Below the dotted line is the proces

Below the dotted line is the processing flow of real-time calculation in the computer room. After the user posts and replies, it is written into the Kafka message queue. First, the whitelist users are filtered. Then, according to the training effect of the classification model, the text is identified and classified. Distinguish between spam and non-spam posts; another way is to calculate the similarity of the text through the algorithm, calculate the similarity statistics, and finally summarize these strategies to generate the entire anti-spam algorithm. Recently, Meiyou has made further iterations on this algorithm, analyzing the user's nickname, posting interval, and posting behavior. Due to the use of anti-spam algorithm, Meiyou APP has prevented spam.