Open AccessFraud Detection Using Multi-layer Heterogeneous EnsembleMethod
Author(s) -
Haritha Rajeev Neenu Kuriakose
Publication year - 2021
Publication title -
international journal of modern trends in science and technology
Language(s) - English
Resource type - Journals
ISSN - 2455-3778
DOI - 10.46501/0706001
Subject(s) - computer science , layer (electronics) , misfortune , credit card fraud , artificial intelligence , property (philosophy) , misrepresentation , ensemble learning , cover (algebra) , cash , deep learning , machine learning , credit card , payment , finance , business , engineering , mechanical engineering , philosophy , chemistry , organic chemistry , epistemology , world wide web , perspective (graphical) , political science , law
Fraudulent detection is a large number of exercises that try to keep cash or property out of the way. Fraudsurveillance is used in many businesses such as banking or security. At the bank, misrepresentation mayinvolve producing checks or using a Credit Card taken. Different types of robberies can include misfortune orcreate a problem with the expectation of only a paid Layer Ensemble Method running other AI fields includingcollecting learning. Recently, there have been one deep group models deployed with a large number ofclassifiers in each layer. These models, as a result, require a much larger calculation. In addition, the deepintegration models are available that use all the separating elements including the unnecessary ones thatcan reduce the accuracy of the group. In this experiment, we propose a multi-layered learning structure calledthe Two-Layer Ensemble System to address the issue of definition. The proposed framework is working witha number of weird filters to get the troupe jumper sity, in these lines being a technology in the use ofequipment.