Deep Reinforcement Learning-based Text Anonymization against Private-Attribute Inference.
Ahmadreza MosallanezhadGhazaleh BeigiHuan LiuPublished in: EMNLP/IJCNLP (1) (2019)
Keyphrases
- reinforcement learning
- privacy preserving
- sensitive attributes
- information retrieval
- text retrieval
- private information
- semantic attributes
- privacy preservation
- database
- bayesian inference
- free text
- belief nets
- private data
- information loss
- function approximation
- belief networks
- learning algorithm
- bayesian networks
- keywords
- text fragments
- optimal policy
- text mining
- natural language processing
- dynamic programming
- model free
- reinforcement learning algorithms
- machine learning
- markov decision processes
- lexical semantics
- learning process
- rough sets
- probabilistic inference
- attribute values
- text documents