Say What? Automatic Modeling of Collaborative Problem Solving Skills from Student Speech in the Wild [Konferenzbericht] Paper presented at the International Conference on Educational Data Mining (EDM) (14th, Online, Jun 29-Jul 2, 2021).

Autor/inn/en	Pugh, Samuel L.; Subburaj, Shree Krishna; Rao, Arjun Ramesh; Stewart, Angela E. B.; Andrews-Todd, Jessica; D'Mello, Sidney K.
Titel	Say What? Automatic Modeling of Collaborative Problem Solving Skills from Student Speech in the Wild [Konferenzbericht] Paper presented at the International Conference on Educational Data Mining (EDM) (14th, Online, Jun 29-Jul 2, 2021).
Quelle	(2021), (13 Seiten) PDF als Volltext kostenfreie Datei Verfügbarkeit
Zusatzinformation	Weitere Informationen
Sprache	englisch
Dokumenttyp	gedruckt; online; Monographie
Schlagwörter	Problem Solving; Cooperation; Middle School Students; High School Students; Automation; Pattern Recognition; Speech; Natural Language Processing; Videoconferencing; Mathematics; Physics; Artificial Intelligence; Computer Uses in Education; Accuracy; Classification + Suchen Sie Ihr Suchwort? Problemlösen; Co-operation; Kooperation; Middle school; Middle schools; Student; Students; Mittelschule; Mittelstufenschule; Schüler; Schülerin; High school; High schools; Oberschule; Studentin; Mustererkennung; Speaking; Sprechen; Natürliche Sprache; Mathematik; Physik; Künstliche Intelligenz; Computernutzung; Classification system; Klassifikation; Klassifikationssystem
Abstract	We investigated the feasibility of using automatic speech recognition (ASR) and natural language processing (NLP) to classify collaborative problem solving (CPS) skills from recorded speech in noisy environments. We analyzed data from 44 dyads of middle and high school students who used videoconferencing to collaboratively solve physics and math problems (35 and 9 dyads in school and lab environments, respectively). Trained coders identified seven cognitive and social CPS skills (e.g., sharing information) in 8,660 utterances. We used a state-of-the-art deep transfer learning approach for NLP, Bidirectional Encoder Representations from Transformers (BERT), with a special input representation enabling the model to analyze adjacent utterances for contextual cues. We achieved a micro-average AUROC score (across seven CPS skills) of 0.80 using ASR transcripts, compared to 0.91 for human transcripts, indicating a decrease in performance attributable to ASR error. We found that the noisy school setting introduced additional ASR error, which reduced model performance (micro-average AUROC of 0.78) compared to the lab (AUROC = 0.83). We discuss implications for real-time CPS assessment and support in schools. [For the full proceedings, see ED615472.] (As Provided).
Anmerkungen	International Educational Data Mining Society. e-mail: admin@educationaldatamining.org; Web site: https://educationaldatamining.org/conferences/
Erfasst von	ERIC (Education Resources Information Center), Washington, DC
Update	2024/1/01