![]() ![]() Now that's two true measures of success: the cheat sellers are finding that it's harder to produce these accounts, reflected in price rises, and some seem to have moved on from PUBG entirely. ![]() We have also observed an increase in the prices of these accounts." Furthermore, the internal monitoring process for suspected disruptive players/accounts has shown continuous improvement, and the number of monitored account vendors has decreased. "The number of bans issued against disruptive accounts has increased by over threefold compared to the period before the introduction of this model. So the PUBG anti-cheat team "initiated the development of a machine learning model that could learn the characteristics and patterns of Mastery Level abuse." It began to be used this year and Krafton has "expanded and refined the criteria for detecting disruptive players" and found great success. It says that while it was previously focused on pattern-spotting to identify hacked accounts, this had issues such as new forms of abuse going under the radar and the low accuracy of detection meaning it was difficult to apply serious punishments like a permanent ban (because there's always a chance that suspicious behaviour may well just be a legitimate player behaving suspiciously rather than cheating). Krafton says it's had enough, and wants to address this supply of accounts that's behind so much of the cheating in PUBG.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |