Show HN: Predict Mailchimp subject line open rate with AI (grewai.com) 7 points by ihmissuti 5y ago ↗ HN
[–] thefedoration 5y ago ↗ Very cool. You should make this for twitter. [–] ihmissuti 5y ago ↗ It's coming ;)
[–] vuijws 5y ago ↗ Is this doable to other channels? [–] ihmissuti 5y ago ↗ yup that's on my list. Mailchimp is the first channel I'm testing but I'm thinking about adding eg. ad channels.
[–] ihmissuti 5y ago ↗ yup that's on my list. Mailchimp is the first channel I'm testing but I'm thinking about adding eg. ad channels.
[–] codeforfun 5y ago ↗ Whats the prediction accuracy? Neat tool anyway! [–] ihmissuti 5y ago ↗ Thanks! Depends on the data (size and quality). We've been able to get 85% accuracy if the data is decent.
[–] ihmissuti 5y ago ↗ Thanks! Depends on the data (size and quality). We've been able to get 85% accuracy if the data is decent.
[–] apinamk 5y ago ↗ Good call to make it actually affordable. [–] ihmissuti 5y ago ↗ Yea thanks! I've been selling basically same stuff to enterprises (customized UI's and algorithms) but I didn't want to overprice this one.
[–] ihmissuti 5y ago ↗ Yea thanks! I've been selling basically same stuff to enterprises (customized UI's and algorithms) but I didn't want to overprice this one.
[–] icebin_icebin 5y ago ↗ Cool! How does the AI work? How does it make it's predictions? [–] ihmissuti 5y ago ↗ In short there's a stack of prediction models that analyze the data: deep-learning neural network, keyword-based machine learning model, and shortly also I'll include a BERT NLP analyzer
[–] ihmissuti 5y ago ↗ In short there's a stack of prediction models that analyze the data: deep-learning neural network, keyword-based machine learning model, and shortly also I'll include a BERT NLP analyzer
10 comments
[ 37.8 ms ] story [ 35.7 ms ] thread