Mafalda - MAtchmaking Features for mAchine Learning Data Analysis

Mafalda is a semantic-enhanced framework for data mining on sensor streams, amenable to resource-constrained pervasive contexts. It merges an ontology-based characterization of data distributions with non-standard reasoning for a fine-grained event detection by treating the typical classification problem of Machine Learning as a resource discovery.

Reference ontologies and datasets used for the framework evaluation are available in the project sub-folders.

References

If you want to refer to Mafalda dataset and ontologies in a publication, please cite the following paper:

@article{mafalda-swj18,
  title={{Machine learning in the Internet of Things: A semantic-enhanced approach}},
  author={Ruta, Michele and Scioscia, Floriano and Loseto, Giuseppe and Pinto, Agnese and {Di Sciascio}, Eugenio},
  journal={Semantic Web},
  volume={10},
  number={1},
  pages={183--204},
  year={2019},
  publisher={IOS Press}
}