Generator of synthetic n-dimensional stream datasets for the evaluation of stream clustering and outlier detection algorithms (MATLAB R2016 or higher)
- Author: Denis Ojdanic, [email protected]
- Supervisor: Félix Iglesias Vázquez, [email protected]
Open MATLAB. From the [mdcstream-matlab] folder run:
mdcstream_help
or, for a quick example using MDCGen as data generator, run:
mdcstream_example
Félix Iglesias, Denis Ojdanic, Alexander Hartl, and Tanja Zseby. 2020. MDCStream: Stream Data Generator for Testing Analysis Algorithms. In Proceedings of the 13th EAI International Conference on Performance Evaluation Methodologies and Tools (VALUETOOLS '20). Association for Computing Machinery, New York, NY, USA, 56–63. DOI:https://doi.org/10.1145/3388831.3388832
In order to fully experience the benefits from mdcstream it is recommended to use MDCGen as an underlying static data set generator. MDCGen can be downloaded using the following link: https://github.com/CN-TU/mdcgen-matlab