US 12,011,667 B2
Apparatus for generating standardized table for classifying the psychology of a game user and an operation thereof
Hye Min Kwon, Busan (KR); In Su Gim, Seoul (KR); Jin Kim, Goyang-si (KR); and Hye Yon Kwon, Yongin-si (KR)
Assigned to Sentience Inc., Seoul (KR)
Filed by Sentience Inc., Seoul (KR)
Filed on Dec. 30, 2021, as Appl. No. 17/565,763.
Claims priority of application No. 10-2020-0146536 (KR), filed on Nov. 5, 2020.
Prior Publication US 2022/0323870 A1, Oct. 13, 2022
Int. Cl. A63F 13/798 (2014.01); A63F 13/792 (2014.01)
CPC A63F 13/798 (2014.09) [A63F 13/792 (2014.09)] 6 Claims
OG exemplary drawing
 
1. A method of operating a server, which includes a processor and a memory, that standardizes game data of a user to classify a user's game propensity, the method comprising:
storing raw data including log data related to activities of users in different games in a raw database in real time;
transforming the raw data related to a plurality of events occurred for a plurality of users in the raw database to standardized values using a data standardization function, the data standardization function filtering the raw data, wherein the raw data includes metadata of a game and play records of the game, and structure of raw data is different for different games;
transforming structure of the raw database including the raw data into structure of a standardized database comprising a plurality of standardized tables;
obtaining a plurality of standardized tables for the plurality of users including the standardized values;
generating a distribution graph based on the plurality of standardized tables, a first axis of the distribution graph representing a first parameter of the standardized tables, and a second axis of the distribution graph representing a second parameter of the standardized tables;
determining reference data based on a maximum point or a minimum point of the distribution graph;
determining whether an event occurs within a game while a user is playing the game;
calling the data standardization function corresponding to the event when the event occurs within the game, the data standardization function receiving raw data related to the event in association with the user and acquiring at least one predetermined parameter by selecting some of the raw data;
acquiring the at least one predetermined parameter based on the data standardization function;
storing the at least one predetermined parameter in a standardized table for the user corresponding to the event;
classifying a user's propensity based on the table; and
outputting, to the user, game content recommended based on the classified user's propensity,
wherein the classifying of the user's propensity includes:
generating the reference data which is a criterion for classifying the user into one of a plurality of predetermined propensities;
generating feature data for an account of the user from the standardized table for the user; and
selecting the user's propensity from among the plurality of predetermined propensities based on a comparison of the reference data and the feature data for the account of the user, and
wherein the method further comprises:
acquiring a machine learning model to determine a relationship between users' propensities and parameters included in a plurality of standardized tables; and
classifying the user's propensity based on the table and the machine learning model.