Search

Abstract The esports industry is one of the prominent business sectors in the digital era, particularly, Multiplayer Online Battle Arena (MOBA) games which gain much attention from gamers and streaming audiences. Among such games, Defense of the Ancient 2 or Dota 2 is the record holder for the highest prize esports tournament. Therefore, various companies and investors start their esports teams to compete in the Dota 2 tournaments, the Internationals. To success in the competition, player recruitment is a crucial process as it usually takes considerable effort to find a skillful player. Watching the game replay to evaluate the player’s skill is one of the approaches. However, it can be too exhaustive, also some player’s ranking, which represent the player’s skill, are often not available. In this paper, we propose an effective machine learning pipeline to evaluate the player’s skill. Our designed pipeline includes data collection, feature engineering, and machine learning modeling. We show the data collection process using open-source API. An effective method for feature engineering is proposed. Features, e.g., end-game, or tactical decision related statistics, are incorporated along with the resource in the game distribution, harassment tactic, or spatiotemporal features, in order to provide effective models. Subsequently, we apply major machine learning models based on a single game data, i.e., logistic regression and random forest, to the processed data. The most effective model can achieve up to 0.7091 precision, 0.5850 recall, 0.6411 F1-score, and 0.8123 ROC AUC score.

The esports industry is one of the prominent business sectors in the digital era, particularly, Multiplayer Online Battle Arena (MOBA) games which gain much attention from gamers and streaming audiences. Among such games, Defense of the Ancient 2 or Dota 2 is the record holder for the highest prize esports tournament. Therefore, various companies and investors start their esports teams to compete in the Dota 2 tournaments, the Internationals. To success in the competition, player recruitment is a crucial process as it usually takes considerable effort to find a skillful player. Watching the game replay to evaluate the player's skill is one of the approaches. However, it can be too exhaustive, also some player's ranking, which represent the player's skill, are often not available. In this paper, we propose an effective machine learning pipeline to evaluate the player's skill. Our designed pipeline includes data collection, feature engineering, and machine learning modeling. We show the data collection process using open-source API. An effective method for feature engineering is proposed. Features, e.g., end-game, or tactical decision related statistics, are incorporated along with the resource in the game distribution, harassment tactic, or spatiotemporal features, in order to provide effective models. Subsequently, we apply major machine learning models based on a single game data, i.e., logistic regression and random forest, to the processed data. The most effective model can achieve up to 0.7091 precision, 0.5850 recall, 0.6411 F1-score, and 0.8123 ROC AUC score. [ABSTRACT FROM AUTHOR]

Understanding player behaviour in online gaming is essential for enhancing user engagement and retention. This study utilizes a dataset from Kaggle, capturing a wide range of player demographics and in-game metrics to predict player engagement levels categorized as 'High,' 'Medium,' or 'Low.' The dataset includes features such as age, gender, location, game genre, playtime, in-game purchases, game difficulty, session frequency, session duration, player level, and achievements. The research employs a Gaussian Naive Bayes model, with data pre-processing steps including feature selection, categorical data encoding, and scaling of numerical features. The dataset is split into training (80%) and testing (20%) sets, and a 5-fold cross-validation is used to ensure model robustness. The model's performance is evaluated using accuracy, precision, recall, and F1-score. The results show consistent performance across different folds, with an average accuracy of 84.27%, precision of 85.59%, recall of 84.27%, and F1-score of 83.98%. These findings indicate that the Gaussian Naive Bayes model can reliably predict player engagement levels, identifying significant predictors such as session frequency and in-game purchases. The study contributes to game analytics by providing a predictive model that can help game developers and marketers design more engaging gaming experiences. Future research should incorporate a broader range of features, including psychological and social factors, and explore other machine learning algorithms to enhance predictive accuracy. This study's insights are valuable for developing strategies to improve player retention and satisfaction in the gaming industry.

Game Provenance has been proposed and employed for Game Analytics tasks as they capture game session data in detail and allow exploratory analysis and visualizations. Games are highly heterogeneous models with several interacting agents and game-world environment elements. Game Provenance Graphs can accommodate the heterogeneous nature of such applications with different types of nodes and edges that tend to share information across themselves, enhancing cause-effect features rarely addressed by any other approach. On the other hand, existing Heterogeneous Graph Neural Network (HGNN) solutions disregard node feature information, overlooking shared features across distinct node types, and rely on naïve approaches, such as projecting each type of node to the same n-dimensional space. We conjecture that leveraging heterogeneous feature information is essential for tackling Game Analytics tasks, especially through Machine Learning based models. To achieve that, we propose a novel approach that allows HGNNs to leverage Game Provenance Graphs' heterogeneous node feature information. Hence, we introduce in this paper three strategies for Heterogeneous Graph Representation Learning that encodes feature set information into the HGNN architecture and projects feature values leveraging similarities across such feature sets. We conduct experiments on two Game Provenance Graphs datasets, the Smoke Squadron and the Game Provenance Profile datasets, which gather game session data from different games. Our results show that encoding feature set information in the representation learning process improves the outcomes of GNN models in non-disjoint feature datasets. [ABSTRACT FROM AUTHOR]

The traditional classroom setting presents challenges when it comes to strengthening adaptive expertise in mathematics education through deliberate practice. This study aimed to investigate whether the Number Navigation Game (NNG) could help promote deliberate practice, and whether students’ performance in the game was related to their development of Adaptive Number Knowledge, perceived challenge, flow, and math interest. NNG is a game-based learning environment that requires students to progress by solving increasingly complex arithmetic problems, which is crucial for promoting adaptive number knowledge. Game performances of 214 Finnish students were analyzed and compared to the best possible performance for each game level. A growth mixture model based on the students' relative performance levels was used to gain insight into how students' game performance changed throughout the game, and how this related to their knowledge gains, perceived challenge, math motivation, and flow. There were four different profiles of students' game performance. The largest profile consisted of students who steadily improved their performance in the game, despite initially having lower-than-average performance. This group experienced lower levels of flow but achieved larger learning gains than the other groups, suggesting that their engagement may be more aligned with deliberate practice.

With the emergence of mobile distribution channels, the traditional game value chain has produced new changes, leading to the emergence of the mobile value chain. Independent (Indie) game developers can upload their games directly through third-party app stores and publish them themselves. However, many indie game developers have issues with game publishing, especially updating the new version, promoting the market, and forecasting revenue for their games. This paper aims to provide a method to guide indie mobile game developers with mobile publishing. This new method mainly focuses on addressing the main challenges from the indie game developer's side. The method includes a new concept of mobile game publishing logic and an online analysis tool along with the guidelines. It shows how to collect and analyze data and guide new version updates, marketing promotion, and revenue forecasts. In practice, the method was provided to six indie game companies and guided their mobile game publishing, and related data were collected and analyzed for evaluation. Based on the survey and interview results, the usefulness, usability, and confidence in the method were positive, and the method improved the indie game developers' mobile game publishing and benefited their game business. [ABSTRACT FROM AUTHOR]

Competitive video game playing, an activity called esports, is increasingly popular to the point that there are now many professional competitions held for a variety of games. These competitions are broadcast in a professional manner similar to traditional sports broadcasting. Esports games are generally fast paced, and due to the virtual nature of these games, camera positioning can be limited. Therefore, knowing ahead of time where to position cameras, and what to focus a broadcast and associated commentary on, is a key challenge in esports reporting. This gives rise to moment-to-moment prediction within esports matches which can empower broadcasters to better observe and process esports matches. In this work we focus on this moment-to-moment prediction and in particular present techniques for predicting if a player will die within a set number of seconds for the esports title Dota 2. A player death is one of the most consequential events in Dota 2. We train our model on ‘telemetry’ data gathered directly from the game itself, and position this work as a novel extension of our previous work on the challenge. We use an enhanced dataset covering 9,822 Dota 2 matches. Since the publication of our previous work, new dataset parsing techniques developed by the WEAVR project enable the model to track more features, namely player status effects, and more importantly, to operate in real time. Additionally, we explore two new enhancements to the original model: one data-based extension and one architectural. Firstly we employ learnt embeddings for categorical features, e.g. which in game character a player has selected, and secondly we explicitly model the temporal element of our telemetry data using recurrent neural networks. We find that these extensions and additional features all aid the predictive power of the model achieving an F1 score of 0.54 compared to 0.17 for our previous model (on the new data). We improve this further by experimenting with the length of the time-series in the input data and find using 15 time steps further improves the F1 score to 0.62. This compares to F1 of 0.1 for a standard RNN on the same task. Additionally a deeper analysis of the Time to Die model is carried out to assess its suitability as a broadcast aid.

This article aims to illuminate how game analytics have discursively shaped the mainstream perception of Southeast Asia as a regional games market through a qualitative analysis of the data and discourses in three market reports by Newzoo, an influential game analytics company that played a pivotal role in pioneering market research about the region. By reconceiving the futurity of Southeast Asia in terms of capitalist temporality, these reports envision the region as a games market of perpetual capitalist growth through data-led approaches. Despite its limitations, the compelling conception of Southeast Asia as "the world's fastest growing games market" has become a powerful myth that exerts profound influence on how the public conceive the region as a gaming space. [ABSTRACT FROM AUTHOR]

Digital educational games have evolved in recent years due to the need to support education and training focused on non-technical skills. Data gathered through interaction with the graphical user interface are explored and exploited to analyze the players' experience. Many researchers have pointed the importance to analyse players' in-game behavior, which can help to enhance the learning process, identify learners' strategies, and improve the effectiveness of the serious game. This study is devoted to the analysis of students' behavior in a simulation game called CLONE, which targets work scheduling, situation awareness, and decision-making. The students performance and their behavioral strategies are examined based on sequences analysis of players' in-game actions. Moreover, outlier detection is proposed as an instrument for obtaining information that might help better understand students behavior. The findings of the study show that such indicators as time spent on planning schedule, time spent on inspecting additional information, and intensity of delegation activity are significantly higher for successful games than for lost ones. The sequences analysis and clustering have revealed students' prevailing in-game strategies, which mostly consist of inspection, reading medical records, delegation, and scheduling. Eventually, outlier detection has disclosed the game sessions with uncertain strategies and unstructured scheduling. [ABSTRACT FROM AUTHOR]

Digital educational games have evolved in recent years due to the need to support education and training focused on non-technical skills. Data gathered through interaction with the graphical user interface are explored and exploited to analyze the players' experience. Many researchers have pointed the importance of analysis of players’ in-game behavior, which can help to enhance the learning process, identify learners' strategies, and improve the effectiveness of the serious game. This study is devoted to the analysis of students' behavior in a simulation game called CLONE, which targets work scheduling, situation awareness, and decision-making. The students’ performance and their behavioral strategies are examined based on sequences analysis of players' in-game actions. Moreover, outlier detection is proposed as an instrument for obtaining information that might help better understand students’ behavior. The findings of the study show that such indicators as time spent on planning schedule, time spent on inspecting additional information, and intensity of delegation activity are significantly higher for successful games than for lost ones. The sequences analysis and clustering reveal students' prevailing in-game strategies, which mostly consist of inspection, reading medical records, delegation, and scheduling. Eventually, outlier detection discloses the game sessions with uncertain strategies and unstructured scheduling.

The large number of new services and applications and, in general, all our everyday activities resolve in data mass production: all these data can become a golden source of information that might be used to improve our lives, wellness and working days. (Interpretable) Machine Learning approaches, the use of which is increasingly ubiquitous in various settings, are definitely one of the most effective tools for retrieving and obtaining essential information from data. However, many challenges arise in order to effectively exploit them. In this paper, we analyze key scenarios in which large amounts of data and machine learning techniques can be used for social good: social network analytics for enhancing cultural heritage dissemination; game analytics to foster Computational Thinking in education; medical analytics to improve the quality of life of the elderly and reduce health care expenses; exploration of work datafication potential in improving the management of human resources (HRM). For the first two of the previously mentioned scenarios, we present new results related to previously published research, framing these results in a more general discussion over challenges arising when adopting machine learning techniques for social good. [ABSTRACT FROM AUTHOR]

Using data from a major commercial online game, Des-tiny, we track the development of player skill across time.From over 20,000 player record we identify 3475 playerswho have played on 50 or more days. Our focus is onhow variability in elements of play affect subsequent skilldevelopment. After validating the persistent influence ofdifferences in initial performance between players, wetest how practice spacing, social play, play mode vari-ability and a direct measure of game-world explorationaffect learning rate. These latter two factors do not af-fect learning rate. Players who space their practice morelearn faster, in line with our expectations, whereas play-ers who coordinate more with other players learn slower,which contradicts our initial hypothesis. We conclude thatnot all forms of practice variety expedite skill acquisition.Online game telemetry is a rich domain for exploring the-ories of optimal skill acquisition.

Artificial Intelligence (AI) and Machine Learning (ML) algorithms are increasingly being adopted to create and filter online digital content viewed by audiences from diverse demographics. From an early age, children grow into habitual use of online services but are usually unaware of how such algorithms operate, or even of their presence. Design decisions and biases inherent in the ML algorithms or in the datasets they are trained on shape the everyday digital lives of present and future generations. It is therefore important to disseminate a general understanding of AI and ML, and the ethical concerns associated with their use. As a response, the digital game ArtBot was designed and developed to teach fundamental principles about AI and ML, and to promote critical thinking about their functionality and shortcomings in everyday digital life. The game is intended as a learning tool in primary and secondary school classrooms. To assess the effectiveness of the ArtBot game as a learning experience we collected data from over 2,000 players across different platforms focusing on the degree of usage, interface efficiency, learners' performance and user experience. The quantitative usage data collected within the game was complemented by over 160 survey responses from teachers and students during early pilots of ArtBot. The evaluation analysis performed in this paper gauges the usability and usefulness of the game, and identifies areas of the game design which need improvement.

Games as a service is similar to software as a service, which provides players with game content on a continuous monetization model. Game revenue forecast is vital to game developers to make the right business decisions, such as determining the marketing budget, controlling the development cost, and setting up benchmarks for evaluating game publishing performance. How to make the revenue forecast and integrate it with the game publishing process is hard for small and medium-sized independent (indie) game developers. This includes all steps of the process, from forecasting to decision-making based on the results. This paper provides a data-driven method that uses the mobile game revenue forecast based on different time-series prediction models to drive the game publishing. We demonstrate how to use the data-driven method to guide an indie game studio to forecast revenue and then set the revenue forecast as the internal benchmark to drive game publishing. In practice, we involve a real game project from an indie game studio and provide guidance for one of their casual game projects. Then, based on the revenue forecast, we discuss how to set the revenue forecast as an internal benchmark and drive the actions for mobile game publishing. Finally, we make a conclusion on how our data-driven method can be used to drive mobile game publishing and also discuss future research work. [ABSTRACT FROM AUTHOR]

As a business model, the essence of games is to provide a service to satisfy the player experience. From a business perspective, development in the game industry has led to the application of Business Intelligence (BI) becoming more and more extensive. However, related research lacks systematic examination and precise classification. This paper provides a comprehensive literature review of BI used in the game industry, focusing primarily on game analytics. This research mainly studies and discusses five aspects. First, we explore game analytics aspects in the available literature based on the traditional game value chain. Second, we find out the main purposes of using analytics in the game industry. Third, we present the problems or challenges in the game area, which can be addressed by using game analytics. Fourth, we also list different algorithms that have been used in game analytics for prediction. Finally, we summarize the research areas that have already been covered in literature but need further development. Based on the categories established after the mapping and the review findings, we also discuss the limitations of game analytics and propose potential research points for future research. [ABSTRACT FROM AUTHOR]

Learning games are becoming popular among teachers as educational tools. However, despite all the game development quality processes (e.g., beta testing), there is no total assurance about the game design appropriateness to the students' cognitive skills until the games are used in the classroom. Furthermore, games designed specifically for Intellectual Disabled (ID) users are even harder to evaluate because of the communication issues that this type of players have. ID users' feedback about their learning experience is complex to obtain and not always fully reliable. To address this problem, we use an evidence-based approach for evaluating the game design of Downtown, A Subway Adventure, a game created to improve independent living in users with ID. In this paper we exemplify the whole process of applying Game Analytics techniques to gather actual users' gameplay interaction data in real settings for evaluating the design. Following this process, researchers were able to validate different game aspects (e.g., mechanics) and could also identify game flaws that may be difficult to detect using formative evaluation or other observational-based methods. Results showed that the proposed evidence-based approach using Game Analytics information is an effective way to evaluate both the game design and the implementation, especially in situations where other types of evaluations that require users' involvement are limited.

The large number of new services and applications and, in general, all our everyday activities resolve in data mass production: all these data can become a golden source of information that might be used to improve our lives, wellness and working days. (Interpretable) Machine Learning approaches, the use of which is increasingly ubiquitous in various settings, are definitely one of the most effective tools for retrieving and obtaining essential information from data. However, many challenges arise in order to effectively exploit them. In this paper, we analyze key scenarios in which large amounts of data and machine learning techniques can be used for social good: social network analytics for enhancing cultural heritage dissemination; game analytics to foster Computational Thinking in education; medical analytics to improve the quality of life of the elderly and reduce health care expenses; exploration of work datafication potential in improving the management of human resources (HRM). For the first two of the previously mentioned scenarios, we present new results related to previously published research, framing these results in a more general discussion over challenges arising when adopting machine learning techniques for social good.

The lostness measure, an implicit and unobtrusive measure originally designed for assessing the usability of hypertext systems, could be useful in Virtual Reality (VR) games where players need to find information to complete a task. VR locomotion systems with node-based movement mimic actions for exploration and browsing found in hypertext systems. For that reason, hypertext usability measures, such as “lostness” can be used to identify how disoriented a player is when completing tasks in an educational game by examining steps made by the player. An evaluation of two different lostness measures, global and local lostness, based on two different types of tasks, is described in a VR educational game using 13 college students between 14 and 18 years old in a first study and extended using 12 extra participants in a second study. Multiple Linear Regression analyses showed, in both studies, that local lostness, and not global lostness, had a significant effect on a post-game knowledge test. Therefore, we argued that local lostness was able to predict how well-participants would perform on a post-game knowledge test indicating how well they learned from the game. In-game experience aspects (engagement, cognitive interest, and presence) were also evaluated and, interestingly, it was also found that participants learned less when they felt more present in the game. We believe these two measures relate to cognitive overload, which is known to have an adverse effect on learning. Further research should investigate the lostness measure for use in an online adaptive game system and design the game system in such a way that the risk of cognitive overload is minimized when learning, resulting in higher retention of information.

Player motivation is a key research area within games research, with the aim of understanding how the motivation of players is related to their experience and behavior in the game. We present the results of a cross-sectional study with data from 750 players of League of Legends, a popular Multiplayer Online Battle Arena game. Based on the motivational regulations posited by Self-Determination Theory and Latent Profile Analysis, we identify four distinct motivational profiles, which differ with regards to player experience and, to a lesser extent, in-game behavior. While the more self-determined profiles “Intrinsic” and “Autonomous” report mainly positive experience-related outcomes, a considerable part of the player base does not. Players of the “Amotivated” and “External” profile derive less enjoyment, experience more negative affect and tension, and score lower on vitality, indicating game engagement that is potentially detrimental to players' well-being. With regards to game metrics, minor differences in the rate of assists in unranked matches and performance indicators were observed between profiles. This strengthens the notion that differences in experiences are not necessarily reflected in differences in behavioral game metrics. Our findings provide insights into the interplay of player motivation, experience, and in-game behavior, contributing to a more nuanced understanding of player-computer interaction.

The lostness measure, an implicit and unobtrusive measure originally designed for assessing the usability of hypertext systems, could be useful in Virtual Reality (VR) games where players need to find information to complete a task. VR locomotion systems with node-based movement mimic actions for exploration and browsing found in hypertext systems. For that reason, hypertext usability measures, such as "lostness" can be used to identify how disoriented a player is when completing tasks in an educational game by examining steps made by the player. An evaluation of two different lostness measures, global and local lostness, based on two different types of tasks, is described in a VR educational game using 13 college students between 14 and 18 years old in a first study and extended using 12 extra participants in a second study. Multiple Linear Regression analyses showed, in both studies, that local lostness, and not global lostness, had a significant effect on a post-game knowledge test. Therefore, we argued that local lostness was able to predict how well-participants would perform on a post-game knowledge test indicating how well they learned from the game. In-game experience aspects (engagement, cognitive interest, and presence) were also evaluated and, interestingly, it was also found that participants learned less when they felt more present in the game. We believe these two measures relate to cognitive overload, which is known to have an adverse effect on learning. Further research should investigate the lostness measure for use in an online adaptive game system and design the game system in such a way that the risk of cognitive overload is minimized when learning, resulting in higher retention of information. [ABSTRACT FROM AUTHOR]

The purpose of this case study is to demonstrate how to utilize machine learning approaches to analyze student process data for validating and informing digital game-based assessments (DGBAs) with an evidence-centered game design (ECgD). The first analysis was conducted to examine whether students' mastery of the overall skill required by the game can be well predicted by task-related behavioral features and if the selected key features map onto the evidence model of the ECgD. Specifically, we extracted 27 behavioral features as the indicators of students' gameplay activities from the evidence trace files and modelled them using a machine learning algorithm—support vector machine with recursive feature elimination—to identify the key features for prediction. The key features were in turn used to predict students' mastery of the overall skill. Results showed that students' retry attempts on two assessment tasks were found to be most influential for prediction with a moderate to high training and testing accuracy. The second analysis was conducted to examine whether the number of learning opportunities is sufficient for evaluating students' mastery of the overall skill as well as determine the optimal number of learning opportunities for evaluation. The approach of long short-term memory networks was used to model students' time-series behavioral features across multiple learning opportunities for predicting their acquisition of the overall skill. Results suggested that five learning opportunities were a good balance between evaluation accuracy and practical feasibility, and they were sufficient for evaluating students' mastery of the overall skill given the DGGA tasks. [ABSTRACT FROM AUTHOR]

The widespread adoption of the freemium business model together with the introduction of cost-efficient analytics tools have made the use of analytics pervasive in the game industry. While big data and analytics have drawn extensive scholarly attention, the research focusing particularly on game analytics is scant and largely descriptive. Thus, there is a need for research focusing on how game companies employ analytics. In this study, we analyze data collected through a set of in-depth interviews of small and medium-sized freemium game developers. We identify four main roles of game analytics: 1) sense-making device, 2) decision-support system, 3) communication tool, and 4) hygiene factor. We employ the attention-based view of the firm to discuss how these roles diverge and converge in terms of organizational attention. The study advances the research on the roles and business value of analytics in the game and software industry. [ABSTRACT FROM AUTHOR]

Player motivation is a key research area within games research, with the aim of understanding how the motivation of players is related to their experience and behavior in the game. We present the results of a cross-sectional study with data from 750 players of League of Legends , a popular Multiplayer Online Battle Arena game. Based on the motivational regulations posited by Self-Determination Theory and Latent Profile Analysis, we identify four distinct motivational profiles, which differ with regards to player experience and, to a lesser extent, in-game behavior. While the more self-determined profiles "Intrinsic" and "Autonomous" report mainly positive experience-related outcomes, a considerable part of the player base does not. Players of the "Amotivated" and "External" profile derive less enjoyment, experience more negative affect and tension, and score lower on vitality, indicating game engagement that is potentially detrimental to players' well-being. With regards to game metrics, minor differences in the rate of assists in unranked matches and performance indicators were observed between profiles. This strengthens the notion that differences in experiences are not necessarily reflected in differences in behavioral game metrics. Our findings provide insights into the interplay of player motivation, experience, and in-game behavior, contributing to a more nuanced understanding of player-computer interaction. [ABSTRACT FROM AUTHOR]

The COVID-19 pandemic has forced faculties including dental schools into a ‘new normal’, where the adoption of remote or distance learning is required to minimise the risk of infection. Synchronous learning historically was favoured due to the perceived advantage of ‘real time’ interactions between instructors and learners; these interactions are not always possible in asynchronous settings. However, serious games can overcome this limitation of asynchronous learning. This integrative review explores the literature on serious games in dental education, to construct a conceptual framework of their strengths in this pandemic. Following consideration of inclusion and exclusion criteria, 15 articles on 11 serious games designed for dental education were included in this review. Our investigation points to an increase in the use of serious games since 2018. The findings of the review support the use of serious games in dental education during the recent crisis. Key strengths include positive educational outcomes, enhanced engagement and motivation, interactive asynchronous distance learning, a safe learning environment, and the advantage of stealth assessment. Consequently, the ‘new normal’ in education appears to support a very promising future for serious games, particularly in dental education. A conceptual framework is proposed to inform further research across all education settings and timeframes.