Fake news detection using recurrent neural network based on bidirectional LSTM and GloVe

Social Network Analysis and Mining, Год журнала: 2024, Номер 14(1)

Опубликована: Фев. 10, 2024

In the world of technology, electronic and technical development fields communication internet has increased, which caused a renaissance in virtual world. This greatly impacted communities for ease speed information transfer through social media platforms, making these platforms likable easy to use. The network faces major challenges due its extensive As result, many people have become involved cybercrimes. There are accounts on that malicious. Platforms networking online, such as Facebook Twitter, allow all users freely generate consume massive volumes material regardless their traits. While individuals businesses utilize this gain competitive edge, spam or phony create important data. According estimates, 1 200 posts contain spam, 21 tweets spam. problem was centered around accuracy detecting false news correcting it preventing dissemination before spread network. A new method is given based improving detection system; level improvement significant preprocessing stage where Glove used, an unsupervised learning algorithm developed by researchers at Stanford University aiming word embeddings aggregating global co-occurrence matrices from corpus. basic idea behind GloVe embedding derive relationship between words statistics. proposed contains deep algorithms convolutional neural (CNN), (DNN), long short-term memory (LSTM). RNN with using Curpos fake dataset enhance system, sequential processes classification, highest 98.974%.

Язык: Английский

---

Machine and Deep Learning Methods for Concrete Strength Prediction: A Bibliometric and Content Analysis Review of Research Trends and Future Directions

Applied Soft Computing, Год журнала: 2024, Номер 164, С. 111956 - 111956

Опубликована: Июль 8, 2024

Язык: Английский

---

A Deep Learning Based Expert Framework for Portfolio Prediction and Forecasting

IEEE Access, Год журнала: 2024, Номер 12, С. 103810 - 103829

Опубликована: Янв. 1, 2024

Stock market forecasting involves predicting fluctuations and trends in the value of financial assets, utilizing statistical machine learning models to analyze historical data for insights into future behavior. This practice aids investors, traders, institutions, governments making informed decisions, managing risks, assessing economic conditions. Forecasting markets is difficult due intricate interplay global economics, politics, investor sentiment, it inherently unpredictable. study introduces a Deep Learning based Expert Framework Market (Portfolio prediction) called DLEF-SM. The methodology begins with an improved jellyfish-induced filtering (IJF-F) technique preprocessing, effectively analyzing raw eliminating artifacts. To address imbalanced enhance quality, pre-trained convolutional neural network (CNN) architectures, VGGFace2 ResNet-50, are used feature extraction. Additionally, black widow optimization (IBWO) algorithm designed selection, reducing dimensionality preventing under-fitting. For precise stock predictions, integrate deep reinforcement artificial (DRL-ANN) proposed. Simulation outcomes reveal that proposed framework achieves maximum accuracy, reaching 99.562%, 98.235%, 98.825% S&P500-S, S&P500-L, DAX markets, respectively.

Язык: Английский

---

Exploring the trend of recognizing apple leaf disease detection through machine learning: a comprehensive analysis using bibliometric techniques

Artificial Intelligence Review, Год журнала: 2024, Номер 57(2)

Опубликована: Янв. 30, 2024

Abstract This study’s foremost objectives were to scrutinize how unexpected weather affects agricultural output and assess well AI-based machine learning deep leaning algorithms work for spotting apple leaf diseases. The researchers carried out a bibliometric study obtain understanding of the current research trends, citation patterns, ownership partnership arrangements, publishing other parameters related early identification illnesses. Comprehensive interdisciplinary scientific maps are limited because syndrome recognition is not restricted any solitary arena research, despite fact that there have been many studies on By employing scientometric technique 109 publications from Scopus database published between 2011 2022, this attempted condition area combine knowledge frameworks. To find important journals, authors, nations, articles, topics, used automated processes VOSviewer Biblioshiny software. Patterns trends discovered using counts, social network analysis, co-citation studies.

Язык: Английский

---

Wetlands contribution and linkage to support SDGs, its indicators and targets‐ A critical review

Sustainable Development, Год журнала: 2024, Номер unknown

Опубликована: Март 27, 2024

Abstract This study marks one of the pioneering efforts to compile comprehensive information on Ramsar sites globally. It delves into significance wetlands and designation across various countries, incorporating a concise exploration utilization Unmanned Aerial Vehicles (UAVs) for wetland monitoring assessment. Additionally, conducts comparative evaluation sites, analyzing their percentage area overall coverage worldwide. Incorporating Scientometric analysis utilizing Scopus database, features co‐occurrence map, thematic evolution trend, country collaboration map. Emphasizing interconnection between Sustainable Development Goals (SDGs), particularly SDG6 (Clean Water & Sanitation), SDG12 (Responsible Consumption Production), SDG13 (Climate‐Action), SDG14 (Life Below Water) SDG15 Land), associated targets indicators. Targets such as 6.1, 6.2, 6.3, 6.4, 6.5, 6a, 6b SDG‐6, 12.1, 12.2, 12.4 SDG‐12, 13.2, 13.3 SDG‐13 align with management conservation. Moreover, it affirms role in supporting 14.1, 14.2, 14.3, 14.4, 14.5, 14.6, 14a‐c SDG‐14, 15.1, 15.5, 15.6, 15.7, 15.8, 15.8 SDG‐15. Policies, regulations plans different countries relevant establishing relationship SDGs are discussed details. The offers detailed these targets, elucidating indicator types each SDG target. By doing so, provides valuable insights future researchers policymakers, underlining indispensable contribution direct indirect fulfillment 6,12,13,14,15 17.

Язык: Английский

---

Extraction of Facial Features using an Efficient Convolution Siamese Network on Customized Dataset

Опубликована: Июнь 14, 2023

In recent years, one of the best applications visual analysis and understanding that has drawn a lot attention is face recognition. Due to its numerous uses in areas including safety, medical marketing, identity verification, surveillance, security etc., it caught interest many researchers. this research, network been proposed for The work performed on customized dataset with training testing images. 1500 photos altogether constitute dataset. simulation was run using hyper parameters such as batch-size value 128. optimizer Adam. outperformed accuracy 76.48%

Язык: Английский

---

Digital tools and challenges in human resource development and its potential within the maritime sector through bibliometric analysis

Journal of International Maritime Safety Environmental Affairs and Shipping, Год журнала: 2023, Номер 7(4)

Опубликована: Окт. 2, 2023

The maritime industry, a cornerstone of global trade and commerce, is currently undergoing significant transformation, primarily driven by technological advancements. Human resource development (HRD) in industry has become focal point, aimed at improving operational efficiency, safety, competitiveness. This research paper conducts an in-depth examination digital tools its challenges the context HRD through bibliometric analysis. findings indicate presence variety technologies, such as e-learning platforms (ELP), learning management systems (LMS), virtual reality (VR), augmented (AR), gamification, artificial intelligence (AI) machine (ML). Nevertheless, sector may encounter range obstacles including issues related to security concerns, skill gaps, strategic planning, change management, budget constraints, regulatory compliance. To effectively address these difficulties, it essential adopt comprehensive strategy that includes many components cybersecurity measures, efforts for talent development, alignment, techniques, budgetary restraint, legal scrutiny. this study have consequences sector, governments, academic community. It recommended can use digitization fundamental component their competitiveness safety measures.

Язык: Английский

---

Mapping the evolution of neurofeedback research: a bibliometric analysis of trends and future directions

Frontiers in Human Neuroscience, Год журнала: 2024, Номер 18

Опубликована: Май 10, 2024

Introduction This study conducts a bibliometric analysis on neurofeedback research to assess its current state and potential future developments. Methods It examined 3,626 journal articles from the Web of Science (WoS) using co-citation co-word methods. Results The identified three major clusters: “Real-Time fMRI Neurofeedback Self-Regulation Brain Activity,” “EEG Cognitive Performance Enhancement,” “Treatment ADHD Using Neurofeedback.” highlighted four key “Neurofeedback in Mental Health Research,” “Brain-Computer Interfaces for Stroke Rehabilitation,” Youth,” “Neural Mechanisms Emotion with Advanced Neuroimaging. Discussion in-depth significantly enhances our understanding dynamic field neurofeedback, indicating treating improving performance. offers non-invasive, ethical alternatives conventional psychopharmacology aligns trend toward personalized medicine, suggesting specialized solutions mental health rehabilitation as growing focus medical practice.

Язык: Английский

---

GBERT: A Hybrid Deep Learning Model Based on GPT-BERT for Fake News Detection

Heliyon, Год журнала: 2024, Номер 10(16), С. e35865 - e35865

Опубликована: Авг. 1, 2024

The digital era has expanded social exposure with easy internet access for mobile users, allowing global communication. Now, people can get to know what is going on around the globe just a click; however, this also resulted in issue of fake news. Fake news content that pretends be true but actually false and disseminated defraud. poses threat harmony, politics, economy, public opinion. As result, bogus detection become an emerging research domain identify given piece text as genuine or fraudulent. In paper, new framework called Generative Bidirectional Encoder Representations from Transformers (GBERT) proposed leverages combination pre-trained transformer (GPT) (BERT) addresses classification problem. This combines best features both cutting-edge techniques-BERT's deep contextual understanding generative capabilities GPT-to create comprehensive representation text. Both GPT BERT are fine-tuned two real-world benchmark corpora have attained 95.30 % accuracy, 95.13 precision, 97.35 sensitivity, 96.23 F1 score. statistical test results indicate effectiveness suggest it promising approach eradicating landscape.

Язык: Английский

---

Understanding the shifting nature of fake news research: Consumption, dissemination, and detection

Journal of the Association for Information Science and Technology, Год журнала: 2025, Номер unknown

Опубликована: Янв. 13, 2025

Abstract Fake news on social media spreads faster and has become a major societal concern, prompting numerous publications knowledge sharing among researchers. This research aims to understand the shifting nature of fake by investigating citation relationships between significant using key route main path analysis (MPA). The process involves generating keywords, collecting selecting relevant data, conducting MPA in media. study analyzes 4.057 from 2010 2023, identifying 27 influential works shaping diffusion research. Findings reveal two phases: understanding consumption patterns analyzing its dissemination detection mechanisms. Through multiple‐global MPA, five trends are identified: health misinformation, fact‐checking, behavior, recognition, physiological interventions. shows continuous rise citations, with current focusing health‐related misinformation. offers insights into development topics media, emphasizing importance historical guiding future uncovering trends. Highlighting progression provides valuable context, enabling more nuanced field.

Язык: Английский

---