Refining Graph Representation for Cross-Domain Recommendation Based on Edge Pruning in Latent Space.
Taisei HirakawaKeisuke MaedaTakahiro OgawaSatoshi AsamizuMiki HaseyamaPublished in: IEEE Access (2022)
Keyphrases
- graph representation
- cross domain
- latent space
- transfer learning
- low dimensional
- graph model
- latent variables
- knowledge transfer
- gaussian process
- text categorization
- manifold learning
- parameter space
- target domain
- labeled data
- active learning
- dimensionality reduction
- gaussian processes
- learning tasks
- lower dimensional
- generative model
- high dimensional
- reinforcement learning
- weighted graph
- feature space
- high dimensional data
- probabilistic latent semantic analysis
- text mining
- machine learning
- matrix factorization
- machine learning algorithms
- probabilistic model
- data mining
- unlabeled data
- semi supervised learning
- e government
- collaborative filtering
- information extraction
- bayesian networks