InPHYNet: Leveraging attention-based multitask recurrent networks for multi-label physics text classification
作者：
Highlights：
• An attention based multi-task recurrent network is proposed for multilabel classification of school physics text data of 6th to 12th grade.
• The proposed solution is contextual and scalable.
• Although related to physics education, the proposed model is generalizable as an approach to other subjects.
• Importantly, it is observed that there is a correlation between information holding capacity and grade.
• The proposed work can contribute to building an intelligent tutoring system for science education.
摘要
•An attention based multi-task recurrent network is proposed for multilabel classification of school physics text data of 6th to 12th grade.•The proposed solution is contextual and scalable.•Although related to physics education, the proposed model is generalizable as an approach to other subjects.•Importantly, it is observed that there is a correlation between information holding capacity and grade.•The proposed work can contribute to building an intelligent tutoring system for science education.
论文关键词：Architectures for educational technology system,AI in physics education,Attention-based multitask recurrent network,Multi-label classification
论文评审过程：Received 15 April 2020, Revised 17 July 2020, Accepted 25 September 2020, Available online 13 October 2020, Version of Record 19 October 2020.
论文官网地址：https://doi.org/10.1016/j.knosys.2020.106487
