Authorship identification using ensemble learning

Scientific Reports, Journal Year: 2022, Volume and Issue: 12(1)

Published: June 9, 2022

Abstract With time, textual data is proliferating, primarily through the publications of articles. this rapid increase in data, anonymous content also increasing. Researchers are searching for alternative strategies to identify author an unknown text. There a need develop system actual texts based on given set writing samples. This study presents novel approach ensemble learning, DistilBERT , and conventional machine learning techniques authorship identification. The proposed extracts valuable characteristics using count vectorizer bi-gram Term frequency-inverse document frequency (TF-IDF). An extensive detailed dataset, “All news” used experimentation. dataset divided into three subsets (article1, article2, article3). We limit scope selected ten authors first 20 second experimental results provide better performance all dataset. In scope, prove that from 10 provides accuracy gain 3.14% 2.44% article1 Similarly, authors, 5.25% 7.17% which than previous state-of-the-art studies.

Language: Английский

---

Big Data ML-Based Fake News Detection Using Distributed Learning

IEEE Access, Journal Year: 2023, Volume and Issue: 11, P. 29447 - 29463

Published: Jan. 1, 2023

Users rely heavily on social media to consume and share news, facilitating the mass dis-semination of genuine fake stories. The proliferation misinformation various platforms has serious consequences for society. inability differentiate between sev-eral forms false news Twitter is a major obstacle effective detection news. Researchers have made progress toward solution by placing greater emphasis methods identifying bogus dataset FNC-1, which includes four categories will be used in this study. state-of-the-art spotting are evaluated compared using big data technology (Spark) machine learning. methodology study employed decentralized Spark cluster create stacked ensemble model. Following feature extraction N-grams, Hashing TF-IDF, count vectorizer, we proposed classification results show that suggested model superior performance 92.45% F1 score 83.10 % baseline approach. achieved an additional 9.35% techniques.

Language: Английский

---

Evolving cybersecurity frontiers: A comprehensive survey on concept drift and feature dynamics aware machine and deep learning in intrusion detection systems

Engineering Applications of Artificial Intelligence, Journal Year: 2024, Volume and Issue: 137, P. 109143 - 109143

Published: Aug. 22, 2024

Language: Английский

---

A Review of Machine Learning and Transfer Learning Strategies for Intrusion Detection Systems in 5G and Beyond

Mathematics, Journal Year: 2025, Volume and Issue: 13(7), P. 1088 - 1088

Published: March 26, 2025

This review systematically explores the application of machine learning (ML) models in context Intrusion Detection Systems (IDSs) for modern network security, particularly within 5G environments. The evaluation is based on 5G-NIDD dataset, a richly labeled resource encompassing broad range behaviors, from benign user traffic to various attack scenarios. examines multiple models, assessing their performance across critical metrics, including accuracy, precision, recall, F1-score, Receiver Operating Characteristic (ROC), Area Under Curve (AUC), and execution time. Key findings indicate that K-Nearest Neighbors (KNN) model excels accuracy ROC AUC, while Voting Classifier achieves superior precision F1-score. Other decision tree (DT), Bagging, Extra Trees, demonstrate strong AdaBoost shows underperformance all metrics. Naive Bayes (NB) stands out its computational efficiency despite moderate other areas. As technologies evolve, introducing more complex architectures, such as slicing, increases vulnerability cyber threats, Distributed Denial-of-Service (DDoS) attacks. also investigates potential deep (DL) Deep Transfer Learning (DTL) enhancing detection Advanced DL Bidirectional Long Short-Term Memory (BiLSTM), Convolutional Neural Networks (CNNs), Residual (ResNet), Inception, are evaluated, with focus ability DTL leverage knowledge transfer source datasets improve sparse data. underscore importance large-scale adaptive security mechanisms addressing evolving threats. concludes by highlighting significant role ML approaches strengthening defense fostering proactive, robust solutions future networks.

Language: Английский

---

Impact of industrial 4.0 on environment along with correlation between economic growth and carbon emissions

International Journal of Systems Assurance Engineering and Management, Journal Year: 2021, Volume and Issue: 13(S1), P. 415 - 423

Published: Nov. 9, 2021

Language: Английский

---

Dynamic Incremental Ensemble Fuzzy Classifier for Data Streams in Green Internet of Things

IEEE Transactions on Green Communications and Networking, Journal Year: 2022, Volume and Issue: 6(3), P. 1316 - 1329

Published: Feb. 16, 2022

Due to the fast, dynamic, and continuous arrival of data streams in green Internet Things (IoT) environment, probability distribution changes over time. In real IoT scenarios such as unmanned aerial vehicle (UAV) detection smart light switch control, have reduced trained model's accuracy for problems classification, making it challenging detect UAV intruders predict whether energy-saving lamps buildings are on or off. this paper, an incremental ensemble classification method is proposed improve prediction IoT. Specifically, a fuzzy rule-based classifier combined with dynamic weighting algorithm improving accuracy. Moreover, model updated by incrementally learning characteristics streams, which can effectively handle concept drift caused streams. Experimental evaluations intrusion detection, buildings, other datasets show that approach yields 2% higher area under curve (AUC) geometric mean (G-mean) than existing methods Detection Occupancy 5% AUC G-mean five benchmarking datasets. For all datasets, 50% faster average training time methods.

Language: Английский

---

Short-Text Classification Detector: A Bert-Based Mental Approach

Computational Intelligence and Neuroscience, Journal Year: 2022, Volume and Issue: 2022, P. 1 - 11

Published: March 10, 2022

With the continuous development of Internet, social media based on short text has become popular. However, sparsity and shortness essays will restrict accuracy classification. Therefore, Bert model, we capture mental feature reviewers apply them for classification to improve its accuracy. Specifically, construct a model at language level fine tune better embed features. To verify this method, compare variety machine learning methods, such as support vector machine, convolution neural networks, recurrent networks. The results show following: (1) Through comparison, it is found that features can significantly (2) Combining input vectors provide more than separating two independent vectors. (3) be integrate text. results. This help promote

Language: Английский

---

A Large-Scale Benchmark Dataset for Anomaly Detection and Rare Event Classification for Audio Forensics

IEEE Access, Journal Year: 2022, Volume and Issue: 10, P. 38885 - 38894

Published: Jan. 1, 2022

With the emergence of new digital technologies, a significant surge has been seen in volume multimedia data generated from various smart devices. Several challenges for analysis have emerged to extract useful information data. One such challenge is early and accurate detection anomalies This study proposes an efficient technique anomaly classification rare events audio In this paper, we develop vast dataset containing seven different (anomalies) with 15 background environmental settings (e.g., beach, restaurant, train) focus on both anomalous sound events—baby cry, gunshots, broken glasses, footsteps) forensics. The proposed approach uses supreme feature extraction by extracting mel-frequency cepstral coefficients (MFCCs) features signals newly created selects minimum number best-performing optimum performance using principal component (PCA). These are input state-of-the-art machine learning algorithms analysis. We also apply realize good results. Experimental results reveal that effectively detects all superior existing approaches environments cases.

Language: Английский

---

Deep learning for religious and continent-based toxic content detection and classification

Scientific Reports, Journal Year: 2022, Volume and Issue: 12(1)

Published: Oct. 19, 2022

Abstract With time, numerous online communication platforms have emerged that allow people to express themselves, increasing the dissemination of toxic languages, such as racism, sexual harassment, and other negative behaviors are not accepted in polite society. As a result, language identification has critical application natural processing. Numerous academic industrial researchers recently researched using machine learning algorithms. However, Nontoxic comments, including particular descriptors, Muslim, Jewish, White, Black, were assigned unrealistically high toxicity ratings several models. This research analyzes compares modern deep algorithms for multilabel comments classification. We explore two scenarios: first is classification Religious second race or ethnicity with various word embeddings (GloVe, Word2vec, FastText) without an ordinary embedding layer. Experiments show CNN model produced best results classifying both scenarios. compared outcomes these performances terms evaluation metrics.

Language: Английский

---

A novel methodology for malicious traffic detection in smart devices using BI-LSTM–CNN-dependent deep learning methodology

Neural Computing and Applications, Journal Year: 2023, Volume and Issue: 35(27), P. 20319 - 20338

Published: July 20, 2023

Language: Английский

---