Difference between revisions of "Feature-Aware Student knowledge Tracing (FAST)"

From PAWS Lab
Jump to: navigation, search
Line 2: Line 2:
  
 
* Download the source code [http://ml-smores.github.io/fast/ here]
 
* Download the source code [http://ml-smores.github.io/fast/ here]
* Read the conference presentation slides [http://www.slideshare.net/huangyun/fast-presentation-48711687 here]
+
* Read the EDM 2014 conference presentation slides [http://www.slideshare.net/huangyun/fast-presentation-48711687 here]
 
* Read the tutorial slides presented in EDM 2015 workshop [http://www.slideshare.net/huangyun/2015edm-featureaware-student-knowledge-tracing-tutorial here]
 
* Read the tutorial slides presented in EDM 2015 workshop [http://www.slideshare.net/huangyun/2015edm-featureaware-student-knowledge-tracing-tutorial here]

Revision as of 03:40, 4 April 2016

Feature-Aware Student knowledge Tracing (FAST) is a novel, efficient student model created by PAWs lab and Pearson with the state-of-the-art predictive performance. It allows general, flexible features into Knowledge Tracing, which is the most popular student model. For the first time it unifies existed specially designed student models based on Knowledge Tracing framework with integrated advantages. We demonstrate FAST’s flexibility with examples of feature sets that are relevant to a wide audience: we haved uses features in FAST to model (i) multiple subskills, (ii) temporal Item Response Theory, and (iii) expert knowledge. Compared with Knowledge Tracing, (1) it improves up to 25% in classification performance, (2) it generates more interpretable, consistent parameters, and (3) it is 300 times faster. In a follow-up study, we compared FAST to the best paper model (a single-purpose model) of the same year with favorable results while FAST is designed as a general-purpose model. The main paper was nominated for the Best Paper Award in a top-tier conference in 2014. Both the main and the follow-up papers were cited by top researchers in the field from Carnegie Mellon University, Stanford, Cornell, ETH Zurich (etc.) with in total 34 citations since 2014 (till 04/2016).

  • Download the source code here
  • Read the EDM 2014 conference presentation slides here
  • Read the tutorial slides presented in EDM 2015 workshop here