首页>
外国专利>
Privacy protection probabilistic inference based on hidden Markov model
Privacy protection probabilistic inference based on hidden Markov model
展开▼
机译:基于隐马尔可夫模型的隐私保护概率推理
展开▼
页面导航
摘要
著录项
相似文献
摘要
Parameter Hidden Markov Model (HMM) is asked by the server based on the observed value series to be stored at the client, the client has the encryption key and decryption key additive homomorphism encryption system, the server only the encryption key have. The server initializes the parameters of HMM, the difference between the probability of the observed value series of iterations prior probability of the observed value series of iterations until the current exceeds a threshold value, update iteratively the parameters, repetitive each time it is, the conditional probability when the parameters of the HMM and observation sequence is given, is updated based on the conditional joint probability that is encrypted for each pair of states, the encrypted parameter to the server is obtained in the domain is encrypted using secure multiparty computation between the client and the (SMC).
展开▼