**8. Conclusions**

In this work, we developed a high performance and interpretable Grey-Box ML model. Our main purpose was to acquire both benefits of black and White-Box models in order to build an interpretable classifier with a better classification performance, compared to a single White-Box model and comparable to that of a Black-Box. More specifically, we utilized a Black-Box model to enlarge an initial labeled dataset and the final augmented dataset was used to train a White-Box model. In contrast to the self-training framework, this trained White-Box model was utilized as the final predictor. This ensemble model falls into the category of intrinsic interpretability since its output predictor is a White-Box which is by nature interpretable. Our experimental results revealed that our proposed Grey-Box model has accuracy comparable to that of a Black-Box but better accuracy comparing to a single White-Box model, being at the same time interpretable as a White-Box model.

In our future work, we aim to extend our experiments of the proposed model to several datasets and improve its prediction accuracy with more sophisticated and theoretically motivated ensemble learning methodologies combining various self-labeled algorithms.

**Author Contributions:** E.P., I.E.L. and P.P. conceived of the idea, designed and performed the experiments, analyzed the results, drafted the initial manuscript and revised the final manuscript. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Conflicts of Interest:** The authors declare no conflict of interest.
