Player negative behavior detection method, system, electronic device, and storage medium

A player negative behavior detection method, applicable to a PVP online game, wherein the method comprises:

• acquiring game data of a player during a single-round game process;
• analyzing and classifying the game data to obtain multi-dimensional detection data, wherein the multi-dimensional detection data comprises:
• game behavior data, game performance data, and social behavior data;
• and
• obtaining multi-dimensional negative detection information respectively based on the multi-dimensional detection data, determining, by a server, whether the player has a negative behavior through combining the multi-dimensional negative detection information and outputting negative behavior reminding information if yes;
• wherein the game behavior data comprises:
• movement counts, skill usage counts, item usage counts, attack counts, real-time positions of the player in the single-round game process;
• the game performance data comprises:
• defeat counts, assist counts, death counts, and total damage;
• and
• the social behavior data comprises:
• negative speech counts and cooperation request response counts;
• wherein the step of determining, by the server, whether the player has a negative behavior based on the game performance data comprises:
• obtaining a plurality of different types of single-round performance averages according to the defeat counts, the assist counts, the death counts, the total damage, and a single-round game duration respectively, and assigning different weight coefficients to each of the single-round performance averages respectively;
• performing a weighting operation on the plurality of single-round performance averages after assigning the weight coefficients, so as to obtain a game performance score of each player;
• obtaining a benchmark score as a reference standard for measuring a performance of the player according to game performance scores of players in the same camp during the single-round game process, wherein the benchmark score reflects a performance level that the player should achieve when actively participating in the game and reasonably exerting an ability of a character under ideal conditions;
• and
• determining, by the server, whether the player has the negative behavior according to a difference between a game performance score of any one of the players and the benchmark score, and differences between the game performance score of any one of the players and that of the other players in the same camp, and outputting the negative behavior reminding information if yes,
• wherein the method further comprises:
• generating a player position heatmap according to a real-time position of the player in the single-round game process;
• and
• determining, by the server, whether the player has the negative behavior according to a proportion of the player position heatmap to a global game map, if yes, calculating a penalty severity and punishing the player accordingly.