Fast online Qn estimator
This repository hosts the source code related to the C implementation of the FQN (Fast Qn) algorithm for online detection of outliers in data streams. We also provide source code for the implementation of the competing algorithm by Nunkesser et al. and for a variant of Nunkesser in which the buffer B is limited to linear size.
Finally, the repository also hosts synthetic datasets generated according to specific distributions.
Massimo Cafaro, Catiuscia Melle, Marco Pulimeno and Italo Epicoco University of Salento, Italy
If you use this software please cite the following paper:
M. Cafaro, C. Melle, M. Pulimeno, I. Epicoco. Fast Online Computation of the Qn Estimator with Applications to the Detection of Outliers in Data Streams. Expert Systems With Applications, Elsevier, Volume 164, February 2021, Article 113831, ISSN: 0957-4174, DOI: 10.1016/j.eswa.2020.113831