Interpretability of AI Systems in Electronic Governance

Modern electronic governance systems require cutting-edge analytical techniques to manage available ever-larger and distributes data, with a known spread of unstructured and unlabeled text documents. Many organizations are turning to data governance to exercise control over the quality of their data and their processes in order to guarantee the delivery of trustworthy decisions. In this context, modern AI breakthroughs give new opportunities to impact many application scenarios, like knowledge extraction and exploration in electronic governance. In this paper we introduce the need to build interpretable AI systems for electronic governance in order to improve trust and consequently user acceptance, highlighting some emergent topics and open challenges, mainly linked to integrating quantitative and qualitative techniques, such as deep learning and knowledge graphs for semantic-aware models.