COLTR: Semi-Supervised Learning to Rank With Co-Training and Over-Parameterization for Web Search.
Yuchen LiHaoyi XiongQingzhong WangLinghe KongHao LiuHaifang LiJiang BianShuaiqiang WangGuihai ChenDejing DouDawei YinPublished in: IEEE Trans. Knowl. Data Eng. (2023)
Keyphrases
- co training
- learning to rank
- web search
- semi supervised
- ranking functions
- supervised learning
- unlabeled data
- semi supervised learning
- query dependent
- web search engines
- unlabelled data
- labeled data
- search engine
- ranking models
- information retrieval
- multi view
- pairwise
- labeled examples
- unsupervised learning
- loss function
- text classification
- active learning
- single view
- gaussian process
- collaborative filtering
- clickthrough data
- machine learning
- semi supervised learning methods
- web pages
- training data
- document retrieval
- named entities
- test collection