Provisionary model of feedback
This model does not reference Scaffold-specific entities such as Candidate, SignalDetector, or any ranking algorithm
Raw inputs are ingested and (minimally) cleaned by the frontend (web, etl, etc.) to produce performance information blocks for processing by signal detectors
- Input to Signal Detectors
- Tabular: performer, measure, target performance month, time/period (for each performance set), performance metrics, comparators, ...
- There may be alternatives for specifying the target performance month, which is a parameter of the Scaffold processing and not a feature of the performance information itself.
Signal detectors process performance information to extract motivating information (trends, gaps, etc.) related to performance and (partially) quantify its contribution (trend length, gap size) to the motivational potential of feedback messages. Note: some moderators of the motivational potential of a feedback messages (preferences, recency) are currently determined in the message selection component ("esteemer", using recency and preferences).