Privacy-Preserving Distributed Expectation Maximization for Gaussian Mixture Model Using Subspace Perturbation.
Qiongxiu LiJaron Skovsted GundersenKatrine TjellRafal WisniewskiMads Græsbøll ChristensenPublished in: ICASSP (2022)
Keyphrases
- privacy preserving
- gaussian mixture model
- expectation maximization
- em algorithm
- horizontally partitioned data
- mixture model
- data perturbation
- multi party
- horizontally partitioned
- partitioned data
- privacy preserving data mining
- privacy guarantees
- privacy sensitive
- maximum likelihood
- gaussian mixture
- probabilistic model
- vertically partitioned data
- feature space
- privacy preserving classification
- generative model
- privacy preservation
- secure multiparty computation
- privacy concerns
- feature vectors
- probability density function
- image segmentation
- data privacy
- density estimation
- private information
- sensitive information
- scalar product
- preserving privacy
- k means
- privacy protection
- high dimensional
- differential privacy
- gaussian model
- privacy preserving association rule mining
- unsupervised learning
- sensitive data
- privacy issues
- private data
- dimensionality reduction
- low dimensional
- distributed environment
- distributed data
- statistical databases
- feature selection
- text classification
- medical images
- data points
- decision trees