Abstract
This paper proposes a simple estimation of the quality of student oral presentations. It is based on the study and analysis of features extracted from the audio and digital slides of 448 presentations. The main goal of this work is to automatically predict the values assigned by professors to different criteria in a presentation evaluation rubric. Ma- chine Learning methods were used to create several models that classify students in two clusters: high and low perform- ers. The models created from slide features were accurate up to 65%. The most relevant features for the slide-base models were: number of words, images, and tables, and the maximum font size. The audio-based models reached up to 69% of accuracy, with pitch and filled pauses related features being the most significant. The relatively high degrees of ac- curacy obtained with these very simple features encourage the development of automatic estimation tools for improving presentation skills.
| Original language | English (US) |
|---|---|
| Title of host publication | MLA 2014 - Proceedings of the 2014 ACM Multimodal Learning Analytics Workshop and Grand Challenge, Co-located with ICMI 2014 |
| Publisher | Association for Computing Machinery, Inc |
| Pages | 37-44 |
| Number of pages | 8 |
| ISBN (Electronic) | 9781450304887 |
| DOIs | |
| State | Published - Jan 1 2014 |
| Event | 3rd Multimodal Learning Analytics Workshop and Grand Challenges, MLA 2014 - Istanbul, Turkey Duration: Nov 12 2014 → Nov 12 2014 |
Other
| Other | 3rd Multimodal Learning Analytics Workshop and Grand Challenges, MLA 2014 |
|---|---|
| Country | Turkey |
| City | Istanbul |
| Period | 11/12/14 → 11/12/14 |
Fingerprint
Keywords
- Audio features
- Multimodal Learning Analytics
- Presentation skills
- Slides features
ASJC Scopus subject areas
- Computer Science Applications
- Education
Cite this
Estimation of presentations skills based on slides and audio features. / Luzardo, Gonzalo; Guamán, Bruno; Chiluiza, Katherine; Castells, Jaime; Ochoa, Xavier.
MLA 2014 - Proceedings of the 2014 ACM Multimodal Learning Analytics Workshop and Grand Challenge, Co-located with ICMI 2014. Association for Computing Machinery, Inc, 2014. p. 37-44.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Estimation of presentations skills based on slides and audio features
AU - Luzardo, Gonzalo
AU - Guamán, Bruno
AU - Chiluiza, Katherine
AU - Castells, Jaime
AU - Ochoa, Xavier
PY - 2014/1/1
Y1 - 2014/1/1
N2 - This paper proposes a simple estimation of the quality of student oral presentations. It is based on the study and analysis of features extracted from the audio and digital slides of 448 presentations. The main goal of this work is to automatically predict the values assigned by professors to different criteria in a presentation evaluation rubric. Ma- chine Learning methods were used to create several models that classify students in two clusters: high and low perform- ers. The models created from slide features were accurate up to 65%. The most relevant features for the slide-base models were: number of words, images, and tables, and the maximum font size. The audio-based models reached up to 69% of accuracy, with pitch and filled pauses related features being the most significant. The relatively high degrees of ac- curacy obtained with these very simple features encourage the development of automatic estimation tools for improving presentation skills.
AB - This paper proposes a simple estimation of the quality of student oral presentations. It is based on the study and analysis of features extracted from the audio and digital slides of 448 presentations. The main goal of this work is to automatically predict the values assigned by professors to different criteria in a presentation evaluation rubric. Ma- chine Learning methods were used to create several models that classify students in two clusters: high and low perform- ers. The models created from slide features were accurate up to 65%. The most relevant features for the slide-base models were: number of words, images, and tables, and the maximum font size. The audio-based models reached up to 69% of accuracy, with pitch and filled pauses related features being the most significant. The relatively high degrees of ac- curacy obtained with these very simple features encourage the development of automatic estimation tools for improving presentation skills.
KW - Audio features
KW - Multimodal Learning Analytics
KW - Presentation skills
KW - Slides features
UR - http://www.scopus.com/inward/record.url?scp=84919394304&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84919394304&partnerID=8YFLogxK
U2 - 10.1145/2666633.2666639
DO - 10.1145/2666633.2666639
M3 - Conference contribution
SP - 37
EP - 44
BT - MLA 2014 - Proceedings of the 2014 ACM Multimodal Learning Analytics Workshop and Grand Challenge, Co-located with ICMI 2014
PB - Association for Computing Machinery, Inc
ER -