Archive processing method based on association rule mining

Abstract

A method comprises the following steps: extracting elements of multiple dimensions of archives, performing a corresponding similarity measurement method on a to-be-tracked archive and a recorded archive based on a cloud model sequence after linear regression, and calculating the similarity between dimension attributes of the cloud model archives; judging whether the stored record file is a parent file of the file to be tracked according to the content relevance between the file to be tracked and the stored record file; and carrying out future archive prediction on the archives with the obtained father-child archive relationship and carrying out clustering verification on validity between the archives to be tracked and the stored record archives. According to the archive processing method based on association rule mining provided by the invention, the relationship among the archives can be tracked and searched through multi-layer screening and filtering based on association rules, the relationship of the archives can be determined, and the relationship determination method can be subjected to validity verification and potential relationship mining of future archives through operations such as prediction and classification.

Description

technical field [0001] Embodiments of the present invention relate to the field of text data processing, and in particular, to a file processing method based on association rule mining. Background technique [0002] Text traceability is mainly used in the fields of academic integrity detection, search engine optimization, etc. The purpose is to determine whether the content of a text is copied or adapted from another or more texts. The principle of obtaining homologous texts is mainly based on the comparison of text similarity. At the same time, with the widespread use of databases and the gradual rise of data sharing, the problem of data leakage is becoming more and more serious. And because data is often shared with multiple parties, it can be difficult to trace the source of a data breach. If there is a widely used method of tracing back to the source of a data breach, it will act as a deterrent to data breachers, thereby mitigating the growing problem of data breaches....

AI Technical Summary

Problems solved by technology

[0004] However, the Jaccard similarity is not sensitive to the word frequency response of the same similar word, and the cosine similarity can make a sensitive judgment on the direction of the vector, but it cannot make a sensitive judgment on the digital data

[0005] Therefore, the function of the existing text similarity comparison method is relatively single, and it can only make accurate judgments in one aspect, but cannot compare all aspects of the text, resulting in inaccurate final results

Images