Power Poetry's Poetry Genome and Machine Learning

"Genome: an organism's complete set of DNA, including all of its genes; each genome contains all of the information needed to build and maintain that organism "(Genetics Home Reference).

What'd we do?

Rap GenomeIf you've been wondering what the words "Poetry Genome" chillin' in the black box on the right-hand side of any poem page on the site mean, we're here to solve the mystery! It's gonna get a little confusing, so take a deep breath and bear with us.

Our Poetry Genome is a tool that uses Machine Learning to generate a percentage comparison to various writers based on the theme, style, and content of any poem on the site. Both classic and contemporary writers are represented in different sections of our Genome (shoutout to Shakespeare and Kendrick).

How'd we do it?

But wait, what's Machine Learning? It's basically a branch of artificial intelligence based on the idea that machines can learn and adapt through experience. Don't get freaked out — the machines aren't going to learn so much that they'll rise up against us and take over the planet...yet.

For our Rap Poetry Genome we've taken 30 influential artists and compiled 10 songs from each artist into our Machine Learning tool table (think Excel spreadsheets). To avoid repetitiveness we've removed the hook (chorus) and added in the verses.

Here comes the awesome part. The Machine Learning tool absorbs this data we've entered into the table and picks up on the subtleties and differences in each artist's work. Then it actually creates an algorithm (a step-by-step solution) that makes it so when the table is compared to any poem on the site it can determine how similar that poem is to the artists in the table. The tool goes a step further and picks out, percentage-wise, the top three artists the poem is most similar to. But wait, there's more. Since this is a learning tool, the more it studies, the more it knows (just like you in school)! So, the more poems it reads on the site, the better it gets at figuring out the similarities between poems and the work of various artists.

What does it mean?

Okay, so you've posted a poem on the site and found out its similarities equal 4.3% Eminem, 1.2% Nicki Minaj, and 4.5% Lil Yachty. Now what? Go investigate! Find out more about these artists: their school years, their favorite books, their personal lives pre and post-career success. Matching with certain artists goes beyond the words and literary techniques. Researching more about the artists you match with can help you discover more about yourself, too.

What are you waiting for? Test out our Poetry Genome and see which artists your poems are most similar to. Does it vary per poem or does the data show that you and Ludacris would be BFFs? Did you match with one of your faves? Share with us on Twitter @powerpoetry and Facebook with #poetrygenome!

Location

Ethiopia