The main feature of mini bar charts is that they display the incidence rates of each type of hepatitis along with the historical series through an interactive control of animation. This allows for an interesting visual exploration of the question, enabling interested parties to ask various spatial and / or spatiotemporal questions, inciting possible causal relationships, gathering formal hypotheses for later modeling, observing patterns among many other perceptions. In addition, interactive controls enable a kind of "spatial filter", in addition to the filters commonly used in data manipulation. For example, the map is zoomed to the national level, but someone interested in, for example, the southern region of Brazil can use the zoom controls to focus on this region (where we can see, for example, that RS has the highest rates of hepatitis C, but the lowest rates of hepatitis B, throughout the historical series).

Based on data from the Brazilian Health Ministry's Epidemiological Bulletin of Viral Hepatitis of 2018, we constructed an interactive map of incidence rates per 100,000 hepatitis A inhabitants between 2004 and 2017 in each state. The whole process of data wrangling, geoprocessing operations and interface with the leaflet was done in the R, especially with the tidyverse, leaflet and leaflet.minicharts packages. Thanks in advance to the package authors.

Criamos dados "na mão", padronizamos os dados, calculamos a matriz de distâncias (euclidianas), fazemos o Hierarchical Clustering utilizando linkages "complete", "average" e "single", plotamos os dendogramas, e por fim fazemos "cortes" no dendograma de average linkage para o numero de clusters que melhor se adapta à interpretação dos dados. Os dados consistem em 6 municípios do Paraná, com 3 variáveis: População Total (PopTotal), Esperança de Vida ao Nascer (EVN) e Taxa de Alfabetização (TxAlfab).

This naive R Markdown file is the result of my first experiences in studying geospatial data and doing maps in R. At the time of publishing this document, I'm currently studying the "Working with Geospatial Data" course at Datacamp, as well doing a short post-graduate course in Geoprocessing. I apologize for errors and the possible naive approach. Hints and suggestions are welcome.