SynTF: Synthetic and Differentially Private Term Frequency Vectors for Privacy-Preserving Text Mining.
Benjamin WeggenmannFlorian KerschbaumPublished in: CoRR (2018)
Keyphrases
- differentially private
- privacy preserving
- text mining
- differential privacy
- text documents
- text classification
- privacy guarantees
- privacy preservation
- privacy preserving data mining
- natural language processing
- retrieval model
- information extraction
- vector space
- private data
- data privacy
- private information
- topic models
- knowledge discovery
- bag of words
- information retrieval
- text categorization
- sensitive information
- machine learning
- data mining
- feature extraction
- data analysis
- feature vectors
- document clustering
- privacy protection
- high dimensional
- sensitive data
- keywords
- metadata
- data warehouse
- query expansion