Bayesian Knowledge Tracing (BKT)
A knowledge tracing model proposed by Corbett and Anderson in 1994 that represents student knowledge state as a hidden binary variable. It characterizes the learning process through four parameters—initial mastery probability, learning rate, guess rate, and slip rate—and outputs the posterior probability that a student has reached a mastered state.
Timeline (last 90 days)
The history of intelligent tutoring systems (ITS) can be traced back to computer-assisted instruction in the 1970s; modern ITS combines knowledge graphs, Bayesian knowledge tracing and natural language processing technology, and can diagnose students' problem-solving processes step by step
BKT estimates the latent mastery probability, while neural models estimate the probability of a correct next response.
Bayesian Knowledge Tracing (BKT) is relatively insensitive to threshold changes, whereas neural network models become noticeably more selective as the threshold increases.
Bayesian Knowledge Tracing (BKT) was proposed by Corbett and Anderson in 1994.
In half of the tested scenarios, the thresholds selected by neural models and BKT differed by more than 0.10.
When the underlying KT model changes (e.g., migrating from BKT to DKT) or instructional priorities are adjusted, the mastery threshold must be recalibrated.
Adaptive learning systems typically use algorithms such as Item Response Theory (IRT) or Bayesian Knowledge Tracing to dynamically adjust content difficulty.
Core technologies of adaptive learning systems include machine learning methods such as Deep Knowledge Tracing (DKT) and Bayesian Knowledge Tracing (BKT), as well as Item Response Theory (IRT).
传统自适应学习平台如Knewton、DreamBox通常基于预设规则树或贝叶斯知识追踪模型（BKT），在固定内容空间内做路径选择
传统自适应学习平台如Knewton、DreamBox通常基于预设规则树或贝叶斯知识追踪模型（BKT）调整题目难度
All Facts (10)
The history of intelligent tutoring systems (ITS) can be traced back to computer-assisted instruction in the 1970s; modern ITS combines knowledge graphs, Bayesian knowledge tracing and natural language processing technology, and can diagnose students' problem-solving processes step by step
50%UnverifiedBKT estimates the latent mastery probability, while neural models estimate the probability of a correct next response.
50%UnverifiedBayesian Knowledge Tracing (BKT) is relatively insensitive to threshold changes, whereas neural network models become noticeably more selective as the threshold increases.
50%UnverifiedBayesian Knowledge Tracing (BKT) was proposed by Corbett and Anderson in 1994.
50%UnverifiedIn half of the tested scenarios, the thresholds selected by neural models and BKT differed by more than 0.10.
50%UnverifiedWhen the underlying KT model changes (e.g., migrating from BKT to DKT) or instructional priorities are adjusted, the mastery threshold must be recalibrated.
50%UnverifiedAdaptive learning systems typically use algorithms such as Item Response Theory (IRT) or Bayesian Knowledge Tracing to dynamically adjust content difficulty.
50%UnverifiedCore technologies of adaptive learning systems include machine learning methods such as Deep Knowledge Tracing (DKT) and Bayesian Knowledge Tracing (BKT), as well as Item Response Theory (IRT).
50%Unverified传统自适应学习平台如Knewton、DreamBox通常基于预设规则树或贝叶斯知识追踪模型（BKT），在固定内容空间内做路径选择
50%Unverified传统自适应学习平台如Knewton、DreamBox通常基于预设规则树或贝叶斯知识追踪模型（BKT）调整题目难度
50%