A Method of MOBA Game Lineup Recommendation Based on NSGA-II.

The existing research on MOBA game lineup recommendation mainly takes the heroes with high winning rate in the historical game as the recommendation results, without comprehensively considering the synergy relationship and the counter relationship between heroes. In the recommended results, there will be a situation where a hero has great ability, but the synergy between the heroes is not good, or the recommended hero is counter by the enemy. This paper summarizes the problem as a multi-objective optimization problem, and proposes a MOBA game lineup recommendation method base on NSGA-II. Firstly, based on the performance of the heroes in historical games, the ability of these heroes is quantified. At the same time, according to different heroes as teammates or opponents, the synergy relationship and the counter relationship between heroes are evaluated. Then, NSGA-II is used to solve the multi-objective optimization problem, and the recommendation result is obtained. Finally, the results are evaluated based on prediction models. The method proposed in this paper integrates the hero’s ability, the synergy relationship and the counter relationship between heroes, which has strong rationality. Experimental results show that the method has high accuracy, and can provide a new idea for solving this kind of problems.