4 results about "Comparison sort" patented technology

The invention provides a multi-stage content quality evaluation method and system based on a large language model, and relates to the technical field of natural language processing. The method comprises the steps of obtaining a to-be-evaluated content data set and an evaluation index corresponding to a data type of the to-be-evaluated content data set; based on the evaluation indexes, utilizing a first large language model to carry out batch multi-dimensional coarse screening on the content data set, obtaining evaluation scores of all content data, sorting the evaluation scores, and filtering and retaining a plurality of content data with the top scores; an improved ELO dynamic scoring mechanism is introduced, a second large language model is used for conducting dynamic pairwise comparison sorting on the filtered content data, final ELO scores of all the content data are obtained and sorted, and an ELO score sorting result serves as a content quality sorting result. According to the method, a large language model is utilized to break through the cognitive depth bottleneck, evaluation objectivity is improved by relying on a dynamic game mechanism, two-stage architecture is combined to balance calculation efficiency and sorting precision, and finally a content quality evaluation universal solution capable of being rapidly adapted to multiple fields is formed.

The invention discloses a two-party privacy intersection method and system supporting any data scale, and relates to the technical field of data security processing. The method comprises the following steps: acquiring an original data set of two parties needing to interact, and performing bucket separation on the original data of the two parties; deduplication is executed on each sub-bucket in parallel; performing safety protection operation on a single bucket based on the data volume owned by a single sub-bucket after the initiator sub-buckets to generate a corresponding transparent vector linear valuation quantity; based on the transparent vector linear valuation number, the initiator and the receiver perform privacy intersection on the data by using oblivious key value storage; and performing comparison, sorting and search operations on the de-duplicated data in each sub-bucket to obtain the position of the intersection element before de-duplication in the original data sets of the two parties. According to the method, rapid privacy intersection of any scale of data scenes is realized, and a complete and efficient solution is provided for wide landing of privacy set intersection in actual services.

The invention provides an order-preserving encryption method, a data query method and a data storage method. The order-preserving encryption method comprises the following steps: acquiring plaintext data, and performing order-preserving random mapping on the plaintext data to obtain first order-preserving coded data; carrying out decryptable encryption processing on the plaintext data to obtain decryptable ciphertext data; and splicing the first order-preserving coded data and the decryptable ciphertext data to obtain order-preserving ciphertext data. And according to the size sequence relationship of the first order-preserving codes, operations such as sequence-related comparison sorting query and the like in the secret state database can be quickly completed, and a plaintext form result corresponding to query is obtained. Compared with a common order-preserving encryption algorithm, the order-preserving coding encryption method and device can realize an efficient and safe order-preserving coding encryption function with a controllable ciphertext length.