Combination of feature engineering and ranking models for paper-author identification in KDD Cup 2013.
Journal
Proceedings of the 2013 KDD Cup 2013 Workshop, KDD Cup 2013, Chicago, Illinois, USA, August 11-14, 2013
Pages
2:1-2:7
Date Issued
2013
Author(s)
Li, Chun-Liang et al.
Su, Yu-Chuan
Lin, Ting-Wei
Tsai, Cheng-Hao
Chang, Wei-Cheng
Huang, Kuan-Hao
Kuo, Tzu-Ming
Lin, Shan-Wei
Lin, Young-San
Lu, Yu-Chen
Yang, Chun-Pai
Chang, Cheng-Xia
Chin, Wei-Sheng
Juan, Yu-Chin
Tung, Hsiao-Yu
Wang, Jui-Pin
Wei, Cheng-Kuang
Wu, Felix
Yin, Tu-Chun
Yu, Tong
Zhuang, Yong
Lin, Hsuan-Tien
Abstract
The track 1 problem in KDD Cup 2013 is to discriminate between papers confirmed by the given authors from the other deleted papers. This paper describes the winning solution of team National Taiwan University for track 1 of KDD Cup 2013. First, we conduct the feature engineering to transform the various provided text information into 97 features. Second, we train classification and ranking models using these features. Last, we combine our individual models to boost the performance by using results on the internal validation set and the official Valid set. Some effective post-processing techniques have also been proposed. Our solution achieves 0.98259 MAP score and ranks the first place on the private leaderboard of Test set.
SDGs
Type
conference paper