Concept and Realization to Automatically Generate Test and Training Data for a Natural Language Processing Algorithm for the Assessment of Free Text Answers for Digital Courses on Work Design in the Context of Psychology Studies in Higher Educational Institutions
Ramona SrbeckyUniversity of Hagen,Multimedia and Internet-Applications,Hagen,GermanyFranz BühlerUniversity of Hagen,Multimedia and Internet-Applications,Hagen,GermanySimon-Alexander WetzelUniversity of Hagen,Multimedia and Internet-Applications,Hagen,GermanyMichael WinterhagenUniversity of Hagen,Multimedia and Internet-Applications,Hagen,GermanyWieland FraasUniversity of Hagen,Work and Organizational Psychology,Hagen,GermanyJan DettmersUniversity of Hagen,Work and Organizational Psychology,Hagen,GermanyMatthias HemmjeUniversity of Hagen,Multimedia and Internet-Applications,Hagen,Germany
2022en
ABI
Annotatsiya
In this full paper for the CSCI-RTED research track, we present an approach on how to generate test and training data for a Natural Language Processing (NLP) algorithm to asses free text answers for a digital university course on the topic of work organization and work design in an automated approach without actual data from the learners' as test and training data. For this purpose, this paper first presents the research methodology, questions, and relevant theoretical background. Based on this, the concept and the implementation will be presented. In the end, the evaluation and future improvements are presented.
Hali tarjima qilinmagan
Identifikatorlar
Iqtiboslar va manbalar
2 ta iqtibos0 ta foydalanilgan manba