-
Presentation
Presentation
The success of a game often depends on the balance of a series of variables which can have a direct implication on its mechanical behavior. For example, competitive games rely on a balanced ecosystem between a wide variety of mechanics, which if left unchecked, can result in excessive exploitative behavior ruining the experience for the majority of players. This course will cover classical topics and consider new popular solutions that arise on a yearly basis which include, but are not limited to, bayesian inferences applied to game design, experience and behavior modeling for player classification, player prediction models, and physiological analysis applied to understanding player emotion and experience during play.
-
Class from course
Class from course
-
Degree | Semesters | ECTS
Degree | Semesters | ECTS
Master Degree | Semestral | 10
-
Year | Nature | Language
Year | Nature | Language
1 | Mandatory | Português
-
Code
Code
ULHT6838-25526
-
Prerequisites and corequisites
Prerequisites and corequisites
Not applicable
-
Professional Internship
Professional Internship
Não
-
Syllabus
Syllabus
CP1. The relationship between player and game from a psychophysiological perspective 1.1. The usage of videogames in science: Neuroscience, Psychology, Medicine and Rehabilitation CP2 Physiological analysis: 2.1. Electrocardiogram: Application, Signal and Processing 2.2. Electrodermal Activity: Application, Signal, and Processing 2.3. Electromyography: Application, Signal and Processing 2.4. Devices: Plux, Biopac, and others. 2.5. Sync between device and game CP3. Data collection 3.1. Scientific protocol design 3.2. Definition of study groups 3.3. Bias and precautions to be taken with participants 3.4. Declarations of Consent and Data Privacy. 3.5. The Pilot Study CP4 Applied Modeling 4.1. Algorithms Review: Bayesian Inference, Linear Regression, SVM and Neural Networks 4.2. Unsupervised Algorithms: K-Mean and EM 4.3. Dealing with temporal data: feature extraction 4.4. Optimize Hyperparameters
-
Objectives
Objectives
The Learning Objectives (LO) of this CU are: LO1. Understand the relationship between the player and game, and basic human computer interaction theory. LO2. Understand the concept of physiological data and human sensory stimulation LO3. Learn about the usability of physiological sensors: Electrocardiogram, Electrodermal Activity and Electromyogram. LO4. Understand the relationship between stimulus and physiological response and how to detect them in the data. LO5. Learn how to process physiological data through scripting languages (e.g. Python) LO6. Apply a data collection process for solving particular problems in the field of videogames. LO7. Understand what types of models can be constructed, what algorithms can be applied to different types of collected data. LO8. Build models capable of classifying player according to skill, emotions felt and archetypes LO9. Apply machine learning for studying players, while understanding its advantages and disadvantages.
-
Teaching methodologies
Teaching methodologies
Classes will be given within a theoretical and practical framework, and will consist of lectures (TM1), practical activities in class (TM2), practical project activities (TM3) and feedback sessions by the teacher (TM4). The assessment component will consist of developing ML models consistent with specific problems commonly acknowledged in the industry with each being concluded with an in-depth report. The assessment is carried out as follows: Several small assignments and theoretical exercises will be given during class to assess weekly topics, and 1 practical project (divided in two) that require students to develop, collect and analyse their own game experimental projects. Final mark = 90% project + 10% class assessment
-
References
References
Togelius, J. (2019). Playing smart: On games, intelligence, and artificial intelligence. MIT Press. Yannakakis, G. N., & Togelius, J. (2018). Artificial intelligence and games. Springer. Karpouzis, K., & Yannakakis, G. N. (2016). Emotion in Games. Cham: Springer. Fenner, M. (2019). Machine Learning with Python for Everyone. Pearson Education. James, G., Witten, D., Hastie, T. and Tibshirani, R. (2021). An Introduction to Statistical Learning. Springer.
-
Assessment
Assessment
The evaluation of this curricular unit is divided into small weekly assessment exercises and one large data collection project, divided in two different phases as detailed below.
Description
Ponderação
Weekly Assessment Exercises 10% Project 1: Scientific Methodology & Protocol Design
Design a scientific methodology and protocol to model one player-related variable for playtesting (e.g., stress, engagement, frustration, arousal).30%
Project 2: Experiment, Data Collection and Reporting
Run the experiment designed in Project 1, collect data, model the chosen variable, and report results.60%
-
Mobility
Mobility
No
