Generally speaking, any plagiarized part will be marked after repeated inspection, and then the repetition rate will be calculated according to the calculation method. Similar words = 1 sentence words * 1 sentence words +2 sentence words * n sentence words, the sentence similarity range is 0.00- 1.00, and the green sentence similarity is calculated as 0.

One-sidedness of database

The paper detection system compares the submitted papers with the database contained in the system itself. Only the system database itself contains the paper data you plagiarized can it be marked. For various reasons, it is impossible for each platform to collect all the paper data, so some plagiarism can't be found out. We can upload the plagiarized articles to the self-built database and then check them, so that the test results will be more accurate.

Two, don't appear in the system calculation more than the standard value of repeated formula.

Take the paper detection system of HowNet as an example. Only consecutive occurrences of 13 similar characters will be marked as plagiarism. If our paper does not reach 13 similar characters, it will not be detected. Different websites have different rules. You can learn more about the website you want to check before revising the paper.

Third, when re-checking papers, we must choose a formal platform for detection, otherwise everyone knows that this is part of plagiarism, but the detection system can't detect it, which means that formal websites will have formal data and services. In order to ensure the accuracy of the test results of our paper.