Using Machine Learning to Improve UI/UX
summary
Claims that small JS models can do text prediction and attach a maliciousness rating to inputs using Node and Brain.js, with a single call to create a neural network and one function call to train and predict. Demo uses supervised examples, three hidden layers and live retraining from recent input to make real-time DOM updates. Two demos run during the session one demo works and one flakes so results are promising but brittle and dataset is tiny.
AI-written summary · about
chapters (10)
- 0:00Introduction
- 0:35Overview
- 1:29Neural Networks
- 3:03Predict User Intent
- 5:13Brain Juice
- 6:35Choosing Your Features
- 8:45Training Your Model
- 13:37Making Predictions
- 15:12Coding
- 20:19Key takeaways
related talks
21mUsing Machine Learning To Improve UI/UX
25mUsing Machine Learning To Improve UI/UX
44mArtificial Intelligence and Machine Learning in the browser
37mMachine Learning Principles, with JavaScript?
22mKeynote talk - The future of machine learning & JS
31mA Feasibility of On-Device ML for Adaptive User Interfaces in Frontend Development
24mThe Future of Machine Learning & Javascript
39mMachine Learning Magic for Your JavaScript Application