TY - GEN
T1 - The application of active learning in identification of students with financial difficulties
AU - Wang, Qiqi
AU - Huang, Ding
AU - Shen, Yongchao
AU - Zhang, Yuxi
N1 - Publisher Copyright:
© 2017 Association for Computing Machinery.
PY - 2017/12/20
Y1 - 2017/12/20
N2 - The previous classifiers tend to achieve unsatisfied performance with class-imbalanced data. In order to identify the poverty students using data with few labels, we adopted the active learning method. The results show that it can get lower error compared with the random sampling strategy, which means that the strategy applied in our problem is effective. With the strategy, we can classify the poverty students with much more accuracy, which is useful in distributing subsidies to students in college.
AB - The previous classifiers tend to achieve unsatisfied performance with class-imbalanced data. In order to identify the poverty students using data with few labels, we adopted the active learning method. The results show that it can get lower error compared with the random sampling strategy, which means that the strategy applied in our problem is effective. With the strategy, we can classify the poverty students with much more accuracy, which is useful in distributing subsidies to students in college.
KW - Active Learning
KW - Imbalanced classification
KW - Poverty student identification
UR - https://www.scopus.com/pages/publications/85044251690
U2 - 10.1145/3175536.3175560
DO - 10.1145/3175536.3175560
M3 - 会议稿件
AN - SCOPUS:85044251690
T3 - ACM International Conference Proceeding Series
SP - 136
EP - 140
BT - Proceedings of the 9th International Conference on Education Technology and Computers, ICETC 2017
PB - Association for Computing Machinery
T2 - 9th International Conference on Education Technology and Computers, ICETC 2017
Y2 - 20 December 2017 through 22 December 2017
ER -