It’s the start of National Novel Writing Month, or NaNoWriMo. One of the challenges of writing at scale is ensuring that your individual chapters and the work overall fit the emotional arc needed to hook a reader and keep them reading, especially if you publish a serial story. To help, Trust Insights has made an emotional arc analyzer tool available to you for free, based on the Syuzhet library in the R programming language.
This easy-to-use tool lets you copy and paste a chapter you’re working on into it and it’ll tell you the strength and sentiment of your chapter’s text.
What It Does
For example, take chapter 7 from Lewis Carroll’s Alice in Wonderland, which is the Mad Hatter’s tea party:
The red line represents the relative sentiment (both intensity and positivity/negativity) throughout the chapter; the vertical axis is measured from 1.0 (most intensely positive) to -1.0 (most intensely negative), with 0 being neutral and no emotional intensity. The horizontal axis measures the percentage of the text; at 40%, a 1,000 word chapter is around word 400.
Most of this chapter of Alice in Wonderland is somewhat negative in sentiment as Alice encounters and is frustrated by the Mad Hatter, until she finally leaves for the beautiful garden at the end, ending the chapter on a positive note.
You can use the tool on an entire work as well. Here’s what all of Alice in Wonderland looks like:
Why Sentiment Analysis Matters
You can use this tool to analyze your own text or the text of others (especially if you’re part of a team doing proofreading) and even paste in your entire book when you’re done.
How does it benefit you? If you’ve written a chapter with the intent of creating a significant emotional impact, but the tool doesn’t reflect that, this tool will help you visualize that quickly. If you’ve written a text that is supposed to have a happy ending, but instead the sentiment takes a nose dive, you’ll see that quickly as well.
If you’re editing, this will help you get immediate insights into your writer’s work and understand the general sentiment of their writing in moments; the percentages help you scroll quickly to the relative place in the writing where sentiment may be off and start editing more quickly.
As with all forms of present-day machine learning, this tool is not a substitute for human judgement or experience. It’s a supplement intended to do quick analysis as you write or edit, to give you faster insights into your writing. Enjoy it!
Q: “How private is our information? Are you capturing people’s stories?”
A: No. We are not capturing any information because [a] we don’t want to violate anyone’s copyrights and [b] our server couldn’t store that much text without blowing up.
Get unique data, analysis, and perspectives on analytics, insights, machine learning, marketing, and AI in the weekly Trust Insights newsletter, Data in the Headlights. Subscribe now for free; new issues every Wednesday!
Want to learn more about data, analytics, and insights? Subscribe to In-Ear Insights, the Trust Insights podcast, with new 10-minute or less episodes every week.
Also published on Medium.