Smell Pittsburgh: community-empowered mobile smell reporting system
summary
Smell Pittsburgh is a mobile app plus public map that collects GPS locations and user smell ratings, sending reports to the health department and visualizing 17,000 reports from 3,900 users over two years; it adds push notifications and a machine model (random forest) that achieves about 0.7 F-score and a decision-tree explanation that links wind + hydrogen sulfide readings, cutting negatives from ~16,000 to ~300. A 25-person survey shows self-efficacy rises ~0.5 Likert and community use helps prompt regulatory action.
AI-written summary · about
chapters (6)
- 0:00Start
- 0:49Background about Pittsburgh
- 3:59The System Usage Study
- 5:24Symptoms
- 10:41Survey Study of 25 Participants
- 12:35Takeaway Messages
related talks
39mSmelly London: visualising historical smells.
26mActionable Data to Protect the People
26mThe Data I Breathe
16mKeynote: Open Source as Endurance Infrastructure for Environmental Justice
53mADEM (Breathe)
14mTelling Traces
26mThe Quantified Self in the Smart City: Geo-Visualizing the Open Data of YOU
18mAir quality sensing