A State of Art Techniques on Machine Learning Algorithms: A Perspective of Supervised Learning Approaches in Data Classification

Saravanan, R.; Sujatha, P.

doi:10.1109/iccons.2018.8663155

Cited by 177 publications

(111 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Another way to categorize ML systems is whether it needs human supervision. Based on this standard, ML systems can be categorized into four different types, that is, supervised, unsupervised, semisupervised, and reinforcement . Semisupervised models could also be used in computational materials science studies when labels are only presented for partial data.…”

Section: Approaches In Computational Materials Sciencementioning

confidence: 99%

Machine learning and artificial neural network accelerated computational discoveries in materials science

Yang

Hou

Jiang

et al. 2019

WIREs Comput Mol Sci

View full text Add to dashboard Cite

Artificial intelligence (AI) has been referred to as the "fourth paradigm of science," and as part of a coherent toolbox of data-driven approaches, machine learning (ML) dramatically accelerates the computational discoveries. As the machinery for ML algorithms matures, significant advances have been made not only by the mainstream AI researchers, but also those work in computational materials science. The number of ML and artificial neural network (ANN) applications in the computational materials science is growing at an astounding rate. This perspective briefly reviews the state-of-the-art progress in some supervised and unsupervised methods with their respective applications. The characteristics of primary ML and ANN algorithms are first described. Then, the most critical applications of AI in computational materials science such as empirical interatomic potential development, ML-based potential, property predictions, and molecular discoveries using generative adversarial networks (GAN) are comprehensively reviewed. The central ideas underlying these ML applications are discussed, and future directions for integrating ML with computational materials science are given. Finally, a discussion on the applicability and limitations of current ML techniques and the remaining challenges are summarized.

show abstract

Section: Approaches In Computational Materials Sciencementioning

confidence: 99%

Machine learning and artificial neural network accelerated computational discoveries in materials science

Yang

Hou

Jiang

et al. 2019

WIREs Comput Mol Sci

View full text Add to dashboard Cite

show abstract

“…A few of its implementations function, including Supervised, Semi-Supervised and Unsupervised ML. Utilizing predictive methods, supervised ML predicts performance in statistical information and classifies the right label [6]. The most widely identified techniques there comprise the methodology to analyze and the trees for decision.…”

Section: Machine Learningmentioning

confidence: 99%

Review of Deployment of Machine Learning in Blockchain Methodology

Solanki¹,

Solanki²

2020

IRJASH

View full text Add to dashboard Cite

The evolution of blockchain methodology has been a remarkable, highly transformative and trend-setting platform in current years. BT's accessible platform reinforces data protection and confidentiality. In addition, the consensus framework in it ensures system is protected and accurate. Nevertheless, it introduces additional security challenges such as invasion by the majority and double consumption. Data analysis on encrypted data centered on blockchain is crucial to manage the existing challenges. Insights on these results elevates the value of emerging of Machine Learning technique. It covers the fair quantity of data needed to make specific choices. Consistency of data and its distribution are very critical in ML to increase findings reliability. The fusion of these two techniques will produce extremely accurate outcomes. In this article, we describe a thorough analysis on ML implementation to make smart applications based on BT further robust to threats. There are numerous standard ML approaches such as Support Vector Machines (SVM), Clustering, Bagging, and Deep Learning (DL) algorithms such as Convolutional Neural Network (CNN) and Long-Term Memory (LSTM) that can be employed to evaluate the threats on a block chain network. Finally, we discuss how two different techniques can be implemented in a number of smart applications like Unmanned Aerial Vehicle (UAV), Smart Grid (SG), medical care and Smart cities.

show abstract

“…The algorithm adapts the system model according to the errors between the achieved and wanted results. With enough training, the system can provide the output to a new input data [21]. Unsupervised learning training data set consists of inputs without any designated outputs (inputs are unlabelled) [14].…”

Section: Introductionmentioning

confidence: 99%

“…Unsupervised learning training data set consists of inputs without any designated outputs (inputs are unlabelled) [14]. In unsupervised learning, the system discovers the hidden patterns in the data on its own since there is no output vector [21]. The fundamental objective is to categorize the data into separate groups by looking at their similarities [20].…”

Section: Introductionmentioning

confidence: 99%

Q-Learning Based Scheduling With Successive Interference Cancellation

Mete

Girici

2020

IEEE Access

View full text Add to dashboard Cite

This work studies the problem of scheduling using Q-learning, which is a reinforcement learning algorithm, in a Successive Interference Cancellation (SIC)-enabled wireless ad hoc network. Distributed Q-learning algorithm tries to find the best schedule for the transmission of maximum number of packets in the presence of the SIC technique. Performance of the algorithm is compared to the case where Q-learning is applied to a wireless network without SIC. In addition to that, the number of successful transmissions of our algorithm is compared to the optimal solution with and without SIC. Numerical results reveal that Q-learning based scheduling with SIC shows an improved performance compared to Q-learning scheduling without SIC and the optimal solution without SIC. Also, Q-learning scheduling with SIC shows similar performance to optimal scheduling with SIC when transmitting a reasonable number of packets. Thus, combining Q-learning and the SIC technique in wireless ad hoc networks is an effective approach to increase the number of transmitted packets.

show abstract

A State of Art Techniques on Machine Learning Algorithms: A Perspective of Supervised Learning Approaches in Data Classification

Cited by 177 publications

References 16 publications

Machine learning and artificial neural network accelerated computational discoveries in materials science

Machine learning and artificial neural network accelerated computational discoveries in materials science

Review of Deployment of Machine Learning in Blockchain Methodology

Q-Learning Based Scheduling With Successive Interference Cancellation

Contact Info

Product

Resources

About