Steam在今天正式推出了一个全新的实验性游戏推荐系统 - 交互式推荐模型。大家可以点击此处查看自己专属的推荐列表。推荐页面分为两部分，左侧为玩家游戏时长前50名的已拥有游戏，按时长从上至下依此排序；右侧则为“为您推荐”一栏，大家可以根据“热门程度”、“发售时间”、包含/排除特定标签，寻找自己心仪的游戏。在设定完条件后，推荐列表就会按推荐程度从上至下依次排序。在Steam博文中，Valve表示该推荐系统“充分利用机器学习的威力，根据玩家各自的游玩模式，向他们进行个性化的推荐”，“可以将输出结果限制在某一特定时间段内发行的游戏，还可以调整偏向更热门或更冷门的游戏...玩家可以选择只在结...
Steam officially launched a new experimental game recommendation system today - - nbsp; interactive recommendation model. You can click here to see your own list of recommendations. Recommendation page is divided into two parts, the left side is the first 50 games that have been played by the players, sorted from top to bottom according to the length of time; the right side is the column of "Recommendation for you". You can find your favorite games according to the "popularity", "sale time", "including/excluding specific tags". After setting the conditions, the recommendation list will be sorted from top to bottom according to the recommendation degree. In Team's blog post, Valve said that the recommendation system "makes full use of the power of machine learning to personalize recommendation to players according to their own game modes", "can limit the output to a certain period of time for the release of the game, can also adjust the bias towards more popular or less popular games... Players can In order to choose only in knot.
