A novel computerized adaptive testing framework with decoupled learning selector
Abstract Computerized adaptive testing (CAT) targets to accurately assess the student’s proficiency in the required subject/area. The key issue is how to design a question selector that adaptively selects the best-suited questions for each student based on previous performance step by step. Most exi...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Springer
2023-03-01
|
| Series: | Complex & Intelligent Systems |
| Subjects: | |
| Online Access: | https://doi.org/10.1007/s40747-023-01019-1 |
| _version_ | 1827808106844782592 |
|---|---|
| author | Haiping Ma Yi Zeng Shangshang Yang Chuan Qin Xingyi Zhang Limiao Zhang |
| author_facet | Haiping Ma Yi Zeng Shangshang Yang Chuan Qin Xingyi Zhang Limiao Zhang |
| author_sort | Haiping Ma |
| collection | DOAJ |
| description | Abstract Computerized adaptive testing (CAT) targets to accurately assess the student’s proficiency in the required subject/area. The key issue is how to design a question selector that adaptively selects the best-suited questions for each student based on previous performance step by step. Most existing question selectors execute via greedy metric functions (e.g., question information and uncertainty), which can not effectively capture data characteristics. There also exist learning-based question selectors that redefine the CAT problem as a bilevel optimization problem, where the parameter learning of the question selector and the student proficiency estimation model are coupled, which is not flexible enough. To this end, in this paper, we propose a novel CAT framework with Decoupled Learning selector (DL-CAT). Specifically, we first use the currently estimated student ability and question characteristics as input and design a deep learning-based question selector to predict question selection scores. Then, to address the issue that there is no ground truth to measure the quality of the selected question, an approximate ground-truth and a pairwise rank loss function are specially designed to update the parameters of the question selector independently. Extensive experiments on two real datasets demonstrate that our proposed DL-CAT has certain advantages in effectiveness and significant advantages in efficiency. |
| first_indexed | 2024-03-11T22:07:36Z |
| format | Article |
| id | doaj.art-17b1f3d468ea49d6bc18bc8ef8ab38d0 |
| institution | Directory Open Access Journal |
| issn | 2199-4536 2198-6053 |
| language | English |
| last_indexed | 2024-03-11T22:07:36Z |
| publishDate | 2023-03-01 |
| publisher | Springer |
| record_format | Article |
| series | Complex & Intelligent Systems |
| spelling | doaj.art-17b1f3d468ea49d6bc18bc8ef8ab38d02023-09-24T11:35:11ZengSpringerComplex & Intelligent Systems2199-45362198-60532023-03-01955555556610.1007/s40747-023-01019-1A novel computerized adaptive testing framework with decoupled learning selectorHaiping Ma0Yi Zeng1Shangshang Yang2Chuan Qin3Xingyi Zhang4Limiao Zhang5Institutes of Physical Science and Information Technology, Anhui UniversityInstitutes of Physical Science and Information Technology, Anhui UniversitySchool of Computer Science and Technology, Anhui UniversityBaidu Talent Intelligence Center, Baidu Inc. BeijingSchool of Computer Science and Technology, Anhui UniversityInformation Materials and Intelligent Sensing Laboratory of Anhui Province, Anhui UniversityAbstract Computerized adaptive testing (CAT) targets to accurately assess the student’s proficiency in the required subject/area. The key issue is how to design a question selector that adaptively selects the best-suited questions for each student based on previous performance step by step. Most existing question selectors execute via greedy metric functions (e.g., question information and uncertainty), which can not effectively capture data characteristics. There also exist learning-based question selectors that redefine the CAT problem as a bilevel optimization problem, where the parameter learning of the question selector and the student proficiency estimation model are coupled, which is not flexible enough. To this end, in this paper, we propose a novel CAT framework with Decoupled Learning selector (DL-CAT). Specifically, we first use the currently estimated student ability and question characteristics as input and design a deep learning-based question selector to predict question selection scores. Then, to address the issue that there is no ground truth to measure the quality of the selected question, an approximate ground-truth and a pairwise rank loss function are specially designed to update the parameters of the question selector independently. Extensive experiments on two real datasets demonstrate that our proposed DL-CAT has certain advantages in effectiveness and significant advantages in efficiency.https://doi.org/10.1007/s40747-023-01019-1Computerized adaptive testingQuestion selectorStudent assessmentLearning selector |
| spellingShingle | Haiping Ma Yi Zeng Shangshang Yang Chuan Qin Xingyi Zhang Limiao Zhang A novel computerized adaptive testing framework with decoupled learning selector Complex & Intelligent Systems Computerized adaptive testing Question selector Student assessment Learning selector |
| title | A novel computerized adaptive testing framework with decoupled learning selector |
| title_full | A novel computerized adaptive testing framework with decoupled learning selector |
| title_fullStr | A novel computerized adaptive testing framework with decoupled learning selector |
| title_full_unstemmed | A novel computerized adaptive testing framework with decoupled learning selector |
| title_short | A novel computerized adaptive testing framework with decoupled learning selector |
| title_sort | novel computerized adaptive testing framework with decoupled learning selector |
| topic | Computerized adaptive testing Question selector Student assessment Learning selector |
| url | https://doi.org/10.1007/s40747-023-01019-1 |
| work_keys_str_mv | AT haipingma anovelcomputerizedadaptivetestingframeworkwithdecoupledlearningselector AT yizeng anovelcomputerizedadaptivetestingframeworkwithdecoupledlearningselector AT shangshangyang anovelcomputerizedadaptivetestingframeworkwithdecoupledlearningselector AT chuanqin anovelcomputerizedadaptivetestingframeworkwithdecoupledlearningselector AT xingyizhang anovelcomputerizedadaptivetestingframeworkwithdecoupledlearningselector AT limiaozhang anovelcomputerizedadaptivetestingframeworkwithdecoupledlearningselector AT haipingma novelcomputerizedadaptivetestingframeworkwithdecoupledlearningselector AT yizeng novelcomputerizedadaptivetestingframeworkwithdecoupledlearningselector AT shangshangyang novelcomputerizedadaptivetestingframeworkwithdecoupledlearningselector AT chuanqin novelcomputerizedadaptivetestingframeworkwithdecoupledlearningselector AT xingyizhang novelcomputerizedadaptivetestingframeworkwithdecoupledlearningselector AT limiaozhang novelcomputerizedadaptivetestingframeworkwithdecoupledlearningselector |