Trialog in ARIES : User Input Assessment in an Intelligent Tutoring System
Trialog in ARIES : User Input Assessment in an Intelligent Tutoring System
复制标题
ARIES 中的 Trialog:智能辅导系统中的用户输入评估
DOI:
--
复制
发表时间:
2011
期刊:
影响因子:
--
通讯作者:
Patricia S. Wallace
中科院分区:
文献类型:
--
作者:
Zhiqiang Cai;A. Graesser;Carol M. Forsyth;Candice Burkett;Keith K. Millis;Patricia S. Wallace
OperationARIES! (or ARIES for short) is an intelligent tutoring system that teaches critical thinking and helps learners acquire scientific inquiry skills. One of the core components of ARIES is “trialogs” which are three-party conversations in natural language among a human student and two artificial pedagogical agents (tutor and fellow student). These tutorial interactions were designed to enhance the students’ learning experience. Assessing student input is essential to the performance of ARIES. Regular expressions and Latent Semantic Analysis (LSA) were used in models that evaluate students’ answers. The resulting computational models were found to be as reliable as human raters. Keywords-Operation ARIES!; AutoTutor; intelligent tutoring systems; LSA; natural language processing