[Feature Proposal] Implementing Adaptive Learning Analytics for Personalized Typing Practice

[ayushmaan-16]

@srevinsaju @chimosky I am proposing an enhancement to the Typing Turtle activity that introduces a lightweight, data-driven "Adaptive Learning" mechanism. The goal is to transform the activity from a static word-sequence game into an intelligent tutor that identifies and helps users overcome their specific typing weaknesses.

The Problem Currently, Typing Turtle delivers content in a linear or semi-random fashion. However, every learner has unique "finger-memory" hurdles—for instance, a student might consistently struggle with the reaching motion for 'p', 'q', or 'b'. In the current version, the game does not "notice" these repeated errors, missing an opportunity to provide targeted reinforcement where it is needed most.

The Solution: Adaptive AI Analytics I propose adding a session-based analytics layer that tracks performance in real-time.

Error Mapping: The activity will maintain a background frequency map (Python dictionary) of mistyped characters.

Weakness Identification: The logic will identify "Target Letters" (letters with an error rate above a specific threshold).

Dynamic Content Generation: The word-selection algorithm will be modified to "weight" words containing Target Letters more heavily.

Example: If a user misses 's' three times, the next five words generated will be filtered to include 's' (e.g., "grass", "small", "system") to force intentional practice.

Success Decay: As the user’s accuracy for a Target Letter improves, its weight will decrease, returning the word pool to its natural distribution.

[quozl]

Not yet. This activity is not being maintained, and is not available in the activity library. More mundane things have to happen first, like testing and release. Once those steps are up to date, then we can consider additions.