Supporting job mediator and job seeker through an actionable dashboard
summary
A prototype dashboard surfaces a recommender’s prediction (probability of finding work within 120 days from a 700,000+‑user model) and gives mediators interactive controls and three visualization levels (circle, bar, forest) so they can control the message, clarify recommendations, and support the mediator. A qualitative evaluation with 12 experienced mediators finds circles easiest to explain, mediators remove misleading parameters to avoid demotivation, and the tool successfully makes mediators “gatekeepers” of recommendations while revealing incomplete profile data.
AI-written summary · about
chapters (13)
- 0:00Introduction
- 0:24Background
- 2:30Interface Tip
- 3:23Goals
- 3:50Experience
- 4:21Preliminary study
- 6:12Design Goal 1
- 8:59Feedback
- 9:28Resource
- 10:53Clarification
- 11:35Support the mediator
- 14:02Clarify recommendations
- 14:20Questions
related talks
24mMind-Reading with Adaptive and Intelligent User Interfaces
34mHow to bridge the gap between Data Scientist & Decision Makers?
26mDesigning Anticipated User Experiences
16mMaking recommendations explainable
02h 52mLive wireframing: lean BI using AI
01h 21mFair Futures at Work Co Creation and AI driven Solutions in Governing
12mWhy am I seeing this: Democratizing End User Auditing for Online Content Recommendations
37mMind-Reading with adaptive UIs