1. Generally, the parts with high similarity will be colored in the duplicate checking report. You can directly and clearly see which parts need to be modified. If some small fragments and sentences are too similar, although there are not many words, they are also too similar parts, and we should also focus on revision. This part can be deleted directly, or the sentence can be reorganized with some attributive adverbials, or the sentence can be changed into word order voice, and then some words with similar meanings can be changed to make it a new sentence. These methods can effectively reduce the similarity of papers.

2. If it is detected that the similarity of the references cited in the whole paragraph is too high, and the detection system of the references cited will generally automatically identify that the similarity is not detected, this situation is generally that the symbol format is incorrect, and it is necessary to modify the symbol format. Try not to use a period after quoting symbols, and the detection system will generally identify the quoted content after a period as plagiarism. Therefore, when the quoted content is not completely finished, it is best to use semicolons and superscript the quoted symbols.

3. Many majors have fixed technical terms and need relevant experiments, so the similarity detection of data terms in this area will be very high. At this time, the data in this area can be made into tables, and the methods and styles of making tables will be different for everyone, and many paper detection systems will not detect tables, so the similarity of papers can be effectively reduced. For technical terms that are difficult to modify, you can insert some pictures and express them with words on them, which can also effectively reduce the similarity of papers.