Game track data preprocessing method and related equipment

The invention provides a game track data preprocessing method and related equipment. The method comprises the following steps: converting game track data into a first grid sequence; combining high-frequency adjacent grids in the first grid sequence, and calculating to obtain a second grid sequence;...

Full description

Saved in:
Bibliographic Details
Main Authors TAO JIANRONG, SHEN QIAOZHI, FAN CHANGJIE, LYU TANGJIE, LIN JIANSHI, LIU HAOYU, WU RUNZE, HU ZHIPENG, ZHAO SHIWEI
Format Patent
LanguageChinese
English
Published 21.10.2022
Subjects
Online AccessGet full text

Cover

Loading…
More Information
Summary:The invention provides a game track data preprocessing method and related equipment. The method comprises the following steps: converting game track data into a first grid sequence; combining high-frequency adjacent grids in the first grid sequence, and calculating to obtain a second grid sequence; extracting an embedded vector of the second grid sequence, and calculating to obtain an embedded vector sequence; the embedded vector sequence serves as a training set to be input into a deep learning model. According to the embodiment of the invention, the game track data is converted into the first grid sequence, the high-frequency adjacent grids of the first grid sequence are merged, and then the embedding vector of the second grid sequence is extracted, so that the prior influence of environmental factors on track distribution and potential context semantic information in the track can be captured; effective compression of the long-track data is realized. 本申请提供一种游戏轨迹数据预处理方法及相关设备。所述方法包括:将游戏轨迹数据转换为第一网格序列;合并所述第一网格
Bibliography:Application Number: CN202210723022