A shiny interface that visualize some data related to Game Of Thrones
The designed GOT interface is structured in 4 Tabs. The first tab contains a visualization of the spatial distribution of the scenes of a character selected by the user. The interface displays a map designed through the data of shp and csv files containing the locations with their respective names and the number of scenes in each indicated location as well as the total sum of the durations of the scenes in dot format with their sizes which vary according to the sums of the durations. The interface also offers the possibility of just viewing the spatial distribution of the scenes of the previously selected character with another character from the series. If there are no scenes that include both people, the interface returns an empty map. In order to allow the user to see the correlation between the number of scenes and the durations, we decided to display them in the map separately.In order not to load the map, only the locations associated with the scenes in question are labelled. We have also used the geom_text_repel function to avoid overlapping data, this function allows us to distance the labels associated with the points. For more creativity, the user can change the colour of the points.
The second tab contains a visualization of the appearance time per episode of the character selected with ggplotly. The ggplotly allows a more interactive visualisation. We have chosen the point cloud (Geom_point) because the evolution of the time of presence per episode is not continuous (the character can appear in one episode and not in another). The episodes have been grouped by season and each season has its own colour. Similarly, the interface also allows you to visualise the presence time per episode of two characters.
The third tab contains a wordcloud of people sharing scenes with the selected character. The greater the number of times a person appears with the character, the greater the name in the word cloud. The user can limit the maximum number of words in the word cloud. The wordcloud is generally used in text mining to visualise the number of times a word appears in the text, which makes it a good visual representation tool for our case.
The last tab allows you to view, in case of existence, the location of the family (clan) to which the selected character belongs. The interface displays a map designed using the data stored in the shp and csv files which visualises the main region of the family with the number of population in this region. In case the character does not have a family, the interface displays an empty GOT map.