service Agent Selection

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 25200 - 25214

Published: Jan. 1, 2024

Personality classification has garnered significant interest in psychology, computational social science, and Machine Learning (ML) due to its wide-ranging applications. This paper presents PersoNet, an innovative framework developed identify personality types using the Myers-Briggs Type Indicator (MBTI), aimed at enhancing customer service experiences by matching customers with suitable support agents. PersoNet employs a Bidirectional Long Short-Term Memory (BiLSTM) neural network architecture achieved impressive accuracy of over 93.98%. Our extensive experiments MBTI dataset reveal that BiLSTM effectively captures both temporal dependencies semantic subtleties textual data, contributing this high level accuracy. Consequently, can accurately select agents who match personalities, achieving Customer Satisfaction Rate (CSR) 97.82%—a notable improvement 20.25% CSR based on our experimental data. These results establish as cutting-edge tool classification, surpassing existing methods efficiency markedly quality.

Language: Английский

---

System development for enhancing social media advertisement engagement through XLNet-based personality classification

Eastern-European Journal of Enterprise Technologies, Journal Year: 2024, Volume and Issue: 4(2 (130)), P. 40 - 51

Published: Aug. 30, 2024

This research focuses on addressing the challenge of implementing personalized advertisements in retail industry, where existing methods often face complexities that hinder their swift and large-scale adoption. The primary objective this study was to develop a scalable efficient social media advertisement personalization system by employing advanced personality classification techniques. utilizes myPersonality dataset, grounded Big 5 OCEAN traits theory, accurately classify user personalities. By integrating XLNet model, optimized for classification, achieves accuracy 97.47 %, with precision, recall, F1-Score values 0.95, 0.94, respectively. findings demonstrate advertisements, driven classified traits, significantly enhance interaction rates, showing 24 % improvement over generalized advertisements. engagement suggests can effectively personalize resonate more deeply users, fostering stronger connections between users advertised content. proposed system's high improved rates make it valuable addition current marketing strategies, enhancing both conversion rates. innovative approach has potential transform advertising, making effective widely adoptable within sector

Language: Английский

---