Abstract:Computerized Adaptive Testing refers to an online student ability testing system. It consists of Selection Algorithm and Cognitive Diagnosis Module (CDM). The Selection Algorithm aims to carefully select a small set of questions for the newly arrived student to answer, and the CDM aims to infer the ability of the newly arrived student based on his answers to the selected questions. The CDM faces the challenge of inferring student ability based on the short question-answer sequence. The state-of-the-art approach employs a meta-learning framework to solve this challenge. Specifically, in the outer-layer of the meta-learning framework, it learns an average student ability from all students as the initial ability for the newly arrived student, while in the inner-layer of the meta-learning framework, it infers the newly arrived student's ability from the short sequence. However, this approach suffers from two major limitations: Firstly, collaborative information between the newly arrived student and all students was not utilized. Secondly, the meta-learning framework has a slow convergence. This paper introduces Student-Adaptive Computerized Adaptive Testing (SACAT), which addresses these issues by: Firstly, use the short sequence of the newly arrived student to retrieve similar students from all students. Secondly, design a fast adaptive function that can adaptively generate an ability representation suitable for the newly arrived student from abilities of similar students. Through extensive experiments on three real-world student response datasets, we validate that SACAT can outperform existing CAT methods. Furthermore, through the fast adaptation, we show that SACAT achieves fast model convergence.

Ability Assessment Based on CAT in Adaptive Learning System

Fully Adaptive Framework: Neural Computerized Adaptive Testing for Online Education.

A Robust Computerized Adaptive Testing Approach in Educational Question Retrieval

Design and Application of a Response-Driven Adaptive Test System

SACAT: Student-Adaptive Computerized Adaptive Testing

The Development of Computerized Adaptive Testing with Cognitive Diagnosis for an English Achievement Test in China

Approach to Fitting Testlet for Computerized Adaptive Language Testing

Survey of Computerized Adaptive Testing: A Machine Learning Perspective

A Bounded Ability Estimation for Computerized Adaptive Testing

Stable and Diverse: A Unified Approach for Computerized Adaptive Testing

Efficiency of computerized adaptive testing with a cognitively designed item bank

Development and Validation of A Computer Adaptive Efl Test

A novel computerized adaptive testing framework with decoupled learning selector

Improving the 3PLM-Based Computerized Adaptive Testing System with Multi-Agent Item Bank.

The Computerized Adaptive Testing System Based on Item Response Theory

Diagnosing Primary Students’ Reading Progression: is Cognitive Diagnostic Computerized Adaptive Testing the Way Forward?

The Generation of Automated Learner Feedback Based on Individual Proficiency Levels

The Adaptive Learning System Based on Learning Style and Cognitive State

Personalized Recommendation in the Adaptive Learning System: the Role of Adaptive Testing Technology

Research on Item Response Theory Based System of Computerized Adaptive Testing

A Comparison of a Computerised Adaptive Test for Mathematics Instruction with the Classical Test