Firstly, the principle and process of duplicate checking system are introduced, and the problem of how to check duplicates when there are tables is put forward. Through the analysis of a large number of sample data, it is found that tables do affect the repetition rate of papers. Specifically, if the paper contains a large number of tables, and most of the table contents come from external resources, then these tables may become "key areas of concern".

In order to solve this problem, an improved algorithm based on table analysis is proposed, which can retain the information of tables and reduce the influence of tables on repetition rate. In practice, we apply this algorithm to a large number of sample data and compare it with the traditional duplicate checking algorithm. The results show that the algorithm can improve the accuracy of paper duplicate checking and effectively reduce the influence of tables on repetition rate.

Generally speaking, this paper mainly discusses the problem that the table will affect the repetition rate when checking duplicates, and puts forward an improved algorithm based on table analysis to improve the accuracy of checking duplicates. In addition, we also conducted a comprehensive analysis and research on the paper duplicate checking and duplicate checking system, which provided some valuable ideas and references for future research. There are many duplicate checking systems in the market, and the database algorithms are different. It is the best to choose the duplicate checking system that is most suitable for our paper. The system has high security, and some cost-effective duplicate checking systems can be used when drafting the first draft. Then Papertime paper duplicate checking system is very suitable for early detection, and the final detection report results are reliable and the algorithm mechanism is strong. After the paper is finalized, the duplicate checking system consistent with the school should be used for final testing.