TY - GEN
T1 - Sentiment analysis techniques and applications in education
T2 - 1st International Conference on Technology and Innovation in Learning, Teaching and Education, TECH-EDU 2018
AU - Dolianiti, Foteini S.
AU - Iakovakis, Dimitrios
AU - Dias, Sofia B.
AU - Hadjileontiadou, Sofia
AU - Diniz, José A.
AU - Hadjileontiadis, Leontios
N1 - Publisher Copyright:
© Springer Nature Switzerland AG 2019.
PY - 2019
Y1 - 2019
N2 - As the interplay between cognition and emotion is involved in every learning process, student profile should be enhanced with information regarding his/her affective state. Sentiment analysis could serve this end, through the analysis of student behavioral traces in teaching-learning environments. The purpose of the present study is to review the status of research on the field of sentiment analysis in the educational domain; exploring different ways in which sentiment analysis has been applied in the educational domain, and analyze the different techniques that researchers have adopted in developing sentiment analysis systems on educational datasets. Five different task types that sentiment analysis has served within the domain were identified, namely: (i) instruction evaluation, (ii) institutional decision/policy making, (iii) intelligent information/learning systems enhancement, (iv) assignment evaluation and feedback improvement, and (v) new research insights. From a technical perspective, a brief explanation of the different sentiment analysis techniques along with representative examples are presented. The character of this work may address the needs of a diverse group of stakeholders, including educators, social sciences researchers as well as researchers, in natural language processing in education.
AB - As the interplay between cognition and emotion is involved in every learning process, student profile should be enhanced with information regarding his/her affective state. Sentiment analysis could serve this end, through the analysis of student behavioral traces in teaching-learning environments. The purpose of the present study is to review the status of research on the field of sentiment analysis in the educational domain; exploring different ways in which sentiment analysis has been applied in the educational domain, and analyze the different techniques that researchers have adopted in developing sentiment analysis systems on educational datasets. Five different task types that sentiment analysis has served within the domain were identified, namely: (i) instruction evaluation, (ii) institutional decision/policy making, (iii) intelligent information/learning systems enhancement, (iv) assignment evaluation and feedback improvement, and (v) new research insights. From a technical perspective, a brief explanation of the different sentiment analysis techniques along with representative examples are presented. The character of this work may address the needs of a diverse group of stakeholders, including educators, social sciences researchers as well as researchers, in natural language processing in education.
KW - Education
KW - Machine learning
KW - Natural language processing
KW - Opinion mining
KW - Sentiment analysis
UR - http://www.scopus.com/inward/record.url?scp=85067243647&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-20954-4_31
DO - 10.1007/978-3-030-20954-4_31
M3 - Conference contribution
AN - SCOPUS:85067243647
SN - 9783030209537
T3 - Communications in Computer and Information Science
SP - 412
EP - 427
BT - Technology and Innovation in Learning, Teaching and Education - 1st International Conference, TECH-EDU 2018, Revised Selected Papers
A2 - Mikropoulos, Tassos A.
A2 - Diniz, José A.
A2 - Tsitouridou, Meni
PB - Springer Verlag
Y2 - 20 June 2018 through 22 June 2018
ER -
