D2.1 Text Mining Technical Specifications

This deliverable describes the role of text mining in the PoliRural project and the technical specifications of Semantic Explorer - the main tool for supporting pilots in the processes of needs gathering and policy assessment. The core semantic model will be built and "trained" on the corpus of EU texts, before being adapted to regional contexts and languages. After several iterations, fine-tuning and testing, the tool will be able to accurately extract the required information from the pre-compiled regional libraries, as well as process survey data gathered at the later stages of the Project. The aim is to create a set of services providing inputs for foresight exercises to assist in the tasks of horizon scanning, needs and challenges analysis, etc. Text Mining (TM) solutions should also serve as a tool to build regional profiles for the process of System Dynamics (SD). In addition to that, the system is intended to collect data from Social Media, in particular opinions and their relationship to various aspects of policies. This will allow for comparing the analysis with quantitative data, obtained through surveys, and to provide "one version of the truth" through interactive visuals.