Chinese Multi-Keyword Fuzzy Rank Search over Encrypted Cloud Data Based on Locality-Sensitive Hashing

Most of the existing Chinese keyword fuzzy searchable encryption schemes realize fuzzy keyword search utilizing the wildcard and gram methods to construct the fuzzy set, which consumes a lot of storage and computation overheads. In this paper, we propose a novel Chinese multi-keyword fuzzy rank searchable encryption scheme, which achieves efficient fuzzy keyword search without constructing a large fuzzy set. First, the Chinese keyword is converted to the pinyin string, which is partitioned based on unigram, or the mandarin consonant, vowel and tone of pinyin. Then, we design two Chinese keyword vector generation algorithms to convert a pinyin string into a keyword vector. Moreover, the locality-sensitive hashing and Bloom filter are utilized to construct the fuzzy keyword search algorithm. We design two schemes to realize the Chinese fuzzy multi-keyword search, and all of them utilize a single Bloom filter as the encryption index of a document. The cloud storage server only needs to add (or delete) an encrypted file and its encrypted index to realize the dynamic update of the files. To improve the accuracy of the rank, a three-factor rank algorithm is proposed. The theoretical analysis and experimental results indicate that the proposed schemes realize Chinese multi-keyword fuzzy search, more accurate search result rank, guarantee the data security, and save a large amount of storage and computation costs.