Entertainment Computing最新文献_第6页

Monetization mechanisms in gacha games: The behavioral triad of pricing strategies, pity systems, and belief of luck gacha游戏的盈利机制：定价策略、同情系统和运气信念的行为三元组合

IF 2.4 3区计算机科学 Q2 COMPUTER SCIENCE, CYBERNETICS

Entertainment Computing

Pub Date : 2025-09-01 DOI: 10.1016/j.entcom.2025.101044

Chang Ma, Jingbo Shao, Pengyu Li

As a dominant monetization mechanism in mobile gaming, gacha systems have raised significant concerns regarding their behavioral impacts due to the probabilistic nature of virtual item acquisition. While previous studies have drawn parallels between gacha engagement and gambling-related disorders, this research adopts a behavioral economics lens to investigate the determinants of user participation in this prevalent virtual economy model. Through a randomized controlled trial (N = 457), we systematically examine how pricing strategies (single-pull gacha cost) and pity systems (guaranteed prize mechanisms) interact to shape players’ intention to pay. Empirical evidence reveals that both pricing strategies of single gacha and pity systems significantly impact the spending intentions through cognitive reappraisal of perceived risk. Notably, individual differences in belief of luck patterns emerged as critical moderators. This study quantifies the economic interplay between system architecture and superstitious cognition, and provides evidence-based recommendations for gacha mechanisms design and regulatory interventions targeting compulsive spending patterns in digital environments.

作为手机游戏中的主要盈利机制，由于虚拟道具获取的概率性，gacha系统引起了人们对其行为影响的极大关注。虽然之前的研究将gacha粘性与赌博相关障碍进行了比较，但本研究采用了行为经济学的视角来调查这种流行的虚拟经济模式中用户参与度的决定因素。通过随机对照试验（N = 457），我们系统地研究了定价策略（单拉动gacha成本）和同情系统（保证奖励机制）如何相互作用，从而塑造玩家的付费意愿。经验证据表明，单个gacha和同情系统的定价策略都通过对感知风险的认知重新评估显著影响消费意愿。值得注意的是，个体对运气模式的信念差异成为关键的调节因素。本研究量化了系统架构与迷信认知之间的经济相互作用，并为gacha机制设计和针对数字环境中强迫性消费模式的监管干预提供了基于证据的建议。

{"title":"Monetization mechanisms in gacha games: The behavioral triad of pricing strategies, pity systems, and belief of luck","authors":"Chang Ma, Jingbo Shao, Pengyu Li","doi":"10.1016/j.entcom.2025.101044","DOIUrl":"10.1016/j.entcom.2025.101044","url":null,"abstract":"<div><div>As a dominant monetization mechanism in mobile gaming, gacha systems have raised significant concerns regarding their behavioral impacts due to the probabilistic nature of virtual item acquisition. While previous studies have drawn parallels between gacha engagement and gambling-related disorders, this research adopts a behavioral economics lens to investigate the determinants of user participation in this prevalent virtual economy model. Through a randomized controlled trial (N = 457), we systematically examine how pricing strategies (single-pull gacha cost) and pity systems (guaranteed prize mechanisms) interact to shape players’ intention to pay. Empirical evidence reveals that both pricing strategies of single gacha and pity systems significantly impact the spending intentions through cognitive reappraisal of perceived risk. Notably, individual differences in belief of luck patterns emerged as critical moderators. This study quantifies the economic interplay between system architecture and superstitious cognition, and provides evidence-based recommendations for gacha mechanisms design and regulatory interventions targeting compulsive spending patterns in digital environments.</div></div>","PeriodicalId":55997,"journal":{"name":"Entertainment Computing","volume":"55 ","pages":"Article 101044"},"PeriodicalIF":2.4,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145415067","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Lock the look: Recommending trendy looks for fashion products using natural language processing 锁定外观：使用自然语言处理为时尚产品推荐时尚外观

IF 2.4 3区计算机科学 Q2 COMPUTER SCIENCE, CYBERNETICS

Entertainment Computing

Pub Date : 2025-09-01 DOI: 10.1016/j.entcom.2025.101049

Manjarini Mallik , Tushti Thakur , Chandreyee Chowdhury

The recreation of looks established by favorite movie characters or fashion icons is a popular trend in this decade. It is difficult to find out the dresses and accessories required to develop that look as current product recommendations are mostly based on history of users’ choices. There exists computer vision-based solutions that check image-wise similarities between the desired looks and available fashion products from e-commerce stores. However, this is a resource hungry complex process as plenty of product images would be analyzed. In this work an NLP-based lightweight look recommendation system is proposed. In the proposed approach, multiple text descriptions of trendy looks are collected from different websites to build the training dataset. A subset of two benchmark datasets (Myntra Products Dataset and Ajio Products Dataset) have been used for recommendation. Using the bag of words technique, text datasets are embedded, and a set of looks is recommended for each product. The system is validated using Cosine similarity and Cohen’s kappa metrics. Products in the test dataset have been mapped to their 1st and 2nd highest recommended looks with positive scores. We observed a minimum score of 0.6 and 0.2 for Cosine similarity and Cohen’s kappa respectively, representing appreciable performance.

在这十年里，模仿最喜欢的电影角色或时尚偶像的造型是一种流行趋势。由于目前的产品推荐大多是基于用户的选择历史，因此很难找到开发这种外观所需的服装和配饰。目前存在基于计算机视觉的解决方案，可以检查期望的外观与电子商务商店中可用的时尚产品之间的图像相似性。然而，这是一个需要大量资源的复杂过程，因为需要分析大量的产品图像。本文提出了一种基于nlp的轻量级外观推荐系统。在提出的方法中，从不同的网站收集时尚外观的多个文本描述来构建训练数据集。两个基准数据集（Myntra产品数据集和Ajio产品数据集）的子集已被用于推荐。使用词包技术，嵌入文本数据集，并为每个产品推荐一组外观。该系统使用余弦相似度和科恩的kappa指标进行验证。测试数据集中的产品被映射到第一和第二高的推荐外观，并获得正分数。我们观察到余弦相似度和科恩kappa的最低得分分别为0.6和0.2，代表了可观的性能。

{"title":"Lock the look: Recommending trendy looks for fashion products using natural language processing","authors":"Manjarini Mallik , Tushti Thakur , Chandreyee Chowdhury","doi":"10.1016/j.entcom.2025.101049","DOIUrl":"10.1016/j.entcom.2025.101049","url":null,"abstract":"<div><div>The recreation of looks established by favorite movie characters or fashion icons is a popular trend in this decade. It is difficult to find out the dresses and accessories required to develop that look as current product recommendations are mostly based on history of users’ choices. There exists computer vision-based solutions that check image-wise similarities between the desired looks and available fashion products from e-commerce stores. However, this is a resource hungry complex process as plenty of product images would be analyzed. In this work an NLP-based lightweight look recommendation system is proposed. In the proposed approach, multiple text descriptions of trendy looks are collected from different websites to build the training dataset. A subset of two benchmark datasets (Myntra Products Dataset and Ajio Products Dataset) have been used for recommendation. Using the bag of words technique, text datasets are embedded, and a set of looks is recommended for each product. The system is validated using Cosine similarity and Cohen’s kappa metrics. Products in the test dataset have been mapped to their 1st and 2nd highest recommended looks with positive scores. We observed a minimum score of 0.6 and 0.2 for Cosine similarity and Cohen’s kappa respectively, representing appreciable performance.</div></div>","PeriodicalId":55997,"journal":{"name":"Entertainment Computing","volume":"55 ","pages":"Article 101049"},"PeriodicalIF":2.4,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145465462","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Hateful tweet detection using a BiLSTM-BiGRU: An ensemble perspective 使用BiLSTM-BiGRU的仇恨推文检测：一个整体的视角

IF 2.4 3区计算机科学 Q2 COMPUTER SCIENCE, CYBERNETICS

Entertainment Computing

Pub Date : 2025-09-01 DOI: 10.1016/j.entcom.2025.101019

Imandi Tejaswini , Venkata Gayathri Ganivada , Appala Srinuvasu Muttipati

Social media hate speech is an emerging issue, and there is a need to create automatic systems to identify and mitigate its effects. The rapid expansion of social media platforms, especially Twitter, has facilitated the dissemination of hate speech, presenting a major challenge for online communities. Such speech can have severe social and psychological consequences, including inciting violence, promoting extremism, and affecting mental health. Thus, it is essential to manage hateful content on Twitter. This paper presents an ensemble deep learning model that combines BiLSTM and BiGRU to enhance prediction accuracy and robustness. The model achieved 98.56% accuracy rate and demonstrated better generalization than existing methods, proving its effectiveness in identifying hate speech with fewer false positives. This paper offers a powerful tool for detecting and preventing harmful online behavior, contributing to a safer and more inclusive digital space.

社交媒体上的仇恨言论是一个新出现的问题，有必要创建自动系统来识别和减轻其影响。社交媒体平台的迅速扩张，尤其是推特，促进了仇恨言论的传播，给在线社区带来了重大挑战。此类言论可能产生严重的社会和心理后果，包括煽动暴力、宣扬极端主义和影响心理健康。因此，管理Twitter上的仇恨内容至关重要。本文提出了一种结合BiLSTM和BiGRU的集成深度学习模型，以提高预测精度和鲁棒性。该模型的准确率达到了98.56%，并且比现有方法具有更好的泛化性，证明了其在识别仇恨言论方面的有效性，并且假阳性较少。本文提供了一种检测和预防有害在线行为的强大工具，有助于建立一个更安全、更包容的数字空间。

引用次数: 0

Can LLMs predict the success of Turkish TV series from their first episodes? 法学硕士能否从土耳其电视剧的第一集就预测其成功？

IF 2.4 3区计算机科学 Q2 COMPUTER SCIENCE, CYBERNETICS

Entertainment Computing

Pub Date : 2025-09-01 DOI: 10.1016/j.entcom.2025.101052

Firat Ismailoglu

Turkey is the third largest exporter of TV series worldwide. However, half of these series are cancelled early leading to economic and social consequences. In this study, we explore whether the success of these series can be predicted from the scripts of their first episodes using LLMs. We built a dataset of first-episode scripts from recently aired Turkish series and trained LLM-based models on it. The main challenge we faced is that these scripts are very long, making them unsuitable for standard BERT models. This led to one of the key contributions of our study, as there is currently no research that specifically focuses on handling long Turkish texts. We pretrained a BigBird model from scratch for Turkish and fine-tuned it for our task. We also developed a Hierarchical Attention Network (HAN) model capable of processing long Turkish texts. While predicting the exact number of episodes is difficult, both HAN and BigBird achieve strong performance in binary classification setup, distinguishing successful series from unsuccessful ones. Additionally, we investigate whether audience preferences in Turkey have changed over time by testing our models on some iconic older Turkish series to see if they would still be classified as successful by today’s standards.

土耳其是世界第三大电视剧出口国。然而，这些节目中有一半被提前取消，导致经济和社会后果。在这项研究中，我们探讨了这些电视剧的成功是否可以用llm来预测它们第一集的剧本。我们建立了最近播出的土耳其电视剧的第一集脚本数据集，并在其上训练了基于法学硕士的模型。我们面临的主要挑战是这些脚本非常长，使得它们不适合标准BERT模型。这导致了我们研究的关键贡献之一，因为目前没有研究专门关注处理长土耳其文本。我们从零开始为土耳其语预训练了一个BigBird模型，并对其进行了微调。我们还开发了一个能够处理长土耳其文本的分层注意网络（HAN）模型。虽然预测剧集的确切数量很困难，但HAN和BigBird在二元分类设置中都取得了很强的性能，可以区分成功的剧集和不成功的剧集。此外，我们通过在一些标志性的土耳其老系列上测试我们的模型来调查土耳其观众的偏好是否随着时间的推移而改变，看看它们是否仍然按照今天的标准被归类为成功。

{"title":"Can LLMs predict the success of Turkish TV series from their first episodes?","authors":"Firat Ismailoglu","doi":"10.1016/j.entcom.2025.101052","DOIUrl":"10.1016/j.entcom.2025.101052","url":null,"abstract":"<div><div>Turkey is the third largest exporter of TV series worldwide. However, half of these series are cancelled early leading to economic and social consequences. In this study, we explore whether the success of these series can be predicted from the scripts of their first episodes using LLMs. We built a dataset of first-episode scripts from recently aired Turkish series and trained LLM-based models on it. The main challenge we faced is that these scripts are very long, making them unsuitable for standard BERT models. This led to one of the key contributions of our study, as there is currently no research that specifically focuses on handling long Turkish texts. We pretrained a BigBird model from scratch for Turkish and fine-tuned it for our task. We also developed a Hierarchical Attention Network (HAN) model capable of processing long Turkish texts. While predicting the exact number of episodes is difficult, both HAN and BigBird achieve strong performance in binary classification setup, distinguishing successful series from unsuccessful ones. Additionally, we investigate whether audience preferences in Turkey have changed over time by testing our models on some iconic older Turkish series to see if they would still be classified as successful by today’s standards.</div></div>","PeriodicalId":55997,"journal":{"name":"Entertainment Computing","volume":"55 ","pages":"Article 101052"},"PeriodicalIF":2.4,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145519277","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Benchmarking reinforcement learning algorithms in first-person shooter games using VizDoom 使用VizDoom测试第一人称射击游戏中的强化学习算法

IF 2.4 3区计算机科学 Q2 COMPUTER SCIENCE, CYBERNETICS

Entertainment Computing

Pub Date : 2025-09-01 DOI: 10.1016/j.entcom.2025.101031

Adil Khan , Aamir Aqeel

Computer games are considered one of the best test beds for evaluating artificial intelligence algorithms, as it is a well-known practice before applying the algorithms in the real world, such as the robotics industry. A machine learning technique, known as reinforcement learning, utilizes positive and negative rewards to guide an artificial intelligence agent as it learns new tactics and strategies. This study compares four reinforcement learning algorithms: Dueling Double Deep Q-Network (Dueling DDQN), Advantage Actor-Critic (A2C), LSTM-Based Advantage Actor-Critic (A2C LSTM), and REINFORCE. The game artificial intelligence (Game AI) based platform VizDoom evaluates and compares these reinforcement learning algorithms. VizDoom is based on the first-person shooter (FPS) video game Doom, which has had a significant influence on artificial intelligence. The results are compared, and, in most cases, Dueling DDQN outperformed all other algorithms in all chosen scenarios. However, in contrast, the A2C performed well for the kills metric in the defending the center scenario only. Finally, the proposed work’s analysis, implications, and limitations are presented, along with the potential future directions for research.

电脑游戏被认为是评估人工智能算法的最佳测试平台之一，因为在将算法应用于现实世界（如机器人行业）之前，它是一种众所周知的实践。一种被称为强化学习的机器学习技术，利用积极和消极的奖励来指导人工智能代理学习新的战术和策略。本研究比较了四种强化学习算法：Dueling Double Deep Q-Network （Dueling DDQN）、Advantage Actor-Critic （A2C）、基于LSTM的Advantage Actor-Critic （A2C LSTM）和REINFORCE。基于游戏人工智能（game AI）的平台VizDoom评估并比较了这些强化学习算法。VizDoom是基于第一人称射击游戏《毁灭战士》开发的，这款游戏对人工智能产生了重大影响。对结果进行比较，在大多数情况下，Dueling DDQN在所有选择的场景中都优于所有其他算法。然而，相比之下，A2C只在防守中路的情况下表现得很好。最后，提出了本研究的分析、意义和局限性，以及未来研究的潜在方向。

{"title":"Benchmarking reinforcement learning algorithms in first-person shooter games using VizDoom","authors":"Adil Khan , Aamir Aqeel","doi":"10.1016/j.entcom.2025.101031","DOIUrl":"10.1016/j.entcom.2025.101031","url":null,"abstract":"<div><div>Computer games are considered one of the best test beds for evaluating artificial intelligence algorithms, as it is a well-known practice before applying the algorithms in the real world, such as the robotics industry. A machine learning technique, known as reinforcement learning, utilizes positive and negative rewards to guide an artificial intelligence agent as it learns new tactics and strategies. This study compares four reinforcement learning algorithms: Dueling Double Deep Q-Network (Dueling DDQN), Advantage Actor-Critic (A2C), LSTM-Based Advantage Actor-Critic (A2C LSTM), and REINFORCE. The game artificial intelligence (Game AI) based platform VizDoom evaluates and compares these reinforcement learning algorithms. VizDoom is based on the first-person shooter (FPS) video game Doom, which has had a significant influence on artificial intelligence. The results are compared, and, in most cases, Dueling DDQN outperformed all other algorithms in all chosen scenarios. However, in contrast, the A2C performed well for the kills metric in the defending the center scenario only. Finally, the proposed work’s analysis, implications, and limitations are presented, along with the potential future directions for research.</div></div>","PeriodicalId":55997,"journal":{"name":"Entertainment Computing","volume":"55 ","pages":"Article 101031"},"PeriodicalIF":2.4,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145265939","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Solutions for Dynamic Difficulty Adjustment in digital games: A Systematic Literature Review 数字游戏中动态难度调整的解决方案：系统文献综述

IF 2.4 3区计算机科学 Q2 COMPUTER SCIENCE, CYBERNETICS

Entertainment Computing

Pub Date : 2025-09-01 DOI: 10.1016/j.entcom.2025.101041

Carlos Henrique R. Souza, Daniela F. Nascimento, Luciana O. Berretta, Sergio T. Carvalho

Dynamic Difficulty Adjustment (DDA) is an important aspect of game design aimed at balancing the difficulty level to enhance player experience and prevent frustration or game abandonment. This paper presents a Systematic Literature Review (SLR) that focuses on the implementation of DDA techniques in digital games. The objective of this study was to identify and analyze DDA solutions (models, methods, and/or frameworks) for digital games in the ongoing research agenda. Of the 547 studies found in four bibliographic databases (ACM, IEEE, Scopus, and Web of Science), 34 were selected. The results revealed a diversity of approaches, mainly involving Artificial Intelligence techniques (50%) and heuristics/parameters manipulation (47%), among other possibilities. In addition to reaffirming the open research problems described in the literature, the need for further research on generalizable, flexible, and modularized approaches is highlighted, allowing the integration of various DDA strategies while minimizing the disadvantages of each strategy and ensuring good results. In this sense, the conducted review brings solution directions based on the exchange of knowledge from self-adaptive systems.

动态难度调整（DDA）是游戏设计的一个重要方面，旨在平衡难度等级，增强玩家体验，防止受挫或放弃游戏。本文是一篇关于DDA技术在数字游戏中的应用的系统性文献综述（SLR）。这项研究的目的是在正在进行的研究议程中确定和分析数字游戏的DDA解决方案（模型、方法和/或框架）。在四个书目数据库（ACM、IEEE、Scopus和Web Of Science）中发现的547项研究中，有34项被选中。结果显示了多种方法，主要涉及人工智能技术（50%）和启发式/参数操作（47%），以及其他可能性。除了重申文献中描述的开放性研究问题外，还强调了对可推广、灵活和模块化方法的进一步研究的必要性，以允许各种DDA策略的集成，同时最大限度地减少每种策略的缺点并确保良好的结果。从这个意义上说，进行的审查带来了基于自适应系统的知识交换的解决方案方向。

{"title":"Solutions for Dynamic Difficulty Adjustment in digital games: A Systematic Literature Review","authors":"Carlos Henrique R. Souza, Daniela F. Nascimento, Luciana O. Berretta, Sergio T. Carvalho","doi":"10.1016/j.entcom.2025.101041","DOIUrl":"10.1016/j.entcom.2025.101041","url":null,"abstract":"<div><div>Dynamic Difficulty Adjustment (DDA) is an important aspect of game design aimed at balancing the difficulty level to enhance player experience and prevent frustration or game abandonment. This paper presents a Systematic Literature Review (SLR) that focuses on the implementation of DDA techniques in digital games. The objective of this study was to identify and analyze DDA solutions (models, methods, and/or frameworks) for digital games in the ongoing research agenda. Of the 547 studies found in four bibliographic databases (ACM, IEEE, Scopus, and Web of Science), 34 were selected. The results revealed a diversity of approaches, mainly involving Artificial Intelligence techniques (50%) and heuristics/parameters manipulation (47%), among other possibilities. In addition to reaffirming the open research problems described in the literature, the need for further research on generalizable, flexible, and modularized approaches is highlighted, allowing the integration of various DDA strategies while minimizing the disadvantages of each strategy and ensuring good results. In this sense, the conducted review brings solution directions based on the exchange of knowledge from self-adaptive systems.</div></div>","PeriodicalId":55997,"journal":{"name":"Entertainment Computing","volume":"55 ","pages":"Article 101041"},"PeriodicalIF":2.4,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145361389","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Are you a Hunter or Camper? A phase-based framework for real-time survival prediction in battle royale using FightScore and ActiveScore 你是猎人还是露营者？一个基于阶段的框架，用于在大逃杀中使用战斗得分和ActiveScore进行实时生存预测

IF 2.4 3区计算机科学 Q2 COMPUTER SCIENCE, CYBERNETICS

Entertainment Computing

Pub Date : 2025-09-01 DOI: 10.1016/j.entcom.2025.101040

Chaeyeon Sagong , Ji Young Woo , Huy Kang Kim

The battle royale genre game’s primary goal is survival. Similarly, the goal of the PlayerUnknown’s Battlegrounds (PUBG), a representative battle royale game, is to be the last man standing in a shrinking zone. Thus, game strategy for survival is the most important part in winning. The game play progress is divided into distinct game phases with Aggressiveness and Activeness as key attributes. Players change their strategy by the game phase. To capture this dynamic, we propose introducing scores. In this study, we analyze phase-specific survival strategies in PUBG from the perspectives of Aggressiveness and Activeness, and predicts player survival in real-time. Using in-game features that reflect Aggressiveness and Activeness, we calculate scores, “FightScore” and “ActiveScore”, to cluster players into four strategic types: Hunter, Sniper, Traveler, and Camper. Unlike existing methods that rely on post-game statistics, we use phase-specific data to derive strategies to predict survival and optimize strategies for the subsequent phases. We evaluate the effectiveness of three machine learning models, Random Forest, Logistic Regression, and XGBoost, to predict survival and win. We focus on the relationship between phase-specific strategy and survival in PUBG, which can be informative for players, e-sports analysts, and broadcasters.

大逃杀类型游戏的主要目标是生存。同样，《绝地求生》（PlayerUnknown’s Battlegrounds， PUBG）这款典型的大逃杀游戏的目标是成为一个不断缩小的区域中的最后一个人。因此，游戏生存策略是获胜的最重要部分。游戏进程被划分为不同的游戏阶段，并以攻击性和主动性作为关键属性。玩家在游戏阶段会改变策略。为了捕捉这种动态，我们建议引入分数。本研究从进攻性和主动性的角度分析了《绝地求生》中特定阶段的生存策略，并实时预测了玩家的生存。我们使用反映侵略性和主动性的游戏功能，计算得分，“战斗得分”和“活动得分”，将玩家分为四种战略类型：猎人、狙击手、旅行者和露营者。与依赖赛后统计的现有方法不同，我们使用特定阶段的数据来推导预测生存和优化后续阶段策略的策略。我们评估了随机森林、逻辑回归和XGBoost三种机器学习模型的有效性，以预测生存和获胜。我们专注于《绝地求生》中特定阶段策略与生存之间的关系，这可以为玩家，电子竞技分析师和广播公司提供信息。

{"title":"Are you a Hunter or Camper? A phase-based framework for real-time survival prediction in battle royale using FightScore and ActiveScore","authors":"Chaeyeon Sagong , Ji Young Woo , Huy Kang Kim","doi":"10.1016/j.entcom.2025.101040","DOIUrl":"10.1016/j.entcom.2025.101040","url":null,"abstract":"<div><div>The battle royale genre game’s primary goal is survival. Similarly, the goal of the PlayerUnknown’s Battlegrounds (PUBG), a representative battle royale game, is to be the last man standing in a shrinking zone. Thus, game strategy for survival is the most important part in winning. The game play progress is divided into distinct game phases with <strong>Aggressiveness</strong> and <strong>Activeness</strong> as key attributes. Players change their strategy by the game phase. To capture this dynamic, we propose introducing scores. In this study, we analyze phase-specific survival strategies in PUBG from the perspectives of Aggressiveness and Activeness, and predicts player survival in real-time. Using in-game features that reflect Aggressiveness and Activeness, we calculate scores, <strong>“FightScore”</strong> and <strong>“ActiveScore”</strong>, to cluster players into four strategic types: Hunter, Sniper, Traveler, and Camper. Unlike existing methods that rely on post-game statistics, we use phase-specific data to derive strategies to predict survival and optimize strategies for the subsequent phases. We evaluate the effectiveness of three machine learning models, Random Forest, Logistic Regression, and XGBoost, to predict survival and win. We focus on the relationship between phase-specific strategy and survival in PUBG, which can be informative for players, e-sports analysts, and broadcasters.</div></div>","PeriodicalId":55997,"journal":{"name":"Entertainment Computing","volume":"55 ","pages":"Article 101040"},"PeriodicalIF":2.4,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145361393","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

“In esports teamwork ranks way higher than anything else” : An exploration of esports training perceptions “在电子竞技中，团队合作比其他任何事情都要重要”：电子竞技训练观念的探索

IF 2.4 3区计算机科学 Q2 COMPUTER SCIENCE, CYBERNETICS

Entertainment Computing

Pub Date : 2025-09-01 DOI: 10.1016/j.entcom.2025.101045

Danielle K. Langlois , Simone Kriglstein

Esports has been a growing area of study in recent years. We conducted two qualitatively focused survey studies to further understand the landscape of esports training. Specifically, via research questions related to coaching and training programs, types of training, gaps in training routines, how different people perceive training, and how important various training categories are perceived. The first was a pilot survey exploring existing esports training, the second was a more in-depth exploration of esports training and how fans and specialists perceive the field. We found that fans and specialists generally agreed on many aspects of esports training. For example, teamwork was widely agreed upon as a crucial aspect of training. However, specialists seemed more focused on mental and physical health than fans. Further, three general domains emerged as overarching trends worthy of focus: technical skills, teamwork, and health support (both physical and mental). These domains align well with prior researchers’ findings. Our results also support the notion that greater organization and focus should be placed on teamwork and mental health support, as those areas are perceived to be under served. We also found a strong undercurrent of a lack of access (primarily due to physical location) and a desire for more guidance in the training process. Therefore, future efforts should be made to make esports training resources, particularly mental health resources, more well-known and accessible.

近年来，电子竞技一直是一个不断发展的研究领域。我们进行了两项定性调查研究，以进一步了解电子竞技培训的现状。具体来说，通过研究与教练和培训计划、培训类型、培训程序中的差距、不同的人如何看待培训以及各种培训类别的重要性有关的问题。第一个是对现有电子竞技培训的试点调查，第二个是对电子竞技培训以及粉丝和专家对该领域的看法进行更深入的探索。我们发现，粉丝和专家对电子竞技培训的许多方面普遍持一致意见。例如，团队合作被广泛认为是培训的一个重要方面。然而，专家们似乎比粉丝们更关注心理和身体健康。此外，三个普遍领域成为值得关注的总体趋势：技术技能、团队合作和健康支持（身体和精神）。这些领域与先前研究人员的发现非常吻合。我们的研究结果也支持这样一种观点，即应该把更多的组织和重点放在团队合作和心理健康支持上，因为这些领域被认为是服务不足的。我们还发现了一股强大的暗流，即缺乏访问权限（主要是由于物理位置），并且希望在培训过程中获得更多指导。因此，未来应该努力使电子竞技培训资源，特别是心理健康资源，更加知名和可及。

{"title":"“In esports teamwork ranks way higher than anything else” : An exploration of esports training perceptions","authors":"Danielle K. Langlois , Simone Kriglstein","doi":"10.1016/j.entcom.2025.101045","DOIUrl":"10.1016/j.entcom.2025.101045","url":null,"abstract":"<div><div>Esports has been a growing area of study in recent years. We conducted two qualitatively focused survey studies to further understand the landscape of esports training. Specifically, via research questions related to coaching and training programs, types of training, gaps in training routines, how different people perceive training, and how important various training categories are perceived. The first was a pilot survey exploring existing esports training, the second was a more in-depth exploration of esports training and how fans and specialists perceive the field. We found that fans and specialists generally agreed on many aspects of esports training. For example, teamwork was widely agreed upon as a crucial aspect of training. However, specialists seemed more focused on mental and physical health than fans. Further, three general domains emerged as overarching trends worthy of focus: technical skills, teamwork, and health support (both physical and mental). These domains align well with prior researchers’ findings. Our results also support the notion that greater organization and focus should be placed on teamwork and mental health support, as those areas are perceived to be under served. We also found a strong undercurrent of a lack of access (primarily due to physical location) and a desire for more guidance in the training process. Therefore, future efforts should be made to make esports training resources, particularly mental health resources, more well-known and accessible.</div></div>","PeriodicalId":55997,"journal":{"name":"Entertainment Computing","volume":"55 ","pages":"Article 101045"},"PeriodicalIF":2.4,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145361447","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Enhancing cyberbullying identification with ELECTRA-BiLSTM: A hybrid approach for improved contextual and sequential understanding 利用ELECTRA-BiLSTM增强网络欺凌识别：一种改善上下文和顺序理解的混合方法

IF 2.4 3区计算机科学 Q2 COMPUTER SCIENCE, CYBERNETICS

Entertainment Computing

Pub Date : 2025-09-01 DOI: 10.1016/j.entcom.2025.101056

Shalini Agrahari, Arvind Kumar Tiwari

Social media has revolutionized how we connect, fostering communities based on shared interests worldwide. However, it also opens the door to cyberbullying, a serious concern in today’s digital age. Unlike traditional bullying, cyberbullying happens online, making it harder to detect and prevent. In this paper, we combine the Efficiently Learning an Encoder that Classifies Token Replacements Accurately (ELECTRA) model with Bidirectional Long Short-Term Memory (BiLSTM), to better spot cyberbullying, improving the model’s ability to understand context and sequential patterns in text. Preprocessing has included tokenization, normalization and data cleaning to ensure consistent input quality. Using a dataset of approximately 48,000 tweets across six categories, the proposed model achieved 88.38% accuracy, outperforming traditional and existing models, highlighting its potential to enhance online safety.

社交媒体彻底改变了我们的联系方式，培育了基于全球共同兴趣的社区。然而，它也为网络欺凌打开了大门，这是当今数字时代的一个严重问题。与传统的欺凌不同，网络欺凌发生在网上，这使得它更难被发现和预防。在本文中，我们将高效学习编码器准确分类Token替换（ELECTRA）模型与双向长短期记忆（BiLSTM）模型相结合，以更好地发现网络欺凌，提高模型理解文本中上下文和顺序模式的能力。预处理包括标记化、规范化和数据清理，以确保一致的输入质量。使用六类约48,000条推文的数据集，所提出的模型达到了88.38%的准确率，优于传统和现有的模型，突出了其提高在线安全的潜力。

引用次数: 0

The predictive role of cognitive flexibility in 21st century skills among esports players 认知灵活性在电子竞技选手21世纪技能中的预测作用

IF 2.4 3区计算机科学 Q2 COMPUTER SCIENCE, CYBERNETICS

Entertainment Computing

Pub Date : 2025-09-01 DOI: 10.1016/j.entcom.2025.101042

Alp Kaan Kilci , Zekeriya Göktaş

This paper presents the role of cognitive flexibility in predicting 21st century skills among esports players. Based on survey data collected from 591 players of esports games selected from among the ten most widely played by worldwide, the study explores how cognitive flexibility relates to key competencies such as digital literacy, innovation, leadership, and career consciousness. The research is guided by two research questions: (1) To what extent does cognitive flexibility predict 21st century skill development in esports players? and (2) Does the frequency of gaming influence this prediction? The results indicate that cognitive flexibility is a significant predictor of information technology literacy, entrepreneurial thinking, and leadership, particularly among those with higher weekly gaming frequency. However, no meaningful association was found between cognitive flexibility and critical thinking/problem-solving skills, suggesting that team-based gameplay dynamics may limit individual-level cognitive reflection. The findings highlight esports as a potent environment for developing transferable competencies and emphasize the importance of cognitive flexibility as both an outcome and enabler of skill acquisition in digital contexts. An integrative framework is proposed for understanding how gameplay experiences translate into real-world readiness. Future research should focus on designing targeted interventions to enhance cognitive flexibility within competitive gaming settings.

本文介绍了认知灵活性在预测电子竞技选手21世纪技能中的作用。基于从全球最受欢迎的十大电子竞技游戏中挑选的591名玩家的调查数据，该研究探讨了认知灵活性与数字素养、创新、领导力和职业意识等关键能力之间的关系。本研究以两个研究问题为指导：(1)认知灵活性在多大程度上预测电子竞技选手21世纪的技能发展？(2)游戏频率是否会影响这一预测？结果表明，认知灵活性是信息技术素养、创业思维和领导力的重要预测因素，特别是在每周游戏频率较高的人群中。然而，在认知灵活性和批判性思维/解决问题能力之间没有发现有意义的联系，这表明基于团队的游戏玩法动态可能会限制个人层面的认知反射。研究结果强调了电子竞技是培养可转移能力的有力环境，并强调了认知灵活性作为数字环境中技能习得的结果和推动者的重要性。本文提出了一个综合框架，用于理解游戏体验如何转化为现实世界的准备。未来的研究应该集中于设计有针对性的干预措施，以增强竞争性游戏环境中的认知灵活性。

{"title":"The predictive role of cognitive flexibility in 21st century skills among esports players","authors":"Alp Kaan Kilci , Zekeriya Göktaş","doi":"10.1016/j.entcom.2025.101042","DOIUrl":"10.1016/j.entcom.2025.101042","url":null,"abstract":"<div><div>This paper presents the role of cognitive flexibility in predicting 21st century skills among esports players. Based on survey data collected from 591 players of esports games selected from among the ten most widely played by worldwide, the study explores how cognitive flexibility relates to key competencies such as digital literacy, innovation, leadership, and career consciousness. The research is guided by two research questions: (1) <em>To what extent does cognitive flexibility predict 21st century skill development in esports players?</em> and (2) <em>Does the frequency of gaming influence this prediction?</em> The results indicate that cognitive flexibility is a significant predictor of information technology literacy, entrepreneurial thinking, and leadership, particularly among those with higher weekly gaming frequency. However, no meaningful association was found between cognitive flexibility and critical thinking/problem-solving skills, suggesting that team-based gameplay dynamics may limit individual-level cognitive reflection. The findings highlight esports as a potent environment for developing transferable competencies and emphasize the importance of cognitive flexibility as both an outcome and enabler of skill acquisition in digital contexts. An integrative framework is proposed for understanding how gameplay experiences translate into real-world readiness. Future research should focus on designing targeted interventions to enhance cognitive flexibility within competitive gaming settings.</div></div>","PeriodicalId":55997,"journal":{"name":"Entertainment Computing","volume":"55 ","pages":"Article 101042"},"PeriodicalIF":2.4,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145361391","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0