SEDNN: Shared and enhanced deep neural network model for cross-prompt automated essay scoring

作者：

Highlights：

• Propose a new approach to extract maximum rating knowledge from source prompts.

• Design an adversarial training process to extract transferable features.

• Enhance the final rating model by additional prompt-dependent features.

• Significant improvement in cross-prompt essay scoring over state-of-the-art methods.

摘要

•Propose a new approach to extract maximum rating knowledge from source prompts.•Design an adversarial training process to extract transferable features.•Enhance the final rating model by additional prompt-dependent features.•Significant improvement in cross-prompt essay scoring over state-of-the-art methods.

论文关键词：Automated essay scoring,Essay evaluation,Cross-prompt knowledge transferring,Natural language processing,Deep neural network

论文评审过程：Received 6 January 2020, Revised 10 July 2020, Accepted 29 September 2020, Available online 1 October 2020, Version of Record 3 October 2020.

论文官网地址：https://doi.org/10.1016/j.knosys.2020.106491