Bayesian Knowledge Tracing

This code is based on Bayesian Knowledge Tracing Parameter Fitting by Simulated Annealing code from Miller, W. L., Baker, R.S. and Rossi, L.M. (2014), which is in turn based on BKT Brute Force code from Baker, Corbett, and Aleven (2008).

Install

Usage

(require [bkt.core :refer :all])
(fit-model [[true true] [true true false]] guess-slip-bounds)

Each nested list passed to fit model constitutes a series of ordered correct(true)/incorrect(false) interactions from a student.

The code fits a single skill at a time.

See examples.clj to replicate the example shown in Miller et al (2014). This should give identical results.

Current TODO List

• Tests
• Increase speed (warning: this is a bit slower than the Java one)
• Clojurescript

References:

1. Baker, R.S.J.d., Corbett, A.T., Aleven, V. (2008) More Accurate Student Modeling Through Contextual Estimation of Slip and Guess Probabilities in Bayesian Knowledge Tracing. Proceedings of the 9th International Conference on Intelligent Tutoring Systems, 406-415.
2. Miller, W. L., Baker, R.S. and Rossi, L.M. (2014) Unifying Computer-Based Assessment Across Conceptual Instruction, Problem-Solving, and Digital Games. Technology, Knowledge and Learning. 165-181. (preprint)