Detecting novel class for sensor-based activity recognition using reject rule

3Citations
Citations of this article
1Readers
Mendeley users who have this article in their library.
Get full text

Abstract

Many researches have shown that data streams are continuous and changeable which make them hard to be classified accurately. The major difficulty in data classification is concept evolution, namely, novel class detection. Learn++ group methods are normally employed for stream data, however, these methods hardly handle the novel class detection problem. Therefore, in this paper, we introduce an approach that combines Learn++.NSE with reject option and our research findings can be used in activity recognition whose data streams are collected by body-worn sensors. In our experiment, the proposed approach shows better performance than Learn++.NSE algorithm.

Cite

CITATION STYLE

APA

Deng, C., Yuan, W., Tao, Z., & Cao, J. (2016). Detecting novel class for sensor-based activity recognition using reject rule. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 9864 LNCS, pp. 34–44). Springer Verlag. https://doi.org/10.1007/978-3-319-45940-0_4

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free