Jogo sério para ensino e prática de detecção de outliers

• Ano de defesa: 2018
• Autor(a) principal: Rangel, Jean Avila
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Tecnológica Federal do Paraná; Curitiba; Brasil; Programa de Pós-Graduação em Computação Aplicada; UTFPR
• Palavras-chave em Português: Jogos eletrônicos; Jogos para computador; Auditoria - Processamento de dados; Mineração de dados (Computação); Sistemas de recuperação da informação; Tecnologia da informação; Electronic games; Computer games; Auditing - Data processing; Data mining; Information storage and retrieval systems; Information technology; CNPQ::CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO::METODOLOGIA E TECNICAS DA COMPUTACAO::ENGENHARIA DE SOFTWARE; Ciência da Computação
• Link de acesso: http://repositorio.utfpr.edu.br/jspui/handle/1/3933

A serious game can be defined as any game in which the main purpose is not purely fun. Another definition that can be attributed to serious games is that it can serve a social purpose for developing citizenship among players. Outlier detection is an area of data mining and it can be used for fraud detection. Many works use outlier detection and fraud detection as a means to avoid losses in enterprise financial systems and organizations. As an outlier detection technique, data visualization presents data graphically to facilitate its understanding and the creation of new knowledge after interpretation. There are few studies that present outlier detection concepts for players of a serious game using data visualization and three-dimensional graphics. The goal of this work is to investigate if users learn or exercise outlier detection concepts when playing a serious game intended to teach this subject. This study has developed and evaluated a serious game with students and professionals related to public auditing. A case study was used to analyse the opinions of the students and professionals who played the game and answered a questionnaire. The game proved to be a good alternative for teaching the data mining subject, especially in the opinion of the interviewees who already had knowledge in the discipline. After the State of the Art study of serious games and outlier detection to define how and what subjects would be presented by the game, we developed and applied a case study in three distinct groups. After playing the game developed, all volunteer game validation players answered a questionnaire. In the responses of the first two groups, the questions regarding the efficacy of the game for teaching the subject of data mining received mostly neutral answers, where the respondents did not know how to comment. Between each case study, the comments and considerations elaborated previously were considered to generate an improved version of the serious game. A teacher of data mining and a teacher of serious games and human-computer interaction played the game and suggested some changes. We consider the appropriate changes for the scope of this work and met them. At the last case study, with undergraduate students who had had contact with the data mining discipline, the game got positive responses as to its effectiveness in teaching outlier detection. We consider that the positive answers in the last case study occurred due the continuous improvement of the software and also because we aimed the serious game at students who had had recent contact with the area. With the results, we can conclude that a serious game can be a good alternative to aid in the teaching of outlier detection, however, only when it is accompanied by an interest of the player in learning the subject and with the support of a course in a traditional teacher-student model.