Updated Actor analysis widget to match story text from document to the concatenated sentences in the story elements dataframe. This way we can lookup the correct storyid without having to maintain two different ids in the widget GUI and in the story elements. This was a nice temporary fix suggested by @ThijsVroegh!
I had a go at fixing #54 as well by changing the preprocessing to a more simpler function. This seems to fix it!
Issue
Attempts at fixing #39, #54 and #37
Description of changes
Includes