Optimizing ranking for response prediction via triplet-wise learning from historical feedback.
Lili ShanLei LinChengjie SunXiaolong WangBingquan LiuPublished in: Int. J. Mach. Learn. Cybern. (2017)
Keyphrases
- learning algorithm
- learning process
- knowledge acquisition
- supervised learning
- online learning
- reinforcement learning
- prior knowledge
- learning tasks
- learning systems
- active learning
- learning problems
- prediction accuracy
- assessment tool
- elementary school
- ranking algorithm
- unsupervised learning
- relevance feedback
- semi supervised
- pairwise
- training set
- decision trees