Naver Uses AI to Analyze Sentence Context and Prevent Malicious Comments

Even Without Profanity, Insulting Comments Will Be Blinded

[Asia Economy Reporter Jin-gyu Lee] Naver is using artificial intelligence (AI) technology to analyze sentence context and filter out malicious comments.

On the 19th, Naver announced that it has enhanced its 'AI Cleanbot,' which prevents the exposure of malicious comments using AI. Until now, AI Cleanbot has automatically detected and blinded comments containing profanity and vulgar language. It has been successively applied to webtoons, Juniver, sports, and news since April last year.

The upgraded AI Cleanbot has expanded its criteria for judging malicious comments from just profanity words to sentence context starting today. Accordingly, even if there is no vulgar language, comments judged to be insulting based on sentence context will be blinded. Naver stated, "We analyzed the characteristics of colloquial comments with many abbreviations and typos based on big data of malicious comments."

AI Cleanbot also automatically identifies habitual writers of malicious comments. Habitual malicious commenters recognized by AI Cleanbot will be restricted from using the comment service for a certain period.

Meanwhile, Naver is reportedly preparing a policy to limit the number of comments that can be written and the number of likes that can be given.

This content was produced with the assistance of AI translation services.