层次分析法下的语法论文检测评分系统研究  

Design of automatic grading system for English papers based on Analytic hierarchy process

作　　者：苏涛涛[1] SU Taotao(Shangluo University,Shangluo Shanxi 726000,China)

机构地区：[1]商洛学院

出　　处：《自动化与仪器仪表》2019年第7期222-225,共4页Automation & Instrumentation

基　　金：陕西省教育厅专项科学研究计划:生态语言学视阈下陕西文学作品话语权构建研究(No.17JK0234)

摘　　要：为了提高英语论文在复杂语义环境下自动评分的准确性,提出一种基于语义层次分析和自适应上下文连贯性分析的英语论文自动评分算法,并结合软件设计方法进行英语论文自动评分系统设计。采用模糊关联映射方法构造英语论文自动评分的语义概念树,考虑英语论文的上下文语义概念间的包含关系和语义等价关系进行语义层次分析,构造基于语义本体层次分析的自动评分模糊决策函数,基于语义相似度进行论文评分规则算法构造,将连接词和短语等实例集纳入模糊论文评分的约束因素,采用准确的自动评分系统进行英文论文中的词语、语句、创新性以及结构性的联合判别,实现英语论文的复杂语义机器自动评分。在IIC总线环境和嵌入式环境下实现英语论文自动评分系统的软件开发设计。系统测试结果表明,采用该系统进行英语论文评分的自动性较好,准确评价度较高。In order to improve the accuracy of automatic scoring of English papers in complex semantic environments,an automatic scoring algorithm based on semantic hierarchy analysis and adaptive context coherence analysis is proposed.Combined with the software design method,the automatic scoring system of English papers is designed.A semantic concept tree for automatic scoring of English papers is constructed by using fuzzy association mapping method,and semantic hierarchy analysis is carried out considering the inclusion relation and semantic equivalence relationship between contextual semantic concepts of English papers.An automatic scoring fuzzy decision function based on semantic ontology hierarchy analysis is constructed.Based on semantic similarity,the paper scoring rules algorithm is constructed,and the connectors and phrases are brought into the constraint factors of fuzzy paper scoring.The preposition is incorporated into the combined form to realize the automatic scoring of English papers with complex semantic machines.In the IIC bus environment and embedded environment,the software development and design of English thesis automatic scoring system is realized.The test results show that the system is more automatic and accurate.

关 键 词：层次分析 英语论文 自动评分系统 IIC总线 

分 类 号：TP319[自动化与计算机技术—计算机软件与理论]