DCDIR: A Deep Cross-Domain Recommendation System for Cold Start Users in Insurance Domain.
Ye BiLiqiang SongMengqiu YaoZhenyu WuJianming WangJing XiaoPublished in: CoRR (2020)
Keyphrases
- cross domain
- cold start
- collaborative filtering
- recommender systems
- transfer learning
- cold start problem
- data sparsity
- recommendation systems
- domain adaptation
- multiple domains
- recommendation algorithms
- personalized recommendation
- matrix factorization
- knowledge transfer
- tag recommendation
- user ratings
- sentiment classification
- implicit feedback
- target domain
- user profiles
- text categorization
- user preferences
- user interaction
- end users
- e government
- information overload
- machine learning
- filtering algorithm
- labeled data
- context aware
- web content
- user generated content
- user interests
- feature selection
- learning algorithm