|
|||||
|
|
| 面向作文自动评分的优美句识别 | |
| 引用本文： | 付瑞吉,王栋,王士进,胡国平,刘挺.面向作文自动评分的优美句识别[J].中文信息学报,2018,32(6):88-97. |
| 作者姓名： | 付瑞吉 王栋 王士进 胡国平 刘挺 |
| 作者单位： | 1.科大讯飞股份有限公司研究院,安徽 合肥 518057; 2.科大讯飞股份有限公司 哈工大讯飞联合实验室,北京 100094; 3.哈尔滨工业大学 计算机科学与技术学院 社会计算与信息检索研究中心,黑龙江 哈尔滨 150001 |
| 基金项目： | 国家863计划课题(2015AA015409) |
| 摘 要： | 语言优美是学生写作能力中重要的一部分。该文提出一个面向作文自动评分的作文优美句识别任务,主要识别中学生中文作文中的优美句。相比传统文本分类任务,优美句识别更加难以用特征工程的方式解决。因此,该文提出一种基于卷积神经网络(CNN)和双向长短时记忆(BiLSTM)网络的混合神经网络结构进行优美句识别,并和CNN、BiLSTM网络进行了对比。实验证明,混合神经网络的准确率最高,达到89.23%,F1值与BiLSTM相当,达到75.39%。此外,该文将优美句子特征用于作文自动评分任务,可使计算机评分和人工评分的大分差比例下降21.41%。 |
| 关 键 词： | 优美句识别 深度神经网络 作文自动评分 |
Elegart Sentence Recognition for Automated Essay Scoring |
|
| FU Ruiji,WANG Dong,WANG Shijin,HU Guoping,LIU Ting.Elegart Sentence Recognition for Automated Essay Scoring[J].Journal of Chinese Information Processing,2018,32(6):88-97. | |
| Authors: | FU Ruiji WANG Dong WANG Shijin HU Guoping LIU Ting |
| Affiliation: | 1.iFLYTEK Research, Hefei, Anhui 518057, China; 2.Joint Laboratory of HIT and iFLYTEK, Beijing 100094, China; 3. Research Center for Social Computing and Information Retrieval, School of Computer Science and Technology, Harbin Institute of Technology, Harbin, Heilongjiang 150001, China |
| Abstract: | This paper proposs the task of elegant sentence recognition in Chinese essays of high school students for Automated Essay Scoring (AES). To deal withthis task clellenging the classical text classification plus feature engineering,this paper presents a deep neural network combining Convolutional Neural Network (CNN) and Bi-directional Long Short-Term Memory (BiLSTM) networks to recognize grace sentences. The experiment results show that our joint neural network ranks to in precision (89.23%),with a comparable F1 score to BiLSTM (75.39%). We finally apply the graceful sentence features to the AES task,which can reduce the large-margin prediction error by 21.41%. |
| Keywords: | graceful sentence recognition deep neural networks automated essay scoring |
| 点击此处可从《中文信息学报》浏览原始摘要信息 | |
| 点击此处可从《中文信息学报》下载全文 | |