This video is a demonstration of the Policy Explainability and Interpretation module of the AI4PublicPolicy VPME, developed by our partners from Netcompany-Intrasoft that aims to produce end to end analytics pipelines that can analyze the policy datasets and produce meaningful insights.

This video is a demonstration of the Policy Explainability and Interpretation module of the AI4PublicPolicy VPME. This task requires the involvement of policymakers, stakeholders and AI experts with the aim to produce end to end analytics pipelines that can analyze the policy datasets and produce meaningful insights. This will be realized by developing and applying AI and explainable models on policy dimensions and social problems, in order to produce interpretations and solutions for the policies.

The first demo of this module that is presented in this video is based on the AI4PublicPolicy Athens pilot and specifically, the optimized parking spaces allocation scenario. The problem addressed is the highest coverage of parking spaces based on data regarding parking tickets’ location, activation and duration, as well as three datasets concerning the total parking spaces number, which will be used in the near future in order to enhance the AI model performance.

The second demonstration presented in the video is again based on the AI4PublicPolicy Athens pilot maintenance incidents prediction for planning purposes, focusing on the number of incidents per period and prediction accuracy, as well as using three datasets concerning the incidents type, the incidents department/service and the incidents status change history.