英语自动作文评分系统实现路径探析

Probe into the implementation path of English automatic composition scoring system

英语自动作文评分系统不仅能节约大量人力、物力和财力,还对提高英语学习者英语写作水平起到非常重要的作用。本文系统分析了当下四种实现自动作文评分系统的路径,它们分别为基于非文本相关特征、基于文本相关特征、基于概率统计分类方法和基于深度神经网络的自动作文评分系统。依据分析结果,提出未来英语自动作文评分系统的主要发展方向是融合四种实现路径的优点、优化作文数据特征维度和提升作文练习的有效反馈能力。

The automatic English composition scoring system not only saves a lot of manpower, material and financial resources, but also plays an important role in improving the English writing level of English learners. This paper systematically analyzes the current four paths for implementing automatic composition scoring system. They are based on non-text related features, text-based features, probabilistic statistical classification methods and automatic composition scoring system based on deep neural network. According to the analysis results, the main development direction of the future automatic English composition scoring system is to combine the advantages of the four realization paths, optimize the feature dimension of the composition data and improve the effective feedback ability of the composition exercise.