Classifying Math Knowledge Components via Task-Adaptive Pre-Trained BERT
Saved in:
| Title: | Classifying Math Knowledge Components via Task-Adaptive Pre-Trained BERT |
|---|---|
| Language: | English |
| Authors: | Jia Tracy Shen, Michiharu Yamashita, Ethan Prihar, Neil Heffernan, Xintao Wu, Sean McGrew, Dongwon Lee |
| Source: | Grantee Submission. 2021Paper presented at the International Conference on Artificial Intelligence in Education (AIED) (2021). |
| Peer Reviewed: | Y |
| Page Count: | 12 |
| Publication Date: | 2021 |
| Sponsoring Agency: | National Science Foundation (NSF) Institute of Education Sciences (ED) Office of Elementary and Secondary Education (OESE) (ED), Education Innovation and Research (EIR) Office of Naval Research (ONR) (DOD) |
| Contract Number: | 1940236 1940076 1940093 1917808 1931523 1917713 1903304 1822830 1759229 R305A170137 R305A170243 R305A180401 U411B190024 N000141812768 |
| Document Type: | Speeches/Meeting Papers Reports - Research |
| Descriptors: | Mathematics Education, Knowledge Level, Video Technology, Educational Technology, Prediction, Classification, Accuracy, Intelligent Tutoring Systems, Natural Language Processing |
| DOI: | 10.1007/978-3-030-78292-4_33 |
| Abstract: | Educational content labeled with proper knowledge components (KCs) are particularly useful to teachers or content organizers. However, manually labeling educational content is labor intensive and error-prone. To address this challenge, prior research proposed machine learning based solutions to auto-label educational content with limited success. In this work, we significantly improve prior research by (1) expanding the input types to include KC descriptions, instructional video titles, and problem descriptions (i.e., three types of prediction task), (2) doubling the granularity of the prediction from 198 to 385 KC labels (i.e., more practical setting but much harder multinomial classification problem), (3) improving the prediction accuracies by 0.5-2.3% using Task-adaptive Pre-trained BERT, outperforming six baselines, and (4) proposing a simple evaluation measure by which we can recover 56-73% of mispredicted KC labels. All codes and data sets in the experiments are available at: https://github.com/tbs17/TAPT-BERT [This paper was published in: "AIED 2021, LNAI1 2748," edited by I. Roll et al., Springer Nature Switzerland AG, 2021, pp. 408-19.] |
| Abstractor: | As Provided |
| IES Funded: | Yes |
| Entry Date: | 2024 |
| Accession Number: | ED637573 |
| Database: | ERIC |
| Abstract: | Educational content labeled with proper knowledge components (KCs) are particularly useful to teachers or content organizers. However, manually labeling educational content is labor intensive and error-prone. To address this challenge, prior research proposed machine learning based solutions to auto-label educational content with limited success. In this work, we significantly improve prior research by (1) expanding the input types to include KC descriptions, instructional video titles, and problem descriptions (i.e., three types of prediction task), (2) doubling the granularity of the prediction from 198 to 385 KC labels (i.e., more practical setting but much harder multinomial classification problem), (3) improving the prediction accuracies by 0.5-2.3% using Task-adaptive Pre-trained BERT, outperforming six baselines, and (4) proposing a simple evaluation measure by which we can recover 56-73% of mispredicted KC labels. All codes and data sets in the experiments are available at: https://github.com/tbs17/TAPT-BERT [This paper was published in: "AIED 2021, LNAI1 2748," edited by I. Roll et al., Springer Nature Switzerland AG, 2021, pp. 408-19.] |
|---|---|
| DOI: | 10.1007/978-3-030-78292-4_33 |
