A New Quality Measure and Visualization of the Short-Quantified Sentences of Natural Language on Maps – A Case on COVID-19 Data

Abstract

Maps are a common tool for visualizing various statistical figures that describe development in our society. Domain experts, journalists, and general public can pose questions on how to emphasize regions where, for instance, most young patients have long stayed in hospitals. One of the visualization’s problems is expressing validities of short-quantified sentences for regions on maps. The truth value of a summary assigns a value from the unit interval, which makes it suitable for interpretation on maps by hues of a selected colour, but it does not reflect the data distribution among regions. To meet this goal, a new quality measure covering data distribution among districts and its aggregation by the ordinal sums of conjunctive and disjunctive functions with the truth value is proposed and documented on examples. The next proposal is a relative quantifier expressing significant proportion of entities. This model is applied to the interpretation of COVID-19 cases development in the Slovak Republic on real data from one health insurance company. Finally, this article discusses the applicability of the proposed approach in other areas where the interpretation of summarized sentences on maps is beneficial.