What Gives Language its Sentiment?
I came across James W. Pennebaker’s book The Secret Life of Pronouns, which explores sentiment analysis in-depth. While we often associate tone and sentiment with action words and key nouns in a piece of writing, Pennebaker’s research concluded that auxiliary words and pronouns actually have significant impact on the tone and style of language. The usage and commonality of different types of auxiliary words used in a text dictates how logical or emotional the writing comes across. I also learned that the use of pronouns can indicated differences in social status and emotional state.
I also looked at how other programmers analyzes sentiment with the use of word affinity resources. Combined with other linguistics research, I developed a few key metrics of analyzing text.
▸Positivity & Negativity
▸Sociability & Intimacy
▸Orderliness & Chaos
▸Distinctiveness & Commonality
How Can Computers Evaluate Emotions in Writing?
The making of Texture can be broken down in two parts. First, I wrote custom literary analysis code that detects genre, tone, sentiment, and status of any given text. Those results are each mapped onto a sentiment scale between -10 and 10.
The sentiment scale contains four major sets, each representing two opposing sentiments. A given text will have a value in between the two sentiment extremes.
Sentiment Scale in Use Example
From Literary Analysis to Generative Design
I created visual designs based on my sentiment values. The design of the posters are purely generative. Each value is directly and indirectly correlated with composition, scale, shape, color, and contrast.
Experimenting with Various Types of Writing
Here are the generative posters created by Texture.
Short Children's Stories
Various Wikipedia Articles on Linguistics
Spotify Top 50 as of March, 2019
Featured News Articles from the Washington Post
Poetry of Sylvia Plath
Designing the Front-end