Please use this identifier to cite or link to this item:
Title: Room occupancy detection using modified stacking
Authors: Somkid Amornsamankur
Keywords: Binary classification;Multiclass classification;Neural network;Occupancy detection;Stack generalization;Stacking
Issue Date: 2017
Publisher: Association for Computing Machinery
Citation: 9th International Conference on Machine Learning and Computing, ICMLC 2017
Abstract: Occupancy detection is a binary classification task. However, in this paper, stacking for multiclass classification is applied to detect occupancy of a room. Neural network with duo outputs are combined with stacking. The outputs of stacking for multiclass classification are then integrated to get a binary classification. The occupancy detection dataset obtained from UCI Machine Learning Repository is used in the experiment. It is found that our proposed stacking technique provides better accuracy result than the traditional stacking for binary classification.
ISSN: 978-145034817-1
Appears in Collections:Mathematics: International Proceedings

Files in This Item:
There are no files associated with this item.

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.