• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>Australasian Joint Conference on Artificial Intelligence >Analysis and Prediction of Player Population Changes in Digital Games During the COVID-19 Pandemic
【24h】

Analysis and Prediction of Player Population Changes in Digital Games During the COVID-19 Pandemic

机译：Covid-19流行病中数字游戏的球员人口变化的分析与预测

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

The demand for video games increased in large scale during the COVID-19 pandemic as people had to stay at home. In this study we investigate the changes in player population of games during the pandemic using our dataset of 1963 games on Steam to generate insights that would be valuable for the game industry to understand the demand in such crisis. We conduct an empirical analysis to analyse changes in player population size and weekly patterns. Also, we investigate the use of machine learning classification models to predict the games that become popular during the pandemic using information about games as features. Our results indicate a 33% of increase of population during the pandemic and diminishing of weekly player population patterns. Also, we identify that the Random Forest model performs better than other classification models in predicting popular games, however, with only a 63% accuracy and tags assigned to games are the most important feature for prediction generation. Our tag analysis reveals Multiplayer, Adventure, Racing and Boardgames are popular during the pandemic.
机译：在Covid-19大流行期间，视频游戏的需求量在大规模增加，因为人们必须留在家里。在这项研究中，我们使用1963年游戏的DataSet在大流行期间调查玩家群体群体的变化，以产生对游戏行业有价值的洞察，以了解这种危机的需求。我们进行实证分析，分析玩家人口大小和每周模式的变化。此外，我们调查了机器学习分类模型的使用，以预测流行于大流行期间流行的游戏，使用有关游戏作为特征的信息。我们的结果表明，在大流行和每周球员人口模式下减少的人口增加33％。此外，我们认为随机林模型比预测流行游戏的其他分类模型更好地表现优于其他分类模型，但是，只有63％的准确性和分配给游戏的标签是预测生成最重要的特征。我们的标签分析显示了多人游戏，冒险，赛车和船上流行期间的流行。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号