A comparison between shallow and deep architecture classifiers on small dataset

Simple item page
dc.contributor.authorPasupa, Kitsuchart
dc.contributor.authorSunhem, Wisuwat
dc.date.accessioned2026-08-06T10:16:19Z
dc.date.available2026-08-06T10:16:19Z
dc.date.issued2017-02-23
dc.description.abstractMany machine learning algorithms have been introduced to solve different types of problem. Recently, many of these algorithms have been applied to deep architecture model and showed very impressive performance. In general, deep architecture model suffers from over-fitting problem when there is a small number of training data. In this paper, we attempted to remedy this problem in deep architecture with regularization techniques including overlap pooling, flipped-image augmentation and dropout, and we also compared a deep structure model (convolutional neural network (CNN)) with shallow structure models (support vector machine and artificial neural network with one hidden layer) on a small dataset. It was statistically confirmed that the shallow models achieved better performance than the deep model that did not use a regularization technique. However, a deep model augmented with a regularization technique-CNN with dropout technique-was competitive to the shallow models.
dc.identifier.citationProceedings of 2016 8th International Conference on Information Technology and Electrical Engineering Empowering Technology for Better Future Icitee 2016, 2017
dc.identifier.doi10.1109/ICITEED.2016.7863293
dc.identifier.other2-s2.0-85016017167
dc.identifier.urihttps://dspace.kmitl.ac.th/handle/123456789/7545
dc.sourceProceedings of 2016 8th International Conference on Information Technology and Electrical Engineering Empowering Technology for Better Future Icitee 2016
dc.subjectdeep learning
dc.subjectmachine learning
dc.subjectshallow learning
dc.subjectsmall dataset
dc.titleA comparison between shallow and deep architecture classifiers on small dataset
dc.typeConference Paper

Files

Collections

Research