Litcius/Paper detail

Experimental Evaluation of Boosting Algorithms for Fuel Flame Extinguishment with Acoustic Wave

Raj Gaurang Tiwari, Ambuj Kumar Agarwal, Rupesh Kumar Jindal, Anshbir Singh

20222022 International Conference on Innovation and Intelligence for Informatics, Computing, and Technologies (3ICT)14 citationsDOI

Abstract

Automated classification and regression tasks are often improved by using ensemble approaches. Both bagging and boosting fall within this category. Combining multiple weak and incorrect rules into one highly accurate prediction rule is known as "boosting" in the field of machine learning. A fire may start for a variety of different causes, making it a multifaceted natural catastrophe. The classification of flame extinction and non-extinction was accomplished by the application of six distinct boosting approaches in this research. Results from this testing demonstrate that the Hist Gradient Boost, Light Gradient Boost Machine, and Cat Boost algorithms had the best classification accuracy of the models tested.

Topics & Concepts

Boosting (machine learning)ExtinguishmentGradient boostingArtificial intelligenceComputer scienceMachine learningAlgorithmStatistical classificationEnsemble learningExtinction (optical mineralogy)Pattern recognition (psychology)Random forestPhysicsPolitical scienceLawOpticsFire Detection and Safety SystemsAnomaly Detection Techniques and ApplicationsTime Series Analysis and Forecasting
Experimental Evaluation of Boosting Algorithms for Fuel Flame Extinguishment with Acoustic Wave | Litcius