Around Valentine’s Day in the US and UK, these things called candy hearts (or conversation hearts or sweethearts) appear: small and sugary, bearing a simple, short Valentine’s message. There are only room for a few characters, so they read something like “LOVE YOU” or “CALL ME” or “BE MINE”.

I collected all the genuine heart messages I could find, and then gave them to a learning algorithm called a neural network. Given a set of data, a neural network will learn the patterns that let it imitate the original data - although its imitation is sometimes imperfect. The candy heart messages it produced… well, you be the judge.

The neural net did produce some that would pass for - and arguably improve upon - the standard messages.

DEAR ME MY MY LOVE BOT CUTE KISS MY BEAR LOVE BUN

Others were in the same spirit, but perhaps not quite as effective.

YOU A LOVE AM GOOD YOU ARE BABE ME MY <3 YOU ARE IT HEART ME SWEET PEAR COOL CUD FANCY MY HERO

Others were, um, strange. I don’t know what they mean, but some of them might work on me.

ALL HOVER OOG LOVE TEAM BEAR TIME HUG PIN A FACE YOU’RE ME SWOOL MAT BOG LOVE I HONKER HOW COT BEAR WIG FANG BE A GOOL TWEET UP BAT WIRLY OOT I WANDER TIME STAR

These will probably not be one of the standard messages anytime soon.

STANK LOVE SWEAT POO LOVE 2000 HOGSYEA HOLE HOW U HOT YOU ARE BOA CHERT FACE LICK SWEAT PEAR LOO GIN YOU A GOO LOOK BIG YOU ARE BAG U HACK

There was yet another category of message, a category you might be able to predict given the prevalence of four-letter words in the original dataset. The neural network thought of some nice new four-letter words to use. Unfortunately, some of those words already had other meanings. Let’s just say that the overall effect was surprisingly suggestive. Fill out the form here and I’ll send them to you.

Also, if you need more love help from the neural network, check out the pick up lines it wrote.

Heart pictures made using https://cryptogram.com/hearts/