作者: Sharon Oviatt , Adrienne Cohen
关键词:
摘要: One aim of multimodal learning analytics is to analyze rich natural communication modalities identify domain expertise and rapidly reliably. In this research, written representations are analyzed from the Math Data Corpus, which involves data (digital pen, speech, images) on collaborating students as they solve math problems. Findings reveal that in 96-97% cases correctness a group's solution was predictable advance based students' work content. addition, linear regression revealed 65% variance individual rankings could be accounted for their A content analysis both spoken input correctly predicted dominant expert group 100% time, exceeding unimodal prediction rates. Further reversal between experts non-experts percentage time match versus mismatch present oral answer contributions, with demonstrating higher mismatches. Implications discussed developing reliable systems incorporate digital pen automatically track consolidation expertise.