Método para ajuste de nível de dificuldade em jogos educacionais fundamentado em aprendizagem de máquina
Serious games are used to boost the learning of people with or without intellectual disabilities in an interactive way with activities and pedagogical content according to the reality experienced by the students. The improvement of games is carried out through the application of gamification techniq...
| Autor principal: | Luz, Vinicius Schultz Garcia da |
|---|---|
| Formato: | Dissertação |
| Idioma: | Português |
| Publicado em: |
Universidade Tecnológica Federal do Paraná
2021
|
| Assuntos: | |
| Acesso em linha: |
http://repositorio.utfpr.edu.br/jspui/handle/1/26235 |
| Tags: |
Adicionar Tag
Sem tags, seja o primeiro a adicionar uma tag!
|
| Resumo: |
Serious games are used to boost the learning of people with or without intellectual disabilities in an interactive way with activities and pedagogical content according to the reality experienced by the students. The improvement of games is carried out through the application of gamification techniques and elements that seek to maintain attention and engage the student to overcome challenges. Additionally, machine learning algorithms are being used in serious games in order to obtain more accurate information about the game’s matches, characteristics, difficulties encountered and the student’s own behavior. This work created the method Tuning Game Level by Machine Learning (TGL-ML) to identify attributes and apply machine learning algorithms in order to obtain patterns, rules and indexes to change levels of difficulty. The developed method is divided into two parts, before and after applying machine learning algorithms. The first part describes the elaboration of a memory game containing the definition of the target audience and its characteristics, the theme of the game, the use of gamification and gamification elements, the creation of the game, the definition of attributes and the game features, ending with the generation of the first version of the game. And at the end of the first part, the game is applied to the target audience, data collection and functionality adjustments based on feedbacks received. The second part contemplates the use of machine learning algorithm with the database prepared for subsequent application of machine learning algorithms for data analysis. After analyzing the data, the acquired knowledge is used to update the game version, adding rules and indexes obtained in order to adjust the difficulty levels in a new version of the serious game. The results achieved through the application of machine learning algorithms indicate that the rules must be added in two of the difficulty levels available in the memory game. |
|---|