Knowledge discovery from data and knowledge acquisition from experts are steps of paramount importance when designing cognitive systems. The literature discusses extensively on the issues related to current knowledge acquisition techniques. In this doctoral work we explore the use of gaming approaches as a knowledge acquisition tools, capitalising on aspects such as engagement, ease of use and ability to access tacit knowledge. More specifically, we explore the use of analytical games for this purpose. Analytical game for decision making is not a new class of games, but rather a set of platform independent simulation games, designed not for entertainment, whose main purpose is research on decision-making, either in its complete dynamic cycle or a portion of it (i.e. Situational Awareness). Moreover, the work focuses on the use of analytical games as knowledge acquisition tools. To this end, the Knowledge Acquisition Analytical Game (K2AG) method is introduced. K2AG is an innovative game framework for supporting the knowledge acquisition task. The framework introduced in this doctoral work was born as a generalisation of the Reliability Game, which on turn was inspired by the Risk Game. More specifically, K2AGs aim at collecting information and knowledge to be used in the design of cognitive systems and their algorithms. The two main aspects that characterise those games are the use of knowledge cards to render information and meta-information to the players and the use of an innovative data gathering method that takes advantage of geometrical features of simple shapes (e.g. a triangle) to easily collect players’ beliefs. These beliefs can be mapped to subjective probabilities or masses (in evidence theory framework) and used for algorithm design purposes. However, K2AGs might use also different means of conveying information to the players and to collect data. Part of the work has been devoted to a detailed articulation of the design cycle of K2AGs. More specifically, van der Zee’s simulation gaming design framework has been extended in order to account for the fact that the design cycle steps should be modified to include the different kinds of models that characterise the design of simulation games and simulations in general, namely a conceptual model (platform independent), a design model (platform independent) and one or more implementation models (platform dependent). In addition, the processes that lead from one model to the other have been mapped to design phases of analytical wargaming. Aspects of game validation and player experience evaluation have been addressed in this work. Therefore, based on the literature a set of validation criteria for K2AG has been proposed and a player experience questionnaire for K2AGs has been developed. This questionnaire extends work proposed in the literature, but a validation has not been possible at the time of writing. Finally, two instantiations of the K2AG framework, namely the Reliability Game and the MARISA Game, have been designed and analysed in details to validate the approach and show its potentialities.

Knowledge Acquisition Analytical Games: games for cognitive systems design

de ROSA, FRANCESCA
2020-02-26

Abstract

Knowledge discovery from data and knowledge acquisition from experts are steps of paramount importance when designing cognitive systems. The literature discusses extensively on the issues related to current knowledge acquisition techniques. In this doctoral work we explore the use of gaming approaches as a knowledge acquisition tools, capitalising on aspects such as engagement, ease of use and ability to access tacit knowledge. More specifically, we explore the use of analytical games for this purpose. Analytical game for decision making is not a new class of games, but rather a set of platform independent simulation games, designed not for entertainment, whose main purpose is research on decision-making, either in its complete dynamic cycle or a portion of it (i.e. Situational Awareness). Moreover, the work focuses on the use of analytical games as knowledge acquisition tools. To this end, the Knowledge Acquisition Analytical Game (K2AG) method is introduced. K2AG is an innovative game framework for supporting the knowledge acquisition task. The framework introduced in this doctoral work was born as a generalisation of the Reliability Game, which on turn was inspired by the Risk Game. More specifically, K2AGs aim at collecting information and knowledge to be used in the design of cognitive systems and their algorithms. The two main aspects that characterise those games are the use of knowledge cards to render information and meta-information to the players and the use of an innovative data gathering method that takes advantage of geometrical features of simple shapes (e.g. a triangle) to easily collect players’ beliefs. These beliefs can be mapped to subjective probabilities or masses (in evidence theory framework) and used for algorithm design purposes. However, K2AGs might use also different means of conveying information to the players and to collect data. Part of the work has been devoted to a detailed articulation of the design cycle of K2AGs. More specifically, van der Zee’s simulation gaming design framework has been extended in order to account for the fact that the design cycle steps should be modified to include the different kinds of models that characterise the design of simulation games and simulations in general, namely a conceptual model (platform independent), a design model (platform independent) and one or more implementation models (platform dependent). In addition, the processes that lead from one model to the other have been mapped to design phases of analytical wargaming. Aspects of game validation and player experience evaluation have been addressed in this work. Therefore, based on the literature a set of validation criteria for K2AG has been proposed and a player experience questionnaire for K2AGs has been developed. This questionnaire extends work proposed in the literature, but a validation has not been possible at the time of writing. Finally, two instantiations of the K2AG framework, namely the Reliability Game and the MARISA Game, have been designed and analysed in details to validate the approach and show its potentialities.
26-feb-2020
File in questo prodotto:
File Dimensione Formato  
phdunige_2822976.pdf

accesso aperto

Tipologia: Tesi di dottorato
Dimensione 11.83 MB
Formato Adobe PDF
11.83 MB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11567/995827
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact