Automated Prediction of Student Participation in Collaborative Dialogs Using Time Series Analyses

Iulia Pasov; Mihai Dascalu; Nicolae Nistor; Stefan Trausan-Matu

Automated Prediction of Student Participation in Collaborative Dialogs Using Time Series Analyses

Iulia Pasov, Mihai Dascalu, Nicolae Nistor, Stefan Trausan-Matu

Research output: Chapter in Book/Report/Conference proceeding › Conference Article in proceeding › Academic › peer-review

Abstract

The massive student participation in Computer Supported Collaborative Learning (CSCL) sessions from online classrooms requires intense tutor engagement to track and evaluate individual student participation. In this study, we investigate how the time evolution of messages predicts students’ participation using two models – a linear regression and a Random Forest model. A corpus of 10 chats involving 47 students was scored by 4 human experts and used to evaluate our models. Our analysis shows that students’ pauses length between consecutive messages within a discussion is the strongest participation predictor accounting for R2 ¼ :796 variance in the human estimations while using a Random Forest model. Our results provide an extended basis for the automated assessment of student participation in collaborative online discussions.

Original language	English
Title of host publication	The Interplay of Data, Technology, Place and People for Smart Learning
Subtitle of host publication	Proceedings of the 3rd International Conference on Smart Learning Ecosystems and Regional Development
Editors	Hendrik Knoche, Elvira Popescu, Antonio Cartelli
Publisher	Springer
Pages	177-185
Number of pages	9
Publication status	Published - 1 Jun 2018
Externally published	Yes
Event	3rd Int. Conf. on Smart Learning Ecosystems and Regional Development (SLERD 2018): The Interplay of Data, Technology, Place and People for Smart Learning - Aalborg, Denmark Duration: 23 May 2018 → 25 May 2018

Publication series

Series	Smart Innovation, Systems and Technologies (SIST)
Volume	95

Conference

Conference	3rd Int. Conf. on Smart Learning Ecosystems and Regional Development (SLERD 2018)
Country/Territory	Denmark
City	Aalborg
Period	23/05/18 → 25/05/18

Keywords

CSCL
Time series analysis
Automated evaluation of participation
RAGE

Rage: Realising an Applied Gaming Eco-system
Westera, W. (PI), Georgiadis, K. (CoI), Saveski, G. (CoI), van Lankveld, G. (CoI), Bahreini, K. (CoI), van der Vegt, W. (CoI), Berkhout, J. (CoI), Nyamsuren, E. (CoI), Kluijfhout, E. (CoI) & Nadolski, R. (CoI)
European Union
1/02/15 → 31/07/19
Project: Research

Cite this

Pasov, I., Dascalu, M., Nistor, N., & Trausan-Matu, S. (2018). Automated Prediction of Student Participation in Collaborative Dialogs Using Time Series Analyses. In H. Knoche, E. Popescu, & A. Cartelli (Eds.), The Interplay of Data, Technology, Place and People for Smart Learning: Proceedings of the 3rd International Conference on Smart Learning Ecosystems and Regional Development (pp. 177-185). Springer.

Pasov, Iulia ; Dascalu, Mihai ; Nistor, Nicolae et al. / Automated Prediction of Student Participation in Collaborative Dialogs Using Time Series Analyses. The Interplay of Data, Technology, Place and People for Smart Learning: Proceedings of the 3rd International Conference on Smart Learning Ecosystems and Regional Development. editor / Hendrik Knoche ; Elvira Popescu ; Antonio Cartelli. Springer, 2018. pp. 177-185 (Smart Innovation, Systems and Technologies (SIST), Vol. 95).

@inproceedings{66647daf9bf747be8294961e8704a83b,

title = "Automated Prediction of Student Participation in Collaborative Dialogs Using Time Series Analyses",

abstract = "The massive student participation in Computer Supported Collaborative Learning (CSCL) sessions from online classrooms requires intense tutor engagement to track and evaluate individual student participation. In this study, we investigate how the time evolution of messages predicts students{\textquoteright} participation using two models – a linear regression and a Random Forest model. A corpus of 10 chats involving 47 students was scored by 4 human experts and used to evaluate our models. Our analysis shows that students{\textquoteright} pauses length between consecutive messages within a discussion is the strongest participation predictor accounting for R2 ¼ :796 variance in the human estimations while using a Random Forest model. Our results provide an extended basis for the automated assessment of student participation in collaborative online discussions.",

keywords = "CSCL, Time series analysis, Automated evaluation of participation, RAGE",

author = "Iulia Pasov and Mihai Dascalu and Nicolae Nistor and Stefan Trausan-Matu",

year = "2018",

month = jun,

day = "1",

language = "English",

series = "Smart Innovation, Systems and Technologies (SIST)",

publisher = "Springer",

pages = "177--185",

editor = "Knoche, {Hendrik } and Popescu, {Elvira } and Cartelli, {Antonio }",

booktitle = "The Interplay of Data, Technology, Place and People for Smart Learning",

address = "United States",

note = "3rd Int. Conf. on Smart Learning Ecosystems and Regional Development (SLERD 2018) : The Interplay of Data, Technology, Place and People for Smart Learning ; Conference date: 23-05-2018 Through 25-05-2018",

}

Pasov, I, Dascalu, M, Nistor, N & Trausan-Matu, S 2018, Automated Prediction of Student Participation in Collaborative Dialogs Using Time Series Analyses. in H Knoche, E Popescu & A Cartelli (eds), The Interplay of Data, Technology, Place and People for Smart Learning: Proceedings of the 3rd International Conference on Smart Learning Ecosystems and Regional Development. Springer, Smart Innovation, Systems and Technologies (SIST), vol. 95, pp. 177-185, 3rd Int. Conf. on Smart Learning Ecosystems and Regional Development (SLERD 2018), Aalborg, Denmark, 23/05/18.

Automated Prediction of Student Participation in Collaborative Dialogs Using Time Series Analyses. / Pasov, Iulia; Dascalu, Mihai; Nistor, Nicolae et al.
The Interplay of Data, Technology, Place and People for Smart Learning: Proceedings of the 3rd International Conference on Smart Learning Ecosystems and Regional Development. ed. / Hendrik Knoche; Elvira Popescu; Antonio Cartelli. Springer, 2018. p. 177-185 (Smart Innovation, Systems and Technologies (SIST), Vol. 95).

Research output: Chapter in Book/Report/Conference proceeding › Conference Article in proceeding › Academic › peer-review

TY - GEN

T1 - Automated Prediction of Student Participation in Collaborative Dialogs Using Time Series Analyses

AU - Pasov, Iulia

AU - Dascalu, Mihai

AU - Nistor, Nicolae

AU - Trausan-Matu, Stefan

PY - 2018/6/1

Y1 - 2018/6/1

N2 - The massive student participation in Computer Supported Collaborative Learning (CSCL) sessions from online classrooms requires intense tutor engagement to track and evaluate individual student participation. In this study, we investigate how the time evolution of messages predicts students’ participation using two models – a linear regression and a Random Forest model. A corpus of 10 chats involving 47 students was scored by 4 human experts and used to evaluate our models. Our analysis shows that students’ pauses length between consecutive messages within a discussion is the strongest participation predictor accounting for R2 ¼ :796 variance in the human estimations while using a Random Forest model. Our results provide an extended basis for the automated assessment of student participation in collaborative online discussions.

AB - The massive student participation in Computer Supported Collaborative Learning (CSCL) sessions from online classrooms requires intense tutor engagement to track and evaluate individual student participation. In this study, we investigate how the time evolution of messages predicts students’ participation using two models – a linear regression and a Random Forest model. A corpus of 10 chats involving 47 students was scored by 4 human experts and used to evaluate our models. Our analysis shows that students’ pauses length between consecutive messages within a discussion is the strongest participation predictor accounting for R2 ¼ :796 variance in the human estimations while using a Random Forest model. Our results provide an extended basis for the automated assessment of student participation in collaborative online discussions.

KW - CSCL

KW - Time series analysis

KW - Automated evaluation of participation

KW - RAGE

M3 - Conference Article in proceeding

T3 - Smart Innovation, Systems and Technologies (SIST)

SP - 177

EP - 185

BT - The Interplay of Data, Technology, Place and People for Smart Learning

A2 - Knoche, Hendrik

A2 - Popescu, Elvira

A2 - Cartelli, Antonio

PB - Springer

T2 - 3rd Int. Conf. on Smart Learning Ecosystems and Regional Development (SLERD 2018)

Y2 - 23 May 2018 through 25 May 2018

ER -

Pasov I, Dascalu M, Nistor N, Trausan-Matu S. Automated Prediction of Student Participation in Collaborative Dialogs Using Time Series Analyses. In Knoche H, Popescu E, Cartelli A, editors, The Interplay of Data, Technology, Place and People for Smart Learning: Proceedings of the 3rd International Conference on Smart Learning Ecosystems and Regional Development. Springer. 2018. p. 177-185. (Smart Innovation, Systems and Technologies (SIST), Vol. 95).

Automated Prediction of Student Participation in Collaborative Dialogs Using Time Series Analyses

Abstract

Publication series

Conference

Keywords

Fingerprint

Projects

Rage: Realising an Applied Gaming Eco-system

Cite this