Before the Commission proposes new initiatives, it is responsible for assessing potential consequences, especially on policy adoption issues, ensuring that legislative proposals correspond to the needs of those most concerned and avoid unnecessary administratively costly amendments. Data-driven, empirical tools can provide valuable insight for policy-makers, enabling them to foresee potential issues or delays of transpositions. In this contribution, we elaborate on the creation of PolNetCast a computational framework that will use an extended version of the CJEU citation network and the conjunction of interactions with the EU mandatory acts (directives, regulations and decisions) over time, where each EU law represents the center of specific clusters capturing the trajectory of potential issues about EU policy adoption in member states. PolNetCast is an application of network topology-based machine learning that can learn from the history of the network and forecast future interactions. We expect this application to be an important step for data-driven legislative proposals on EU and national level, paving the way to FAIR policy making.
|Number of pages||5|
|Publication status||Published - 2020|
|Event||Data for Policy 2020: 5th International Conference - Online, London, United Kingdom|
Duration: 15 Sept 2020 → 17 Sept 2020
|Conference||Data for Policy 2020|
|Period||15/09/20 → 17/09/20|