Server-Based User Psychology Classification for Game Personalization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional game recommendation systems fail to accurately match users with game content that aligns with their psychological tendencies, leading to user fatigue and dissatisfaction due to the gap between survey data and actual game behavior, and the inability of AI algorithms to reflect users' psychological states.
Innovation Solution
A server-based method that analyzes user psychology by generating reference data from game service provider databases, using various models to classify users based on their play patterns, character usage, and financial interactions, to recommend content tailored to their specific tendencies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If conventional AI algorithms are used for content recommendation, then recommendation speed is improved, but psychological accuracy deteriorates because they only learn from purchase probability without reflecting users' psychological states
Solution Approach 1:
The patent segments the recommendation system into multiple independent modules: a psychological state analysis module that processes game behavior data to determine user psychology, and a content recommendation module that uses these psychological states for targeted recommendations. This segmentation allows each module to specialize in its function, improving both speed and accuracy.
Solution Approach 2:
The patent introduces psychological state information as an intermediary between raw game behavior data and content recommendations. Instead of directly mapping purchase probability to recommendations, the system first translates behavior data into psychological states (e.g., competitive, cooperative, exploratory), which then guide content selection, thereby capturing nuanced user psychology.
2Loss of information
If survey methods are used to understand users' tendencies, then direct user feedback is obtained, but there is a gap between survey data and actual game behavior making it difficult to accurately understand tendencies
Solution Approach 1:
The system uses game behavior data that users generate naturally during gameplay as self-service data sources. Instead of requiring users to complete surveys, the system automatically collects and analyzes in-game actions, character selections, and interaction patterns, eliminating the gap between stated preferences and actual behavior.
Solution Approach 2:
The patent implements continuous feedback loops where game behavior data is constantly collected, analyzed to update psychological state profiles, and used to refine future recommendations. This dynamic feedback mechanism ensures that the system adapts to changing user tendencies based on actual gameplay rather than static survey responses.
Data Source
AI summary
Disclosed is an operating method of a server which analyzes a user's psychology to classify the user. The operating method includes receiving a raw database from a server of a game service provider, acquiring setup information of at least one module to acquire reference data corresponding to one of a plurality of models from the raw database, generating reference data from data of users included in the raw database on the basis of at least one module which is set on the basis of the setup information, and acquiring psychological state information of a user to be analyzed on the basis of the reference data.