scikit-dyn2sel -- A Dynamic Selection Framework for Data Streams
This work provides a tool for researchers and practitioners dealing with incremental learning in data streams, though it is incremental as it adapts existing DCS techniques to a new context.
The paper tackles the challenge of applying dynamic selection of classifiers (DCS) to data streams by introducing scikit-dyn2sel, an open-source Python library designed for this purpose, which addresses issues like concept drift and integrates with streaming ensembles.
Mining data streams is a challenge per se. It must be ready to deal with an enormous amount of data and with problems not present in batch machine learning, such as concept drift. Therefore, applying a batch-designed technique, such as dynamic selection of classifiers (DCS) also presents a challenge. The dynamic characteristic of ensembles that deal with streams presents barriers to the application of traditional DCS techniques in such classifiers. scikit-dyn2sel is an open-source python library tailored for dynamic selection techniques in streaming data. scikit-dyn2sel's development follows code quality and testing standards, including PEP8 compliance and automated high test coverage using codecov.io and circleci.com. Source code, documentation, and examples are made available on GitHub at https://github.com/luccaportes/Scikit-DYN2SEL.