Privacy protection K-NN (Kth Nearest Neighbor) classification method based on vector homomorphic encryption

Abstract

The invention relates to a privacy protection K-NN (Kth Nearest Neighbor) classification method based on vector homomorphic encryption, which comprises the steps of: A, receiving an inquiry vector set and a standard vector set; B, generating a matrix G by the inquiry vector set, and enabling the standard vector set to use a secret key S to generate a ciphertext group and a new secret key GS by vector homomorphic encryption; C, carrying out secret key conversion on the new secret key GS to form a converted secret key S' so as to obtain a converted matrix M and a converted ciphertext group at the moment; D, carrying out decryption on the converted ciphertext group by using the converted secret key S' so as to obtain a decrypted vector set; and E, according to components of decrypted vectors of k minimum values, attaching a classification tag to each corresponding inquiry vector. According to the privacy protection K-NN classification method disclosed by the invention, privacy data of a user can be excellent protected; and moreover, in a case that the privacy data is protected, the inquiry vectors of the user are efficiently and accurately classified by a K-NN algorithm, so that efficiency of judging a vector type is improved, and an application range of judgment on the vector type is enlarged.

Description

Technical field [0001] The present invention relates to a vector classification method of K-NN algorithm under privacy protection, and specifically is a privacy protection K-NN classification method based on vector homomorphic encryption. Background technique [0002] K-Nearest Algorithm (K-NN) is widely used in pattern recognition, and it also has a good performance in data classification. In the era of big data, complex calculations are often outsourced to third-party clouds. However, in this process, the outsourced data may include some sensitive data of sensitive users. Outsourcing the user's information directly to a third-party cloud may cause some of the user's private information to leak out and cause losses to the user. A direct solution is to encrypt the data before uploading it to the cloud, but this also brings some new challenges to the efficiency of the K-NN algorithm. Homomorphic encryption technology can operate under ciphertext, and the result after decryption ...

AI Technical Summary

Problems solved by technology

Due to the K-nearest neighbor rule for probability estimation density, there are two disadvantages. One is that the integral of the estimated density must be diffused to infinity, and the other is that there is a heavy burden on the calculation.

Images