Quantifying & Visualizing Literature

In terms of my own thoughts about extracting data from literature and/or art, I think that it's a very interesting topic with its respective pros and cons. It certainly has its pros in that it brings a more mathematical and objective approach to the arts, however one of the biggest cons also presents itself in this area. Literature and the arts are inherently subjective and can be quite difficult to quantify in most aspects, especially when data is extracted from a digital format of something such as a sculpture, which instantly creates issues in that a 2D rendition is being used to assess a 3D work of art. This goes for other mediums too, such as errors occurring when scribing verbal stories to text, when relating a previously experienced event, interpreting what's happening in a photo, and so on. To put it short, extracting data from a subjective source is an interesting field of study, however certainly comes with drawbacks and errors simply from how it is conducted.

On a more relevant note, these extraction processes can be used on literary works as well, as we have experienced in running short stories through Voyant for analysis. In my experience, doing this proved to me even further how data analysis on subjective content such as a fictional story loaded with metaphors can bring its fair share of errors and misconceptions. I think that applying methods such as visualization through graphs, charts, and word repetition counts holds its own importance and can be useful in some aspects, however it just doesn't paint an accurate picture of what's going on in the literary work in most cases. Again this is only based on my experience working with the tools, and I suppose this kind of software may be more relevant to journals or other works of non-fiction, however I'm not super impressed.

What the Voyant tools have shown me about my Poe story are how many times certain words are said in the story, and when. The words chosen for most times repeated aren't even that relevant to the story, in my opinion. They provide substance, of course, as the author chose them to be in the story on purpose, however they don't paint a realistic picture of the story as one would think **extracted data** would. Just because it comes out of the literary work doesn't necessarily mean its the most useful. I think this is where data analytics really shows its use, because in terms of numbers these tools would be great. But for words in a fiction piece... not so much. So essentially, no. I don't think data mining and visualization methods are exceptionally useful for literary analysis. But I can see their purpose.