A Survey of Ensemble Learning: Concepts, Algorithms, Applications, and Prospects

IEEE Access, Journal Year: 2022, Volume and Issue: 10, P. 99129 - 99149

Published: Jan. 1, 2022

Ensemble learning techniques have achieved state-of-the-art performance in diverse machine applications by combining the predictions from two or more base models. This paper presents a concise overview of ensemble learning, covering three main methods: bagging, boosting, and stacking, their early development to recent algorithms. The study focuses on widely used algorithms, including random forest, adaptive boosting (AdaBoost), gradient extreme (XGBoost), light (LightGBM), categorical (CatBoost). An attempt is made concisely cover mathematical algorithmic representations, which lacking existing literature would be beneficial researchers practitioners.

Language: Английский

---

Cyber Security in IoT-Based Cloud Computing: A Comprehensive Survey

Electronics, Journal Year: 2021, Volume and Issue: 11(1), P. 16 - 16

Published: Dec. 22, 2021

Cloud computing provides the flexible architecture where data and resources are dispersed at various locations accessible from industrial environments. has changed using, storing, sharing of such as data, services, applications for applications. During last decade, industries have rapidly switched to cloud having more comprehensive access, reduced cost, increased performance. In addition, significant improvement been observed in internet things (IoT) with integration computing. However, this rapid transition into raised security issues concerns. Traditional solutions not directly applicable sometimes ineffective cloud-based systems. platforms’ challenges concerns addressed during three years, despite successive use proliferation multifaceted cyber weapons. The evolution deep learning (DL) artificial intelligence (AI) domain brought many benefits that can be utilized address cloud. findings proposed research include following: we present a survey enabling IoT architecture, configurations, models; classification four major categories (data, network service, applications, people-related issues), which discussed detail; identify inspect latest advancements attacks; identify, discuss, analyze each category limitations general, perspective; provide technological identified literature then gaps IoT-based infrastructure highlight future directions blend cybersecurity

Language: Английский

---

Adventures in data analysis: a systematic review of Deep Learning techniques for pattern recognition in cyber-physical-social systems

Multimedia Tools and Applications, Journal Year: 2023, Volume and Issue: 83(8), P. 22909 - 22973

Published: Aug. 9, 2023

Language: Английский

---

Blockchain Technology on Smart Grid, Energy Trading, and Big Data: Security Issues, Challenges, and Recommendations

Wireless Communications and Mobile Computing, Journal Year: 2022, Volume and Issue: 2022, P. 1 - 26

Published: Jan. 18, 2022

The smart grid idea was implemented as a modern interpretation of the traditional power to find out most efficient way combine renewable energy and storage technologies. Throughout this way, big data Internet always provide revolutionary solution for ensuring that electrical linked intelligent grid, also known Internet. blockchain has some significant features, making it an applicable technology standards solve security issues trust challenges. This study will present rigorous review implementations with cyber perception protections in grids. As result, we describe major scenarios can solve. Then, identify variety recent blockchain-based research works published various literature discuss concerns on systems. We numerous similar practical designs, experiments, items have recently been developed. Finally, go through important problems possible directions using address concerns.

Language: Английский

---

A Comprehensive Survey on Computer Forensics: State-of-the-Art, Tools, Techniques, Challenges, and Future Directions

IEEE Access, Journal Year: 2022, Volume and Issue: 10, P. 11065 - 11089

Published: Jan. 1, 2022

With the alarmingly increasing rate of cybercrimes worldwide, there is a dire need to combat timely and effectively. Cyberattacks on computing machines leave certain artifacts target device storage that can reveal identity behavior cyber-criminals if processed analyzed intelligently. Forensic agencies law enforcement departments use several digital forensic toolkits, both commercial open-source, examine evidence. The proposed research survey focuses identifying current state-of-the-art forensics concepts in existing research, sheds light gaps, presents detailed introduction different computer domains toolkits used for era. also comparative analysis based tool's characteristics facilitate investigators tool selection during process. Finally, identifies derives challenges future directions forensics.

Language: Английский

---

Intrusion Detection in the IoT Under Data and Concept Drifts: Online Deep Learning Approach

IEEE Internet of Things Journal, Journal Year: 2022, Volume and Issue: 9(20), P. 19706 - 19716

Published: April 12, 2022

Although the existing machine learning-based intrusion detection systems in Internet of Things (IoT) usually perform well static environments, they struggle to preserve their performance over time, dynamic environments. Yet, IoT is a highly and heterogeneous environment, leading what known as data drift concept drift. Data phenomenon which embodies change that happens relationships among independent features, mainly due changes quality time. Concept depicts between input output learning model To detect drifts, we first propose technique capitalizes on principal component analysis (PCA) method study variance features across streams. We also discuss an online outlier identifies outliers diverge both from historical temporally close points. counter these deep neural network (DNN) dynamically adjusts sizes hidden layers based Hedge weighting mechanism, thus enabling steadily learn adapt new come. Experiments conducted IoT-based set suggest our solution stabilizes training testing compared DNN model, widely used for detection.

Language: Английский

---

COVID-19 Related Sentiment Analysis Using State-of-the-Art Machine Learning and Deep Learning Techniques

Frontiers in Public Health, Journal Year: 2022, Volume and Issue: 9

Published: Jan. 14, 2022

The coronavirus disease 2019 (COVID-19) pandemic has influenced the everyday life of people around globe. In general and during lockdown phases, worldwide use social media network to state their viewpoints feelings concerning that hampered daily lives. Twitter is one most commonly used platforms, it showed a massive increase in tweets related coronavirus, including positive, negative, neutral tweets, minimal period. researchers move toward sentiment analysis analyze various emotions public COVID-19 due diverse nature tweets. Meanwhile, have expressed regarding vaccinations' safety effectiveness on networking sites such as Twitter. As an advanced step, this paper, our proposed approach analyzes by focusing users who share opinions site. collected tweets' sentiments for classification using feature sets classifiers. early detection from allow better understanding handling pandemic. Tweets are categorized into classes. We evaluate performance machine learning (ML) deep (DL) classifiers evaluation metrics (i.e., accuracy, precision, recall, F1-score). Experiments prove provides accuracy 96.66, 95.22, 94.33, 93.88% COVISenti, COVIDSenti_A, COVIDSenti_B, COVIDSenti_C, respectively, compared all other methods study well existing approaches traditional ML DL algorithms.

Language: Английский

---

Cardiovascular Disease Detection using Ensemble Learning

Computational Intelligence and Neuroscience, Journal Year: 2022, Volume and Issue: 2022, P. 1 - 9

Published: Aug. 16, 2022

One of the most challenging tasks for clinicians is detecting symptoms cardiovascular disease as earlier possible. Many individuals worldwide die each year from disease. Since heart a major concern, it must be dealt with timely. Multiple variables affecting health, such excessive blood pressure, elevated cholesterol, an irregular pulse rate, and many more, make to diagnose cardiac Thus, artificial intelligence can useful in identifying treating diseases early on. This paper proposes ensemble-based approach that uses machine learning (ML) deep (DL) models predict person’s likelihood developing We employ six classification algorithms Models are trained using publicly available dataset cases. use random forest (RF) extract important features. The experiment results demonstrate ML ensemble model achieves best prediction accuracy 88.70%.

Language: Английский

---

Context-aware Emotion Detection from Low-resource Urdu Language Using Deep Neural Network

ACM Transactions on Asian and Low-Resource Language Information Processing, Journal Year: 2022, Volume and Issue: 22(5), P. 1 - 30

Published: April 1, 2022

Emotion detection (ED) plays a vital role in determining individual interest any field. Humans use gestures, facial expressions, and voice pitch choose words to describe their emotions. Significant work has been done detect emotions from the textual data English, French, Chinese, other high-resource languages. However, emotion classification not well studied low-resource languages (i.e., Urdu) due lack of labeled corpora. This article presents publicly available Urdu Nastalique Emotions Dataset ( UNED ) sentences paragraphs annotated with different proposes deep learning (DL)-based technique for classifying corpus. Our corpus six both sentences. We perform extensive experimentation evaluate quality further classify it using machine DL approaches. Experimental results show that developed DL-based model performs better than generic approaches an F1 score 85% on sentence-based 50% paragraph-based

Language: Английский

---

Current Studies and Applications of Shuffled Frog Leaping Algorithm: A Review

Archives of Computational Methods in Engineering, Journal Year: 2022, Volume and Issue: 29(5), P. 3459 - 3474

Published: Jan. 24, 2022

Language: Английский

---