---
title: "写作智能评测研究综述和发展前景 — 薛嗣媛 2023"
description: "This survey reviews automated essay scoring (AES) and gives particular attention to Chinese, where research is still at an early stage compared with…"
authors: ["薛嗣媛", "周建设", "任福继"]
year: 2023
venue: "中文信息学报 (Journal of Chinese Information Processing)"
source: "http://jcip.cipsc.org.cn/article/id/zwxxxb_3472"
licence: "Free access (jcip.cipsc.org.cn), no open licence stated — readonly; © 中文信息学报"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/papers/xue-2023-systematic-automated-essay/"
---
[Elamur Research](https://md.research.elamur.ai/) · [this page in HTML](https://research.elamur.ai/free/papers/xue-2023-systematic-automated-essay/)

[home](https://md.research.elamur.ai/) / [papers](https://md.research.elamur.ai/free/papers/) / [2023](https://md.research.elamur.ai/free/papers/2023/page/7/) / 写作智能评测研究综述和发展前景

# 写作智能评测研究综述和发展前景

**Authors:** 薛嗣媛, 周建设, 任福继 · **Year:** 2023 · **Venue:** 中文信息学报 (Journal of Chinese Information Processing) · **Licence:** Free access (jcip.cipsc.org.cn), no open licence stated — readonly; © 中文信息学报

**Links:** [Publisher's page](http://jcip.cipsc.org.cn/article/id/zwxxxb_3472) · [PDF](http://jcip.cipsc.org.cn/cn/article/pdf/preview/zwxxxb_3472.pdf)

## Summary
This survey reviews automated essay scoring (AES) and gives particular attention to Chinese, where research is still at an early stage compared with English. The authors argue that earlier reviews were written mainly by computer scientists, focused on English texts, and largely ignored pre-trained language models and Chinese data. The paper divides AES history into four periods, from Page's PEG system to recent deep learning work. It sorts the task into four settings: holistic or trait scoring, each done either within one prompt or across prompts. It also lists English benchmark corpora and common agreement metrics. Techniques fall into two groups: feature-based learning (regression, ranking and classification) and neural approaches (recurrent, coherence-aware, pre-trained and multimodal). A comparison of results on ASAP shows that deep models now beat feature-engineered ones and lead the field. The survey then covers Chinese AES separately for native and non-native writers, describes open problems and suggests future directions. It works as a structured entry point for building essay assessment tools in less-studied languages.

## Key points
- **Four historical stages:**
  - 1966–1990: linear-regression models, starting with PEG.
  - 1990–2000: statistical systems such as E-rater, Intelligent Essay Assessor and IntelliMetric.
  - 2000–2013: wide deployment and a shift to multi-aspect feedback (MyAccess!, Bookette, LightSIDE, etc.).
  - 2013–2022: end-to-end deep learning.
- **English datasets reviewed:** CREE, CS, ICLE, CLC-FCE, BEETLE, ASAP (AES and SAS), ASAP++, RTA, SCIENTSBANK, TOEFL11, Powergrading, AAE and FCE. Each is described by writer population, size, number of prompts, task type and score range.
- **Evaluation metrics:** QWK, MAE, RMSE and Pearson's correlation coefficient (PCC).
- **Single-prompt neural models:**
  - Taghipour & Ng's recurrent DNN-AES.
  - Score-specific word embeddings (SSWEs) from Alikaniotis et al.
  - Dong et al.'s hierarchical CNN with attention.
  - SKIPFLOW, which models coherence.
  - BERT fine-tuning, including Yang et al.'s multi-loss variant (MSE plus ranking loss), reported as best on ASAP.
  - Gao et al.'s bimodal scorer for handwritten essays, which combines Fast RCNN image features, OCR text and a VisualBERT/VisualRoBERTa/VisualALBERT ensemble.
- **Cross-prompt methods:**
  - TDNN, which builds pseudo-labels with RankSVM.
  - SEDNN, which uses an adversarial shared model.
  - Cao et al.'s self-supervised sentence reordering and noise identification.
  - Ridley et al.'s cross-prompt multi-trait scoring task.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

Fields: [Linguistics & languages](https://md.research.elamur.ai/free/shelves/linguistics/) · [Natural language processing](https://md.research.elamur.ai/free/shelves/nlp/)

## Cite

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@article{2023,
  title = {{写作智能评测研究综述和发展前景}},
  author = {薛嗣媛 and 周建设 and 任福继},
  journal = {中文信息学报 (Journal of Chinese Information Processing)},
  year = {2023},
  url = {http://jcip.cipsc.org.cn/article/id/zwxxxb_3472}
}
```

- [← previous 中文句子级性别无偏数据集构建及预训练语言模型的性别偏度评估](https://md.research.elamur.ai/free/papers/zhao-2023-construction-chinese-sentence/)
- [next → 名詞役割入れ替えに頑健な日本語含意関係認識](https://md.research.elamur.ai/free/papers/amemiya-2023-japanese-textual-entailment-robust/)
- [↑ back to papers of 2023](https://md.research.elamur.ai/free/papers/2023/page/7/)

## References

26 references, 0 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. 682. [ 40] MAYFELD E,BLACK A W. Shou l d youf i ne - t une //Pr BERTf o rau t oma t ede s s ays c o r i ng? [ C] o c e ed - i ngso ft he Wo r kshoponI nnova t i veUs eo fNLPf o r Bu i l d i ngEduc a t i ona lApp l i c a t i ons,2020:151 - 162. [find](https://scholar.google.com/scholar?q=682%2E+%5B+40%5D+MAYFELD+E%2CBLACK+A+W%2E+Shou+l+d+youf+i+ne+%2D+t+une+%2F%2FPr+BERTf+o+rau+t+oma+t+ede+s+s+ays+c+o+r+i+ng%3F+%5B+C%5D+o+c+e+ed+%2D+i+ngso+ft+he+Wo+r+kshoponI+nnova+t+i+veUs+eo+fNLPf+o+r+Bu+i+l+d+i+ngEduc+a+t)
2. [ 41] RODRIGUEZP U,JAFARIA,ORMEROD C M. Language mode l sandau t oma t ede s s ays c o r i ng[ J]. a rXi v:190909482,2019. [find](https://scholar.google.com/scholar?q=%5B+41%5D+RODRIGUEZP+U%2CJAFARIA%2CORMEROD+C+M%2E+Language+mode+l+sandau+t+oma+t+ede+s+s+ays+c+o+r+i+ng%5B+J%5D%2E+a+rXi+v%3A190909482%2C2019%2E)
3. [ 42] YANG R,CAOJ,WEN Z,e ta l.Enhanc i ngau t o - ma t ede s s ays c o r i ngpe r f o rmanc ev i af i ne t un i ngp r e - t r a i nedl anguage mode l swi t hc omb i na t i ono fr eg r e s - //Pr s i on and r ank i ng [C ] o c e ed i ngs o f Emp i r i c a l Me t hods i n Na t u r a l Language Pr o c… [find](https://scholar.google.com/scholar?q=%5B+42%5D+YANG+R%2CCAOJ%2CWEN+Z%2Ce+ta+l%2EEnhanc+i+ngau+t+o+%2D+ma+t+ede+s+s+ays+c+o+r+i+ngpe+r+f+o+rmanc+ev+i+af+i+ne+t+un+i+ngp+r+e+%2D+t+r+a+i+nedl+anguage+mode+l+swi+t+hc+omb+i+na+t+i+ono+fr+eg+r+e+s+%2D+%2F%2FPr+s+i+)
4. [ 43] GAOJ,YANG Q,ZHANGY,e ta l.Ab i -moda lau - t oma t ede s s ays c o r i ngsy s t emf o rhandwr i t t ene s s ay s [ //Pr C] o c e ed i ngso ft heCon f e r enc eo fI n t e r na t i ona l J o i n tCon f e r enc eonNeu r a lNe two r ks,2021:1 - 8. [find](https://scholar.google.com/scholar?q=%5B+43%5D+GAOJ%2CYANG+Q%2CZHANGY%2Ce+ta+l%2EAb+i+%2Dmoda+lau+%2D+t+oma+t+ede+s+s+ays+c+o+r+i+ngsy+s+t+emf+o+rhandwr+i+t+t+ene+s+s+ay+s+%5B+%2F%2FPr+C%5D+o+c+e+ed+i+ngso+ft+heCon+f+e+r+enc+eo+fI+n+t+e+r+na+t+i+ona+l+J+o+i+n+t)
5. [ 44] 梁茂成 . 中国学生英语作文自动评分模型的构建[ D]. 南京:南京大学博士学位论文, 2005. 799. [ 45] 唐芳,庄翠娟,巩艺超 . 作 文 自 动 评 分 系 统 在 大 学 英 语写作教学 中 的 应 用:以 句 酷 批 改 网 为 例 [ J]. 海 外 , : 英语,2017 20 48 - 49. [find](https://scholar.google.com/scholar?q=%5B+44%5D+%E6%A2%81%E8%8C%82%E6%88%90+%2E+%E4%B8%AD%E5%9B%BD%E5%AD%A6%E7%94%9F%E8%8B%B1%E8%AF%AD%E4%BD%9C%E6%96%87%E8%87%AA%E5%8A%A8%E8%AF%84%E5%88%86%E6%A8%A1%E5%9E%8B%E7%9A%84%E6%9E%84%E5%BB%BA%5B+D%5D%2E+%E5%8D%97%E4%BA%AC%3A%E5%8D%97%E4%BA%AC%E5%A4%A7%E5%AD%A6%E5%8D%9A%E5%A3%AB%E5%AD%A6%E4%BD%8D%E8%AE%BA%E6%96%87%2C+2005%2E+799%2E+%5B+45%5D+%E5%94%90%E8%8A%B3%2C%E5%BA%84%E7%BF%A0%E5%A8%9F%2C%E5%B7%A9%E8%89%BA%E8%B6%85+%2E+%E4%BD%9C+%E6%96%87+%E8%87%AA+%E5%8A%A8+%E8%AF%84+%E5%88%86+%E7%B3%BB+%E7%BB%9F+%E5%9C%A8+%E5%A4%A7+%E5%AD%A6+%E8%8B%B1+%E8%AF%AD%E5%86%99%E4%BD%9C%E6%95%99%E5%AD%A6+%E4%B8%AD+%E7%9A%84+%E5%BA%94+%E7%94%A8%3A%E4%BB%A5+%E5%8F%A5+%E9%85%B7+%E6%89%B9+%E6%94%B9+%E7%BD%91+%E4%B8%BA+%E4%BE%8B+%5B+J%5D%2E+%E6%B5%B7+%E5%A4%96+%2C+%3A+%E8%8B%B1%E8%AF%AD%2C2017+20+48+%2D+49%2E)
6. [ 利用深层语言分析改进 46] 魏思,巩 捷 甫,王 士 进,等 . 中文作文自 动 评 分 方 法 [ 中 文 信 息 学 报, J]. 2022, 36 ( 04):111 - 123. [find](https://scholar.google.com/scholar?q=%5B+%E5%88%A9%E7%94%A8%E6%B7%B1%E5%B1%82%E8%AF%AD%E8%A8%80%E5%88%86%E6%9E%90%E6%94%B9%E8%BF%9B+46%5D+%E9%AD%8F%E6%80%9D%2C%E5%B7%A9+%E6%8D%B7+%E7%94%AB%2C%E7%8E%8B+%E5%A3%AB+%E8%BF%9B%2C%E7%AD%89+%2E+%E4%B8%AD%E6%96%87%E4%BD%9C%E6%96%87%E8%87%AA+%E5%8A%A8+%E8%AF%84+%E5%88%86+%E6%96%B9+%E6%B3%95+%5B+%E4%B8%AD+%E6%96%87+%E4%BF%A1+%E6%81%AF+%E5%AD%A6+%E6%8A%A5%2C+J%5D%2E+2022%2C+36+%28+04%29%3A111+%2D+123%2E)
7. [ 47] IN 课堂智能教育平 台 . IN 课 堂:语 文 作 文 智 能 批 改 // 教育迈向 智 能 化 阶 段 [ N].h t t i nke t ang. c om/v8/ p: news_de t a i l_00. h tml,2018. [find](https://scholar.google.com/scholar?q=%5B+47%5D+IN+%E8%AF%BE%E5%A0%82%E6%99%BA%E8%83%BD%E6%95%99%E8%82%B2%E5%B9%B3+%E5%8F%B0+%2E+IN+%E8%AF%BE+%E5%A0%82%3A%E8%AF%AD+%E6%96%87+%E4%BD%9C+%E6%96%87+%E6%99%BA+%E8%83%BD+%E6%89%B9+%E6%94%B9+%2F%2F+%E6%95%99%E8%82%B2%E8%BF%88%E5%90%91+%E6%99%BA+%E8%83%BD+%E5%8C%96+%E9%98%B6+%E6%AE%B5+%5B+N%5D%2Eh+t+t+i+nke+t+ang%2E+c+om%2Fv8%2F+p%3A+news%5Fde+t+a+i+l%5F00%2E+h+tml%2C2018%2E)
8. [ 基于回归分析 的 中 文 作 文 自 动 评 分 技 术 研 究 48] 陈一乐 . [ D]. 哈尔滨:哈尔滨工业大学硕士学位论文, 2016. [find](https://scholar.google.com/scholar?q=%5B+%E5%9F%BA%E4%BA%8E%E5%9B%9E%E5%BD%92%E5%88%86%E6%9E%90+%E7%9A%84+%E4%B8%AD+%E6%96%87+%E4%BD%9C+%E6%96%87+%E8%87%AA+%E5%8A%A8+%E8%AF%84+%E5%88%86+%E6%8A%80+%E6%9C%AF+%E7%A0%94+%E7%A9%B6+48%5D+%E9%99%88%E4%B8%80%E4%B9%90+%2E+%5B+D%5D%2E+%E5%93%88%E5%B0%94%E6%BB%A8%3A%E5%93%88%E5%B0%94%E6%BB%A8%E5%B7%A5%E4%B8%9A%E5%A4%A7%E5%AD%A6%E7%A1%95%E5%A3%AB%E5%AD%A6%E4%BD%8D%E8%AE%BA%E6%96%87%2C+2016%2E)
9. [ 使用 潜 语 义 分 析 的 汉 语 作 文 自 动 评 分 49] 曹亦薇,杨晨 . 研究[ J]. 考试研究,2007,1( 2):65 - 73. [find](https://scholar.google.com/scholar?q=%5B+%E4%BD%BF%E7%94%A8+%E6%BD%9C+%E8%AF%AD+%E4%B9%89+%E5%88%86+%E6%9E%90+%E7%9A%84+%E6%B1%89+%E8%AF%AD+%E4%BD%9C+%E6%96%87+%E8%87%AA+%E5%8A%A8+%E8%AF%84+%E5%88%86+49%5D+%E6%9B%B9%E4%BA%A6%E8%96%87%2C%E6%9D%A8%E6%99%A8+%2E+%E7%A0%94%E7%A9%B6%5B+J%5D%2E+%E8%80%83%E8%AF%95%E7%A0%94%E7%A9%B6%2C2007%2C1%28+2%29%3A65+%2D+73%2E)
10. [ 面向语文作文 自 动 评 阅 的 修 辞 手 法 识 别 系 统 50] 巩捷甫 . 的设计与实现[ D]. 哈 尔 滨:哈 尔 滨 工 业 大 学 硕 士 学 位论文,2016. [find](https://scholar.google.com/scholar?q=%5B+%E9%9D%A2%E5%90%91%E8%AF%AD%E6%96%87%E4%BD%9C%E6%96%87+%E8%87%AA+%E5%8A%A8+%E8%AF%84+%E9%98%85+%E7%9A%84+%E4%BF%AE+%E8%BE%9E+%E6%89%8B+%E6%B3%95+%E8%AF%86+%E5%88%AB+%E7%B3%BB+%E7%BB%9F+50%5D+%E5%B7%A9%E6%8D%B7%E7%94%AB+%2E+%E7%9A%84%E8%AE%BE%E8%AE%A1%E4%B8%8E%E5%AE%9E%E7%8E%B0%5B+D%5D%2E+%E5%93%88+%E5%B0%94+%E6%BB%A8%3A%E5%93%88+%E5%B0%94+%E6%BB%A8+%E5%B7%A5+%E4%B8%9A+%E5%A4%A7+%E5%AD%A6+%E7%A1%95+%E5%A3%AB+%E5%AD%A6+%E4%BD%8D%E8%AE%BA%E6%96%87%2C2016%2E)
11. [ 51] WEIS,DONG W,RUI J IF,e ta l.Di s c ou r s emodei - //Pr den t i f i c a t i oni ne s s ay s[ C] o c e ed i ngso ft he55 t h Annua lMe e t i ngo ft heAs s o c i a t i onf o rCompu t a t i ona l L i ngu i s t i c, 2017:112 - 122. [find](https://scholar.google.com/scholar?q=%5B+51%5D+WEIS%2CDONG+W%2CRUI+J+IF%2Ce+ta+l%2EDi+s+c+ou+r+s+emodei+%2D+%2F%2FPr+den+t+i+f+i+c+a+t+i+oni+ne+s+s+ay+s%5B+C%5D+o+c+e+ed+i+ngso+ft+he55+t+h+Annua+lMe+e+t+i+ngo+ft+heAs+s+o+c+i+a+t+i+onf+o+rCompu+t+a+t+i+ona+l+L)
12. [ 52] 马晓丽,刘杰,周 建 设,等 . 一 种 中 小 学 汉 语 作 文 表 现手法分 类 方 法 [ J]. 计 算 机 应 用 与 软 件,2018,35 ( 10):49 - 54. [find](https://scholar.google.com/scholar?q=%5B+52%5D+%E9%A9%AC%E6%99%93%E4%B8%BD%2C%E5%88%98%E6%9D%B0%2C%E5%91%A8+%E5%BB%BA+%E8%AE%BE%2C%E7%AD%89+%2E+%E4%B8%80+%E7%A7%8D+%E4%B8%AD+%E5%B0%8F+%E5%AD%A6+%E6%B1%89+%E8%AF%AD+%E4%BD%9C+%E6%96%87+%E8%A1%A8+%E7%8E%B0%E6%89%8B%E6%B3%95%E5%88%86+%E7%B1%BB+%E6%96%B9+%E6%B3%95+%5B+J%5D%2E+%E8%AE%A1+%E7%AE%97+%E6%9C%BA+%E5%BA%94+%E7%94%A8+%E4%B8%8E+%E8%BD%AF+%E4%BB%B6%2C2018%2C35+%28+10%29%3A49+%2D+54%2E)
13. [ 面向作文自动评分的优美 53] 付瑞 吉, 王 栋, 王 士 进, 等 . 句识别[ J]. 中文信息学报,2018,32( 06):88 - 97. [find](https://scholar.google.com/scholar?q=%5B+%E9%9D%A2%E5%90%91%E4%BD%9C%E6%96%87%E8%87%AA%E5%8A%A8%E8%AF%84%E5%88%86%E7%9A%84%E4%BC%98%E7%BE%8E+53%5D+%E4%BB%98%E7%91%9E+%E5%90%89%2C+%E7%8E%8B+%E6%A0%8B%2C+%E7%8E%8B+%E5%A3%AB+%E8%BF%9B%2C+%E7%AD%89+%2E+%E5%8F%A5%E8%AF%86%E5%88%AB%5B+J%5D%2E+%E4%B8%AD%E6%96%87%E4%BF%A1%E6%81%AF%E5%AD%A6%E6%8A%A5%2C2018%2C32%28+06%29%3A88+%2D+97%2E)
14. [ 基于 文 采 特 征 的 高 考 作 文 自 动 评 54] 刘明杨, 秦兵, 刘挺 . 分[ 智能计算机与应用,2016,6( J]. 01):1 - 4. [find](https://scholar.google.com/scholar?q=%5B+%E5%9F%BA%E4%BA%8E+%E6%96%87+%E9%87%87+%E7%89%B9+%E5%BE%81+%E7%9A%84+%E9%AB%98+%E8%80%83+%E4%BD%9C+%E6%96%87+%E8%87%AA+%E5%8A%A8+%E8%AF%84+54%5D+%E5%88%98%E6%98%8E%E6%9D%A8%2C+%E7%A7%A6%E5%85%B5%2C+%E5%88%98%E6%8C%BA+%2E+%E5%88%86%5B+%E6%99%BA%E8%83%BD%E8%AE%A1%E7%AE%97%E6%9C%BA%E4%B8%8E%E5%BA%94%E7%94%A8%2C2016%2C6%28+J%5D%2E+01%29%3A1+%2D+4%2E)
15. [ 高考作文自动评分关键技术研究 [ 哈尔 55] 刘明 杨 . D]. 滨:哈尔滨工业大学硕士学位论文, 2015. [find](https://scholar.google.com/scholar?q=%5B+%E9%AB%98%E8%80%83%E4%BD%9C%E6%96%87%E8%87%AA%E5%8A%A8%E8%AF%84%E5%88%86%E5%85%B3%E9%94%AE%E6%8A%80%E6%9C%AF%E7%A0%94%E7%A9%B6+%5B+%E5%93%88%E5%B0%94+55%5D+%E5%88%98%E6%98%8E+%E6%9D%A8+%2E+D%5D%2E+%E6%BB%A8%3A%E5%93%88%E5%B0%94%E6%BB%A8%E5%B7%A5%E4%B8%9A%E5%A4%A7%E5%AD%A6%E7%A1%95%E5%A3%AB%E5%AD%A6%E4%BD%8D%E8%AE%BA%E6%96%87%2C+2015%2E)
16. [ 基于语言学特征的小学生作文流畅 56] 吴 恩 慈, 田 俊 华 . 性自动评价[ 教育测量与评价, J]. 2020( 03):41 - 50. [find](https://scholar.google.com/scholar?q=%5B+%E5%9F%BA%E4%BA%8E%E8%AF%AD%E8%A8%80%E5%AD%A6%E7%89%B9%E5%BE%81%E7%9A%84%E5%B0%8F%E5%AD%A6%E7%94%9F%E4%BD%9C%E6%96%87%E6%B5%81%E7%95%85+56%5D+%E5%90%B4+%E6%81%A9+%E6%85%88%2C+%E7%94%B0+%E4%BF%8A+%E5%8D%8E+%2E+%E6%80%A7%E8%87%AA%E5%8A%A8%E8%AF%84%E4%BB%B7%5B+%E6%95%99%E8%82%B2%E6%B5%8B%E9%87%8F%E4%B8%8E%E8%AF%84%E4%BB%B7%2C+J%5D%2E+2020%28+03%29%3A41+%2D+50%2E)
17. [ 中文作文句间逻辑合理性智 57] 刘杰, 孙 娜, 袁 克 柔, 等 . 能判 别 方 法 研 究 [ 计 算 机 应 用 与 软 件, J]. 2019, 36 ( 01):71 - 77. [find](https://scholar.google.com/scholar?q=%5B+%E4%B8%AD%E6%96%87%E4%BD%9C%E6%96%87%E5%8F%A5%E9%97%B4%E9%80%BB%E8%BE%91%E5%90%88%E7%90%86%E6%80%A7%E6%99%BA+57%5D+%E5%88%98%E6%9D%B0%2C+%E5%AD%99+%E5%A8%9C%2C+%E8%A2%81+%E5%85%8B+%E6%9F%94%2C+%E7%AD%89+%2E+%E8%83%BD%E5%88%A4+%E5%88%AB+%E6%96%B9+%E6%B3%95+%E7%A0%94+%E7%A9%B6+%5B+%E8%AE%A1+%E7%AE%97+%E6%9C%BA+%E5%BA%94+%E7%94%A8+%E4%B8%8E+%E8%BD%AF+%E4%BB%B6%2C+J%5D%2E+2019%2C+36+%28+01%29%3A71+%2D+77%2E)
18. [ 58] HAO S, XU Y, PENG H, e ta l.Au t oma t edCh i ne s e e s s ays c o r i ngf r omt op i cpe r spe c t i veus i ngr egu l a r i z ed //Pr l a t en ts eman t i ci ndex i ng[ C] o c e ed i ngso ft heI n - t e r na t i ona lCon f e r enc eonPa t t e r nRe c ogn i t i on, 2014: 3092 - 3097. [find](https://scholar.google.com/scholar?q=%5B+58%5D+HAO+S%2C+XU+Y%2C+PENG+H%2C+e+ta+l%2EAu+t+oma+t+edCh+i+ne+s+e+e+s+s+ays+c+o+r+i+ngf+r+omt+op+i+cpe+r+spe+c+t+i+veus+i+ngr+egu+l+a+r+i+z+ed+%2F%2FPr+l+a+t+en+ts+eman+t+i+ci+ndex+i+ng%5B+C%5D+o+c+e+ed+i+ngso+ft+he)
19. [ 59] XU Y,KE D,SU K.Con t ex t ua l i z edl a t en ts eman t i c i ndex i ng:A newapp r oa cht oau t oma t edCh i ne s ee s - s ays c o r i ng[ J]. J ou r na lo fI n t e l l i tSy s t ems,2017, gen 26( 2):263 - 285. [find](https://scholar.google.com/scholar?q=%5B+59%5D+XU+Y%2CKE+D%2CSU+K%2ECon+t+ex+t+ua+l+i+z+edl+a+t+en+ts+eman+t+i+c+i+ndex+i+ng%3AA+newapp+r+oa+cht+oau+t+oma+t+edCh+i+ne+s+ee+s+%2D+s+ays+c+o+r+i+ng%5B+J%5D%2E+J+ou+r+na+lo+fI+n+t+e+l+l+i+tSy+s+t+ems%2C2017%2C+gen+2)
20. [ 基于文本语义离散度的 60] 王耀 华, 李 舟 军, 何 跃 鹰, 等 . 自动 作 文 评 分 关 键 技 术 研 究 [ J]. 中 文 信 息 学 报, 2016, 30( 6):173 - 181. [find](https://scholar.google.com/scholar?q=%5B+%E5%9F%BA%E4%BA%8E%E6%96%87%E6%9C%AC%E8%AF%AD%E4%B9%89%E7%A6%BB%E6%95%A3%E5%BA%A6%E7%9A%84+60%5D+%E7%8E%8B%E8%80%80+%E5%8D%8E%2C+%E6%9D%8E+%E8%88%9F+%E5%86%9B%2C+%E4%BD%95+%E8%B7%83+%E9%B9%B0%2C+%E7%AD%89+%2E+%E8%87%AA%E5%8A%A8+%E4%BD%9C+%E6%96%87+%E8%AF%84+%E5%88%86+%E5%85%B3+%E9%94%AE+%E6%8A%80+%E6%9C%AF+%E7%A0%94+%E7%A9%B6+%5B+J%5D%2E+%E4%B8%AD+%E6%96%87+%E4%BF%A1+%E6%81%AF+%E5%AD%A6+%E6%8A%A5%2C+2016%2C+30%28+6%29%3A173+%2D+181%2E)
21. [ 嵌入语言深度感知的汉语作文评分 61] 钟 启 东, 张 景 祥 . 算法[ 计算机工程与应用, J]. 2020, 56( 08):124 - 129. [find](https://scholar.google.com/scholar?q=%5B+%E5%B5%8C%E5%85%A5%E8%AF%AD%E8%A8%80%E6%B7%B1%E5%BA%A6%E6%84%9F%E7%9F%A5%E7%9A%84%E6%B1%89%E8%AF%AD%E4%BD%9C%E6%96%87%E8%AF%84%E5%88%86+61%5D+%E9%92%9F+%E5%90%AF+%E4%B8%9C%2C+%E5%BC%A0+%E6%99%AF+%E7%A5%A5+%2E+%E7%AE%97%E6%B3%95%5B+%E8%AE%A1%E7%AE%97%E6%9C%BA%E5%B7%A5%E7%A8%8B%E4%B8%8E%E5%BA%94%E7%94%A8%2C+J%5D%2E+2020%2C+56%28+08%29%3A124+%2D+129%2E)
22. [ 少数民族汉语考试的作文辅助评 62] 蔡黎, 彭 星 源, 赵 军 . 分系统研究[ 中文信息学报, J]. 2011, 25( 05):120 - [ 63] GEORGE L,MARKJ.Me t apho r swel i veby[M ]. Un i ve r s i t fCh i c agoPr e s s,2008. yo [ 64] LIU LZ,HU X,SONG W,e ta l.Neu r a lmu l t i t a sk //Pr l e a r n i ngf o rs imi l er e c ogn i t i on[C] o c e ed i ngs o… [find](https://scholar.google.com/scholar?q=%5B+%E5%B0%91%E6%95%B0%E6%B0%91%E6%97%8F%E6%B1%89%E8%AF%AD%E8%80%83%E8%AF%95%E7%9A%84%E4%BD%9C%E6%96%87%E8%BE%85%E5%8A%A9%E8%AF%84+62%5D+%E8%94%A1%E9%BB%8E%2C+%E5%BD%AD+%E6%98%9F+%E6%BA%90%2C+%E8%B5%B5+%E5%86%9B+%2E+%E5%88%86%E7%B3%BB%E7%BB%9F%E7%A0%94%E7%A9%B6%5B+%E4%B8%AD%E6%96%87%E4%BF%A1%E6%81%AF%E5%AD%A6%E6%8A%A5%2C+J%5D%2E+2011%2C+25%28+05%29%3A120+%2D+%5B+63%5D+GEORGE+L%2CMARKJ%2EMe+t+apho+r+swel+i+veby%5BM+%5D%2E+Un+i+ve+r+s+i+t+fCh+i+c+agoPr+e+s+s%2C2008%2E+yo+%5B+64%5D+LIU+LZ%2CHU+X%2CSONG+W%2Ce+t)
23. [ 同义词词林[M]. 上海:商务印书馆,1984. 65] 梅家驹 . [find](https://scholar.google.com/scholar?q=%5B+%E5%90%8C%E4%B9%89%E8%AF%8D%E8%AF%8D%E6%9E%97%5BM%5D%2E+%E4%B8%8A%E6%B5%B7%3A%E5%95%86%E5%8A%A1%E5%8D%B0%E4%B9%A6%E9%A6%86%2C1984%2E+65%5D+%E6%A2%85%E5%AE%B6%E9%A9%B9+%2E)
24. [ 66] SONG W,LIU T,FU RJ,e ta l. Le a r n i ngt oi den t i - //Pr f en t enc epa r a l l e l i sm i ns t uden te s s ay s[ C] o - ys c e ed i ngso ft heCon f e r enc eo fI n t e r na t i ona lCon f e r enc e onCompu t a t i ona lL i ngu i s t i c s,2016:794 - 803. [find](https://scholar.google.com/scholar?q=%5B+66%5D+SONG+W%2CLIU+T%2CFU+RJ%2Ce+ta+l%2E+Le+a+r+n+i+ngt+oi+den+t+i+%2D+%2F%2FPr+f+en+t+enc+epa+r+a+l+l+e+l+i+sm+i+ns+t+uden+te+s+s+ay+s%5B+C%5D+o+%2D+ys+c+e+ed+i+ngso+ft+heCon+f+e+r+enc+eo+fI+n+t+e+r+na+t+i+ona+lCon+f+e+)
25. [ 67] 穆婉青,廖 健,王 素 格 . 融 合 CNN 和 结 构 相 似 度 计 算的排比句 识 别 及 应 用 [ 中 文 信 息 学 报, J]. 2018, 32 ( 02):139 - 146. [find](https://scholar.google.com/scholar?q=%5B+67%5D+%E7%A9%86%E5%A9%89%E9%9D%92%2C%E5%BB%96+%E5%81%A5%2C%E7%8E%8B+%E7%B4%A0+%E6%A0%BC+%2E+%E8%9E%8D+%E5%90%88+CNN+%E5%92%8C+%E7%BB%93+%E6%9E%84+%E7%9B%B8+%E4%BC%BC+%E5%BA%A6+%E8%AE%A1+%E7%AE%97%E7%9A%84%E6%8E%92%E6%AF%94%E5%8F%A5+%E8%AF%86+%E5%88%AB+%E5%8F%8A+%E5%BA%94+%E7%94%A8+%5B+%E4%B8%AD+%E6%96%87+%E4%BF%A1+%E6%81%AF+%E5%AD%A6+%E6%8A%A5%2C+J%5D%2E+2018%2C+32+%28+02%29%3A139+%2D+146%2E)
26. [ 基 于 BERT 模 型 的 排 比 句 自 动 识 别 68] 朱晓亮, 谯 宇 同 . 方法[ 计算机应用与软件, J]. 2021, 38( 07):153 - [ 汉语 书 面 语 体 庄 雅 度 的 自 动 测 量 69] 冯胜利, 王洁, 黄梅 . [find](https://scholar.google.com/scholar?q=%5B+%E5%9F%BA+%E4%BA%8E+BERT+%E6%A8%A1+%E5%9E%8B+%E7%9A%84+%E6%8E%92+%E6%AF%94+%E5%8F%A5+%E8%87%AA+%E5%8A%A8+%E8%AF%86+%E5%88%AB+68%5D+%E6%9C%B1%E6%99%93%E4%BA%AE%2C+%E8%B0%AF+%E5%AE%87+%E5%90%8C+%2E+%E6%96%B9%E6%B3%95%5B+%E8%AE%A1%E7%AE%97%E6%9C%BA%E5%BA%94%E7%94%A8%E4%B8%8E%E8%BD%AF%E4%BB%B6%2C+J%5D%2E+2021%2C+38%28+07%29%3A153+%2D+%5B+%E6%B1%89%E8%AF%AD+%E4%B9%A6+%E9%9D%A2+%E8%AF%AD+%E4%BD%93+%E5%BA%84+%E9%9B%85+%E5%BA%A6+%E7%9A%84+%E8%87%AA+%E5%8A%A8+%E6%B5%8B+%E9%87%8F+69%5D+%E5%86%AF%E8%83%9C%E5%88%A9%2C+%E7%8E%8B%E6%B4%81%2C+%E9%BB%84%E6%A2%85+%2E)