The Influence of Variance in Learner Answers on Automatic Content Scoring

被引：26
作者：
Horbach, Andrea [1 ]
Zesch, Torsten [1 ]
机构：
[1] Univ Duisburg Essen, Language Technol Lab, Duisburg, Germany
来源：
FRONTIERS IN EDUCATION | 2019年 / 4卷
关键词：
automatic content scoring; short-answer questions; natural language processing; linguistic variance; machine learning;
D O I：
10.3389/feduc.2019.00028
中图分类号：
G40 [教育学];
学科分类号：
040101 ; 120403 ;
摘要：
Automatic content scoring is an important application in the area of automatic educational assessment. Short texts written by learners are scored based on their content while spelling and grammar mistakes are usually ignored. The difficulty of automatically scoring such texts varies according to the variance within the learner answers. In this paper, we first discuss factors that influence variance in learner answers, so that practitioners can better estimate if automatic scoring might be applicable to their usage scenario. We then compare the two main paradigms in content scoring: (i) similarity-based and (ii) instance-based methods, and discuss how well they can deal with each of the variance-inducing factors described before.
引用
收藏
页数：14
相关论文
共 47 条
  • [1] Anderson Lorin W, 2001, A Taxonomy for Learning, Teaching, and Assessing: A Revision of Blooms Taxonomy of Educational Objectives
  • [2] Andresen M., 2017, P 4 INT C DEPENDENCY, P4
  • [3] [Anonymous], 2005, CONT LINGUISTICS
  • [4] [Anonymous], 1998, WORDNET ELECT LEXICA, DOI DOI 10.7551/MITPRESS/7287.001.0001
  • [5] Bailey S., 2008, P 3 ACL WORKSHOP INN, P107
  • [6] Bar D., 2012, P 6 INT WORKSH SEM E, P435
  • [7] Basu S., 2013, T ASS COMPUTATIONAL, V1, P391, DOI [DOI 10.1162/TACL_A_00236, DOI 10.1162/TACLA00236]
  • [8] Beinborn L., 2016, P 11 WORKSH INN US N, P73, DOI [10.18653/v1/W16-0508, DOI 10.18653/V1/W16-0508]
  • [9] Bentivogli L., 2013, SEM 2013 1 JOINT C L
  • [10] What Is a Paraphrase?
    Bhagat, Rahul
    Hovy, Eduard
    [J]. COMPUTATIONAL LINGUISTICS, 2013, 39 (03) : 463 - 472
12345