To increase people’s trust in adopting AI models, it is important a complete interpretability of the behaviour of AI-models is shown.
Context of the research activity Artificial intelligence (AI) is playing an increasingly important role in various applicative areas, including industry, which require the analysis of data and information of various kinds, including texts and images. However, the dominant role played by AI models has generated a concern due to possible biases, transparency and interpretability of the model responses. To increase people’s trust in adopting AI models, it is important a complete interpretability of the behaviour of AI-models.