Pearson's Happiness v. Sadness Word Usage Over Time

Although Pearson seems upset during his earlier diary entries, it is surprising that his overall happy sentiment declines throughout the time that he writes, despite his trouble with his health and his family's financial status from 1829 to 1831. Tableau is software that allows us to see the average word usages throughout the text. In order to create the graph pictured, a list of happy and sad words was retrieved through NRC Word-Emotion Association Lexicon, which scores words based on the sentiment they are used to describe. I then went through the text to find all the occasions that words categorized as 'happy' or 'sad' were used. From this, Lexos was used to subtract the number of happy words from the average of sad words, creating a file that could be uploaded to Tableau. While this is a very useful tool, it is worth noting that it fails to take into account if a word such as 'not' is in front of a sentiment word. Therefore, if Pearson wrote 'I am not sad,' it would still be understood that he was writing about a sad occasion, and the same would be true if he wrote 'I am not happy.'