AI Snips
Chapters
Transcript
Episode notes
Tinder Automation
- Oscar Alseng, tired of manually swiping on Tinder, automated the process using a convolutional neural network (CNN).
- He trained the CNN on thousands of images, achieving 75% accuracy in predicting his swiping preferences.
Focus on Faces
- Alseng's CNN focused solely on facial features, using OpenCV and Haar cascades to extract faces and ensure consistency.
- This approach disregarded other profile aspects like body type or activities, concentrating only on facial attractiveness.
Dating Outcomes
- Alseng went on five dates resulting from his AI-driven Tinder experiment, revealing the practical application of his project.
- One date ended early due to the woman's negative perception of AI, highlighting societal views on the technology.