Abstract
Emotionally adaptive games are one of the holy grails of modern affective game research. However, current state of the art affective games rely on static game adaptation mechanics that assume a fixed emotional reaction from players every time. Not only this, most commercial titles have no affective adaptation loop whatsoever and their design is based on game design optimizations via typical beta-testing procedures, which falls short of ideal both in the level design and long-term gameplay experience fronts. In this paper, we demonstrate a generalizable approach for building predictive models of players' emotional reactions across different games and game genres. We describe a physiological approach for modelling players' emotional reactions, which relies on features extracted from players' emotional responses to game events, which were collected and extrapolated through their physiological data during actual gameplay sessions. Based on the optimal feature sets found by three feature selection algorithms (best first, sequential feature selection and genetic search), the collected features are used to create computational models of players' emotional reactions on the arousal and valence dimensions of emotion, using several machine learning algorithms. We expect this approach will allow both a more objective and quicker prototyping for digital games, as well as foster a future generation of affective games capable of modelling players' affective profiles over time, thus adapting to their changing preferences and needs.

Abstract
Current approaches to game design improvements rely on gameplay testing, an iterative process following the test, try and fix pattern, relying on target audience feedback via standard questionnaires. Besides being a very time consuming phase, it is also highly subjective. In this paper, we demonstrate a generalizable approach for building predictive models of players' affective reactions across games and genres. Our aim is two-fold: 1) That game developers can use these models to more easily and accurately tune game parameters, allowing improved gaming experiences, and 2) That these models can be used as the basis for parameterisable and adaptive affective gaming. This paper describes our preliminary results regarding a novel, physiological-based method for emotional player profiling, which consists on the following three phases: (i) monitoring players' emotional states and game events, (ii) identifying player's emotional reactions to game events and (iii) individual and cluster-based modelling of players emotional reactions.

Abstract
Current approaches to game design improvements rely on time-consuming gameplay testing processes, which rely on highly subjective feedback from a target audience. In this paper, we propose a generalizable approach for building predictive models of players' emotional reactions across different games and game genres, as well as other forms of digital stimuli. Our input agnostic approach relies on the following steps: (a) collecting players' physiologically-inferred emotional states during actual gameplay sessions, (b) extrapolating the causal relations between changes in players' emotional states and recorded game events, and (c) building hierarchical cluster models of players' emotional reactions that can later be used to infer individual player models via fuzzy cluster membership vectors. We expect this work to benefit game designers by accelerating the affective playtesting process through the offline simulation of players' reactions to game design adaptations, as well as to contribute towards individually-tailored affective gaming.

Abstract
With the rising popularity of affective computing techniques, there have been several advances in the field of emotion recognition systems. However, despite the several advances in the field, these systems still face scenario adaptability and practical implementation issues. In light of these issues, we developed a nonspecific method for emotional state classification in interactive environments. The proposed method employs a two-layer classification process to detect Arousal and Valence (the emotion's hedonic component), based on four psychophysiological metrics: Skin Conductance, Heart Rate and Electromyography measured at the corrugator supercilii and zygomaticus major muscles. The first classification layer applies multiple regression models to correctly scale the aforementioned metrics across participants and experimental conditions, while also correlating them to the Arousal or Valence dimensions. The second layer then explores several machine learning techniques to merge the regression outputs into one final rating. The obtained results indicate we are able to classify Arousal and Valence independently from participant and experimental conditions with satisfactory accuracy (97% for Arousal and 91% for Valence).

Abstract
Exergames require obtaining or computing information regarding the players’ physical activity and context. Additionally, ensuring that the players are assigned challenges that are adequate to their physical ability, safe and adapted for the current context (both physical and spatial) is also important, as it can improve both the gaming experience and the outcomes of the exercise. However, the impact adaptivity has in the specific case of virtual reality exergames still has not been researched in depth. In this paper, we present a virtual reality exergame and an experimental design aiming to compare the players’ experience when playing both adaptive and regular versions of the game. © Springer International Publishing AG, part of Springer Nature 2018.

Abstract
Physical maps of a city or region are important pieces of geographical information for tourists and local citizens. Unfortunately the amount of information that can be presented on a piece of paper is limited. In order to extend the map information we propose an augmented reality (AR) system, ARTourMap, for additional information visualization and interaction. This system provides an abstraction layer to develop applications based on the concept of separated logic map tiles taking advantage of a multi-target system where several regions of the map trigger different superimposed graphics. This allows the map to be folded, to be partially occluded, and to have dematerialized information. To demonstrate the proposed system ARTourMap, three layers were developed: a location-based game with points of interest (POIs), a 3D building visualization and an historical map layer. © 2016 The Eurographics Association.