The paper repetition rate = the number of repeated words in the paper/the total number of words in the paper * 100%, which is the formula for calculating the paper repetition rate. Note that it is a formula, not an algorithm! Then, according to this algorithm, the duplication checking system determines the content duplication by comparing it with the content in the database. The repetition rate of undergraduate graduation thesis is generally below 30%, strictly below 20%, and some graduation thesis requirements are below 10%. Generally, the repetition rate of high-quality graduation thesis is within 15%; Less than 10% of graduate students and less than 5% of doctoral students.

The paper detection system will divide the paper into sections and clauses according to its own algorithm, extract the segmented content and compare it with the content in the system's own literature collection. Generally, 7-8 words are plagiarism. After the comparison is completed, if obvious plagiarism is found, it will be marked with obvious color. Finally, the results will be obtained according to the formula of repetition rate and displayed in the generated test report. In the paper inspection report, the number of repeated characters will be displayed. Generally speaking, repeated content will be expressed in two colors, one is yellow, which means that the text has been "quoted"; The other is red, indicating that the text has been "copied". From the reference data of the maximum monji replication rate of a single report, we can clearly see the repeated words in the yellow position and the total words on the right, so we can know the repetition rate of our paper according to the repetition rate formula.