Pick up lines from a neural network, more miss than hit | Inquirer Technology

Pick up lines from a neural network, more miss than hit

/ 05:49 PM April 12, 2017

neural network generates pick up lines

Image: INQUIRER.net

From the neural network that brought you fake Pokemon comes a new stream of gibberish designed to help humans break the ice.

Research scientist Janelle Shane has moved forward in her neural network training project and began teaching a neural network about pick up lines.

ADVERTISEMENT

“Gathering the dataset was much more painful than I had expected,” says Shane. “I hadn’t really read many of these before, and most were obscene, or aggressive, or kind of insulting. I began to regret the whole project.”

FEATURED STORIES

While the neural network managed to produce a list, the system predictably failed to grasp the intricacies of word play inherent to the source material.

Some of the lines it generated were “I want to see you to my heart,” “I have a really falling for you,” the hanging “If I were to ask you out?” the absurd “I don’t know you,” and “I have to give you a book, because you’re the only thing in your eyes.” Some were just kind of plain cheesy, such as, “You are so beautiful that you make me feel better to see you.”

Possibly the wittiest line it generated is “You look like a thing and I love you.” More of the list can be found on Shane’s Tumblr account under the handle @lewisandquark.

The neural network may not be a Gigolo Joe but it may be on to something when it also generated “hello”—because doesn’t every encounter start with “Hello?” JB

RELATED STORY:

Computer neural network creates new unofficial Pokémon

Your subscription could not be saved. Please try again.
Your subscription has been successful.

Subscribe to our daily newsletter

By providing an email address. I agree to the Terms of Use and acknowledge that I have read the Privacy Policy.

TOPICS: machine learning, neural network
TAGS: machine learning, neural network

© Copyright 1997-2024 INQUIRER.net | All Rights Reserved

We use cookies to ensure you get the best experience on our website. By continuing, you are agreeing to our use of cookies. To find out more, please click this link.