چكيده به لاتين
Nowadays online games are very profitable for the gaming companies and that is why many companies tend to be active in this field. With the rise of online gaming in the world and easy access to the Internet, gaming companies have come across a wealth of data that tends to make data mining techniques more useful and profitable. This plan aims to discover player patterns) Pattern of players' victory( and methods. The winner of the online gambling game offers a good model of the methods and accessories used to win the new players in the game, as well as offers suggestions to the gaming companies through the discovered patterns and methods. Companies Playing the importance and impact of accessories that make players win and understand the benefits of their company is well planned, as well as the more effective steps to develop and deliver their services. This research looks at the historical background of machine learning for classic games and follows an online game called Rooster Wars. This study provides a comprehensive analysis of the board factors and updating the data in this interval using an intelligent online game analysis system. In this study, we present a new structure for Rooster Wars online game analysis and the effects of combat and weapon upgrades that utilize ELM-based intelligent machine learning systems. The existence of high learning speed and adjustment of one parameter in the training phase, as opposed to setting many parameters in the training phase in neural networks, are the reasons for using this algorithm. In general, ELM has a Feed-Forward structure and uses quasi-inverse structures to calculate real-time synaptic weights, which results in faster learning and data testing. According to the results of analysis and research, according to the maximum victory score, the best weapon used during the analysis was the spear with the maximum duration and the best cost.
