Conference proceeding
Input feature selection for sensor data mining based on mutual information
DYNAMICS OF CONTINUOUS DISCRETE AND IMPULSIVE SYSTEMS-SERIES B-APPLICATIONS & ALGORITHMS, Vol.13E, pp.1203-1208
01 Dec 2006
Featured in Collection : UN Sustainable Development Goals @ Drexel
Abstract
Input feature selection aims to reduce the dimensionality of problems by selecting the most informative instead of irrelevant and/or redundant measurements for sensor data mining. In this paper, two novel measures for feature ranking are presented: one is an improved formula to estimate the conditional mutual information between the candidate feature and the target class given the subset of selected features under the assumption of uniform distributions; the other is a criterion that is able to capture both irrelevant and redundant input features under arbitrary distributions of information. Furthermore, two new feature selection algorithms are proposed, and experimental results demonstrate the good performances of them on benchmark datasets.
Metrics
1 Record Views
Details
- Title
- Input feature selection for sensor data mining based on mutual information
- Creators
- J. J. HuangY. Z. CaiX. M. Xu
- Publication Details
- DYNAMICS OF CONTINUOUS DISCRETE AND IMPULSIVE SYSTEMS-SERIES B-APPLICATIONS & ALGORITHMS, Vol.13E, pp.1203-1208
- Publisher
- Watam Press
- Number of pages
- 6
- Resource Type
- Conference proceeding
- Language
- English
- Academic Unit
- Computer Science (Computing)
- Identifiers
- 991020638508804721
UN Sustainable Development Goals (SDGs)
This output has contributed to the advancement of the following goals:
InCites Highlights
These are selected metrics from InCites Benchmarking & Analytics tool, related to this output
- Web of Science research areas
- Mathematics, Applied