A Review on Large Language Models: Architectures, Applications, Taxonomies, Open Issues and Challenges

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 26839 - 26874

Published: Jan. 1, 2024

Large Language Models (LLMs) recently demonstrated extraordinary capability, including natural language processing (NLP), translation, text generation, question answering, etc. Moreover, LLMs are a new and essential part of computerized processing, having the ability to understand complex verbal patterns generate coherent appropriate replies for situation. Though this success has prompted substantial increase in research contributions, rapid growth made it difficult overall impact these improvements. Since lot on is coming out quickly, getting tough get an overview all them short note. Consequently, community would benefit from but thorough review recent changes area. This article thoroughly overviews LLMs, their history, architectures, transformers, resources, training methods, applications, impacts, challenges, paper begins by discussing fundamental concepts with its traditional pipeline phase. It then provides existing works, history evolution over time, architecture transformers different resources methods that have been used train them. also datasets utilized studies. After that, discusses wide range applications biomedical healthcare, education, social, business, agriculture. illustrates how create society shape future AI they can be solve real-world problems. Then explores open issues challenges deploying scenario. Our aims help practitioners, researchers, experts pre-trained goals.

Language: Английский

---

A Review on Large Language Models: Architectures, Applications, Taxonomies, Open Issues and Challenges

Published: Sept. 27, 2023

Large Language Models (LLMs) recently demonstrated extraordinary capability in various natural language processing (NLP) tasks including translation, text generation, question answering, etc.Moreover, LLMs are new and essential part of computerized processing, having the ability to understand complex verbal patterns generate coherent appropriate replies a given context.Though this success has prompted substantial increase research contributions, rapid growth made it difficult overall impact these improvements.Since plethora on have been appeared within short time, is quite impossible track all get an overview current state area.Consequently, community would benefit from but thorough review recent changes area.This article thoroughly overviews LLMs, their history, architectures, transformers, resources, training methods, applications, impacts, challenges, etc.This paper begins by discussing fundamental concepts with its traditional pipeline phase.Then provides existing works, history evolution over architecture transformers different resources methods that used train them.The also demonstrates datasets utilized studies.After that, discusses wide range applications biomedical healthcare, education, social, business, agriculture.The study illustrates how create society shape future AI they can be solve real-world problems.Finally, explores open issues challenges deploy scenario.Our aims help practitioners, researchers, experts pre-trained goals.

Language: Английский

---

The role of ICT investment, digital financial inclusion, and environmental tax in promoting sustainable energy development in the MENA region: Evidences with Dynamic Common Correlated Effects (DCE) and instrumental variable-adjusted DCE

PLoS ONE, Journal Year: 2024, Volume and Issue: 19(5), P. e0301838 - e0301838

Published: May 6, 2024

His research investigates the interplay among investment in Information and Communication Technology [ICT], digital financial inclusion, environmental tax policies, their impact on progression of sustainable energy development within Middle East North Africa [MENA] region. Recognizing distinctive hurdles impeding advancement, effective policy formulation implementation MENA necessitate a comprehensive understanding these variables. Employing Dynamic Common Correlated Effects [DCE] model alongside an instrumental variable-adjusted DCE approach, this study explores relationship between ICT investment, tax, development. The facilitates analysis dynamic effects potential correlations, while addresses issues pertaining to endogeneity. results indicate that both promotion inclusion significantly positively Additionally, underscores importance fostering highlighting critical role interventions. Based findings, governmental prioritization initiatives for service integration is recommended bolster growth MENA. Furthermore, adoption efficient measures essential incentivize practices mitigate degradation. These recommendations aim create conducive environment region, contributing economic prosperity conservation.

Language: Английский

---

Enhancing cervical cancer diagnosis with graph convolution network: AI-powered segmentation, feature analysis, and classification for early detection

Multimedia Tools and Applications, Journal Year: 2024, Volume and Issue: 83(30), P. 75343 - 75367

Published: Feb. 16, 2024

Abstract Cervical cancer is a prevalent disease affecting the cervix cells in women and one of leading causes mortality for globally. The Pap smear test determines risk cervical by detecting abnormal cells. Early detection diagnosis this can effectively increase patient’s survival rate. advent artificial intelligence facilitates development automated computer-assisted diagnostic systems, which are widely used to enhance screening. This study emphasizes segmentation classification various cell types. An intuitive but effective technique segment nucleus cytoplasm from histopathological images. Additionally, handcrafted features include different properties generated distinct area. Two feature rankings techniques conducted evaluate study’s significant set. Feature analysis identifies critical pathological then divides them into 30, 40, 50 sets features. Furthermore, graph dataset constructed using strongest correlated features, prioritizes relationship between robust convolution network (GCN) introduced efficiently predict proposed model obtains sublime accuracy 99.11% 40-feature set SipakMed dataset. outperforms existing study, performing both simultaneously, conducting an in-depth analysis, attaining maximum efficiently, ensuring interpretability model. To validate model’s outcome, we tested it on Herlev highlighted its robustness 98.18%. results methodology demonstrate dependability effectively, early stages upholding significance lives women.

Language: Английский

---

Empowering agricultural security with IoT and deep learning driven real-time alert systems

International Journal of Information Technology, Journal Year: 2024, Volume and Issue: 16(7), P. 4627 - 4635

Published: July 16, 2024

Language: Английский

---

Sample imbalance remote sensing small target detection based on discriminative feature learning and imbalanced feature semantic enrichment

Expert Systems with Applications, Journal Year: 2025, Volume and Issue: unknown, P. 126753 - 126753

Published: Feb. 1, 2025

Language: Английский

---

Automated diagnosis of respiratory diseases from lung ultrasound videos ensuring XAI: an innovative hybrid model approach

Frontiers in Computer Science, Journal Year: 2024, Volume and Issue: 6

Published: Dec. 12, 2024

Introduction An automated computerized approach can aid radiologists in the early diagnosis of lung disease from video modalities. This study focuses on difficulties associated with identifying and categorizing respiratory diseases, including COVID-19, influenza, pneumonia. Methods We propose a novel method that combines three dimensional (3D) models, model explainability (XAI), Decision Support System (DSS) utilizes ultrasound (LUS) videos. The objective is to improve quality frames, boost diversity dataset, maintain sequence create hybrid 3D [Three-Dimensional Time Distributed Convolutional Neural Network-Long short-term memory (TD-CNNLSTM-LungNet)] for precise classification. proposed methodology involves applying morphological opening contour detection frame quality, utilizing geometrical augmentation dataset balance, introducing graph-based sequencing, implementing combining time-distributed CNN LSTM networks vast ablation study. Model ensured through heatmap generation, region interest segmentation, Probability Density Function (PDF) graphs illustrating feature distribution. Results Our TD-CNN-LSTM-LungNet attained remarkable accuracy 96.57% classifying LUS videos into pneumonia, normal, other classes, which above compared ten traditional transfer learning models experimented this eleven-ablation case reduced training costs redundancy. K-fold cross-validation accuracy-loss curves demonstrated generalization. DSS, incorporating Layer Class Activation Mapping (LayerCAM) heatmaps, improved interpretability reliability, PDF facilitated decision-making by boundaries. DSS facilitates clinical marker analysis, validation using algorithms highlights its impact reliable outcome. Discussion could assist accurately detecting comprehending patterns disorders.

Language: Английский

---

Artificial Intelligence-Driven Advancements in Otitis Media Diagnosis: A Systematic Review

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 99282 - 99307

Published: Jan. 1, 2024

Otitis Media (OM), predominantly affecting children, is a significant global health issue, with an estimated 360 million pediatric cases yearly worldwide. OM causes mild and moderate conductive hearing loss which can be disabling for young particularly during the first three years of life when brain growth rapid, resulting in poor speech language development, communication skills, increased vulnerability on entering school. therefore contributes to burden all-cause loss. This systematic review seeks provide comprehensive evaluation pre-trained Artificial Intelligence (AI) models, including both classical Machine Learning (ML) Deep (DL), context OM. proposes six research questions, it summarizes body across multiple domains, diversity quantity source material training testing otoscopy images, videos, tympanometry, methods used assess quality effectiveness real-time settings. In addition, aims insight into impact potential AI improving diagnosis cast light existing challenges, such as model interpretability, limited medical expert involvement, need knowledge discovery unanswered evolving landscape within this domain. The findings emphasize importance developing more interpretable models that incorporate still images tympanic membrane video recordings (with frames) maximize sensitivity specificity model. collaboration consumers professionals specialties (general practice, pediatricians, audiologists ear, nose, throat (ENT) surgeons) needed ensure applicability confidence these diagnostic digital support systems real-world healthcare

Language: Английский

---

Smart Agriculture by Leveraging AI and IoT for Intelligent Insect Identification and Control

Published: June 28, 2024

Language: Английский

---

Reimagining Otitis Media Diagnosis: A Fusion of Nested U-Net Segmentation with Graph Theory-Inspired Feature Set

Array, Journal Year: 2024, Volume and Issue: 23, P. 100362 - 100362

Published: Sept. 1, 2024

Language: Английский

---