Linguistic Analysis

I have been primarily working with the story's structure, its sentences' syntax, the introduction of characters, their motivations, and the descriptions of these characters. I rewrote a number of the more generic and inquisitive closing bookends in order to have endings that fit better with the narrative flow. Upon doing this, it occurred to me that while yes, the data being used is important in order to have a good story (one could have great code, but if the data it pulls from is limited or does not match a story-telling style then the overall product will fall short), language is a versatile mode of communication. Eventually, the connotation of a word changes, new slang arises, phrases become arbitrary and awkward. The usage of the work "folks" struck me in one of the sentences. While it is still a word that is used and can be understood, it is older and less commonly employed by the younger generation (also, its use varies by geographical location). That had me thinking that if we want to have a machine that tells stories, that appears creative and almost cognizant, the use and reuse of set expressions and templates would limit us in this endeavor. What would further our pursuit, I believe, would be having a software that is able to analyze the changes in the usage of words. With the internet - and through limiting the program to specific types of sites in order to decrease the chance of improper usage of words - the amount of linguistic data a machine would have access to would enable it to use words relatively accurately with the change of time. The computer would thus keep with the time's usage of words and phrases and the risk of it becoming arbitrary would decrease. Of course, things are rarely that simple and it would take a lot of time and effort to develop, but I believe it would certainly prove useful in the field of computational creativity as it would allow the machine to move beyond its creator's limited knowledge, data, and biases (as far as stories, archetypes, positive and negative ambiences go), and to pull on the data of a collective whole (narrowed down and designated) in order to produce more "original" and creative tales.

Scholarly Lite is a free theme, contributed to the Drupal Community by More than Themes.