New machine learning technique could make mobile app notifications much smarter

Machine learning can stop intrusive notifications from bothering users

Two developers have created a new artificial intelligence system that aims to make managing push notifications easier.

Taiwan-based developers, TonTon Hsien-de Huang and Hung-Yu Kao, have figured out a way to input user data and habits into a neural network that’s able to learn what notifications users care about and what ones they would rather not show up on their device

They call this process of machine learning ‘Clicksequence-aware deeP neural network (DNN)-based Pop-uPs recOmmendation’ or C-3PO. It works by combining what notifications users interact with and what ones they don’t. From there, the algorithm shows you notifications that you constantly find useful and hides those that you rarely click on.

The hope for this project is to better facilitate a positive relationships between apps and users by showing them what they want and not bombarding them with intrusive, useless notifications.

Source: ArXiv.org Via: Engaget