A method and device for storing and analyzing game player data

Abstract

The invention discloses a game player data storage and analysis method and apparatus. The analysis method comprises the following steps: 1), storing player data of game role groups in the form of a first table and a second table, wherein in the first table, a row name is date, column names include a column cluster name and sub-column names, the column cluster name is a game behavior class of a player, the sub-column names are specific behaviors in the game behavior class, and cells of the first table store roles participating in the specific behaviors and measurement values; in the second table, a row name is a date and role combination, column names include a column cluster name and sub-column names, and cells of the second table store measurement values of corresponding roles participating in the specific behaviors at corresponding date; 2), extracting a screening condition according to an analysis condition and screening out role groups meeting the condition from the first table; and 3), according to the analysis condition, obtaining measurement value information of the behaviors of the role groups from the second table. According to the method and the apparatus, high screening analysis efficiency can be achieved.

Description

【Technical field】 [0001] The invention relates to a method and device for storing and analyzing massive player data of a game character group in a game. 【Background technique】 [0002] Games, such as online games, mobile games, etc., involve a large number of game characters, and each game character corresponds to a part of player data, such as the number of times of participating in the instance, the length of login time, activity level, consumption status, etc. Correspondingly, a large amount of player data will be generated in the game. Existing player game data is stored using a traditional relational database. Generally speaking, game data needs to save daily player data. Therefore, when using relational database storage, the day is generally used as the primary key of the table, and then the id of the game character and various behaviors of the character are stored in the table as other columns. As shown in Table 1. [0003] Table 1 [0004] date role ...

AI Technical Summary

Problems solved by technology

[0006] In the above process, there is a problem that the analysis efficiency is very poor

The reason is that a general game has tens of millions or even tens of millions of game characters. It is difficult and inefficient to find a qualified character ID from such a large number of records.

Furthermore, after filtering out qualified character ids, the number of these character groups may not be small, and it is very inefficient to further search for their other behavior data based on these character ids

Therefore, the efficiency of traditional storage and analysis methods is very poor

Images