During the last decade, game developers and researchers have started to extensively collect in-game data of players in order to guide decision-making. This has led to the emergence of the new field of game analytics, that is, data analytics in the context of games. Over the last years, game analytics has considerably contributed to our understanding of player behavior. However, with the increase in volume and variety of data, a need for effective data visualization has arisen to support the interpretation of the complex and multimodal data sets and to provide various stakeholders with actionable intelligence. Despite increased attention, data visualization for game analytics is still in its infancy and many challenges in visualizing in-game data remain. Paramount, among other challenges, is the dynamic nature of game data and its high dimensionality. Dynamic data entails the fact that in live games such as Destiny or The Division, players are regularly generating a stream of data that affects their profile constantly. High dimensionality of game data means that it must be possible to account for where and when events occur as well as the virtual context when decisions are made. Thus, there is much research looking into how to reduce the dimensionality or to build interactive visualizations systems to tackle this problem. On the contrary, stakeholders who would like to look at game data usually have different needs and would like to see different level-of-details in their visualizations. This necessitates customization and interactive adaptive techniques. How to design such systems for effective knowledge discovery is thus still a challenge. This Special Issue of Information Visualization collects novel work on visualization techniques in game analytics. Beyond that, another important goal of this issue is to raise awareness and to strengthen the visibility of this emerging interdisciplinary research area. The call for papers attracted eight submissions. After a double-blind review process, four submissions were selected for inclusion. Despite being a rather young field of research, these papers are testament to the diversity of research currently being carried out. ‘‘Visualizing the strategic landscape of arbitrary games’’ co-authored by Joseph Carter Osborn, Benjamin Samuel, and Michael Mateas describes a game-independent visualization of sequences of actions which players take, that is, play traces. Play traces are visualized as vertical lines with symbols along the lines representing game events and lines bending toward and away from each other depending on how similar the respective play traces are to each other. Two cases studies illustrate how the approach can be used in practice. In ‘‘Visualizing log file data from a game using timed word trees,’’ Paula Ceccon Ribeiro and colleagues are concerned with investigating gameplay patterns and pedagogically relevant traits in educational games. For that purpose, they make use of VisCareTrails—originally intended for visualizing electronic health records—which, in turn, builds upon the concept of timed word trees. Timed word trees not only show the sequence of events but also the timing of the sequence and thus seem well suited for analyzing player strategies as demonstrated in this article. In ‘‘Enjoying death among gamers, viewers, and users: A network visualization of Dark Souls 3’s trends on Twitch.tv and Steam platforms,’’ Enrico Gandolfi is not so much concerned with the game playing

中文翻译：

---

视觉游戏分析特刊介绍

在过去的十年中，游戏开发商和研究人员开始广泛收集玩家的游戏内数据以指导决策。这导致了新的游戏分析领域的出现，即游戏背景下的数据分析。在过去的几年里，游戏分析极大地促进了我们对玩家行为的理解。然而，随着数据量和种类的增加，需要有效的数据可视化来支持对复杂和多模式数据集的解释，并为各种利益相关者提供可操作的情报。尽管受到越来越多的关注，但用于游戏分析的数据可视化仍处于起步阶段，在可视化游戏数据方面仍然存在许多挑战。除其他挑战外，最重要的是，是游戏数据的动态特性及其高维度。动态数据需要这样一个事实，即在《命运》或《全境封锁》等实时游戏中，玩家会定期生成不断影响其个人资料的数据流。游戏数据的高维度意味着必须能够说明事件发生的地点和时间以及做出决策时的虚拟环境。因此，有很多研究在研究如何降低维度或构建交互式可视化系统来解决这个问题。相反，想要查看游戏数据的利益相关者通常有不同的需求，并希望在他们的可视化中看到不同层次的细节。这需要定制和交互式自适应技术。因此，如何设计此类系统以进行有效的知识发现仍然是一个挑战。本期《信息可视化》特刊收集了有关游戏分析中可视化技术的新作品。除此之外，本期的另一个重要目标是提高认识并加强这一新兴跨学科研究领域的知名度。论文征集吸引了八份提交。经过双盲审查过程，选择了四份提交内容。尽管是一个相当年轻的研究领域，但这些论文证明了目前正在进行的研究的多样性。约瑟夫·卡特·奥斯本 (Joseph Carter Osborn)、本杰明·塞缪尔 (Benjamin Samuel) 和迈克尔·马泰斯 (Michael Mateas) 合着的“可视化任意游戏的战略格局”描述了一种独立于游戏的玩家所采取行动序列的可视化，即，玩痕迹。游戏轨迹被可视化为垂直线，沿线的符号代表游戏事件，线条朝向和远离彼此弯曲，具体取决于各个游戏轨迹彼此的相似程度。两个案例研究说明了如何在实践中使用该方法。在“使用定时词树可视化游戏日志文件数据”中，Paula Ceccon Ribeiro 及其同事关注研究教育游戏中的游戏模式和教学相关特征。为此，他们利用 VisCareTrails——最初用于可视化电子健康记录——反过来，它建立在定时词树的概念之上。定时词树不仅显示事件的顺序，还显示顺序的时间，因此似乎非常适合分析本文中演示的玩家策略。在《在游戏玩家、观众和用户中享受死亡：黑暗之魂 3 在 Twitch.tv 和 Steam 平台上的趋势的网络可视化》中，恩里科·甘多菲 (Enrico Gandolfi) 不太关心游戏玩法