Visualisation Strategies for Comparing Political Ideas with the ORATIO Platform

Data visualisation has become one of the most relevant DH topics, due to the advent of Big Data in Humanities research practices, and to the need to make complex statistical analyses accessible to users without a technical background. Although several visualisation libraries, such as d3.js, are now freely available online and are relatively easy to use, it is still a challenging task to provide simple and effective interface design, avoiding both over-complex and over-simplified solutions. When the data to be displayed have undergone complex processing, for instance automated text analysis, it is of paramount importance to preserve all the information conveyed by such analyses, while making it understandable to the users. In this work, we present a collaboration between communication design and natural language processing (NLP) researchers, devising effective strategies to display different aspects of the semantic content of texts. The outcome of the collaboration is the ORATIO platform, specifically developed to compare different points of view automatically extracted from text. The most challenging tasks, indeed, concerned the visualisation and the exploration of differences and overlaps detected through automated text processing.