Scientific research is inseparable from reading a large number of documents, but sometimes time is too tight to take care of all the documents, and sometimes the content of concern belongs to research hotspots and the number of documents is huge, so it is impossible to read the documents comprehensively. How to solve these problems? First, document tracking Many databases have the function of document tracking. For example, SCI database and goolge academic all have citation tracking. Considering the immediacy of documents, goolge is relatively good academically, and will be tracked at any time as long as it appears, but the disadvantage is that documents are miscellaneous, such as blog documents. SCI database is relatively accurate, but many documents often need to wait for a long time to be included, so it is difficult to track the documents at any time. Therefore, from the perspective of classical literature analysis, SCI database can be used, but from the perspective of the latest literature tracking, goolge is better academically. Other databases have their own advantages and disadvantages and can be selected according to their own needs. I usually choose a few very classic documents and set up the tracking and feedback function of all databases, so that I can learn from each other's strong points, track them at any time and analyze them comprehensively. Second, key intensive reading The so-called key intensive reading means reading important documents in full text. You can read it again and again, understand it carefully, and even follow the references. The fundamental purpose is to fully understand the progress in this field, deeply understand some important progress, and try to summarize and refine on the basis of these readings.