Abstract—Novelty detection represents the detection of
anomalous data based on a training set consisting of only the
normal data. In this study, we propose a new probabilistic
approach for novelty detection to effectively detect anomalous
data, particularly for the case of multimodal training dataset.
Our method is inspired by the Least-Squares Probabilistic
Classifier (LSPC), which is an efficient multi-class
classification method. Numerical experimental results based on
multimodal datasets show that the proposed method
outperforms the related methods.
Index Terms—Novelty detection, multimodal datasets,
least-square probabilistic analysis.
The authors are with University of Tsukuba, Ibaraki, Japan (e-mail:
yoda.hiroyuki.wy@alumni.tsukuba.ac.jp, imakura@cs.tsukuba.ac.jp,
matsuda.momo.ww@alumni.tsukuba.ac.jp, yexiucai2013@gmail.com,
sakurai@cs.tsukuba.ac.jp).
Cite: Hiroyuki Yoda, Akira Imakura, Momo Matsuda, Xiucai Ye, and Tetsuya Sakurai, "Novelty Detection in Multimodal Datasets Based on Least Square Probabilistic Analysis," International Journal of Machine Learning and Computing vol. 10, no. 4, pp. 527-533, 2020.
Copyright © 2020 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).