Search

Apollo & Rosetta is an Exergame developed for Inhibitory Control stimulation in Elementary School children. This works' goal has been to demonstrate the behavior of the difficulty curves planned for seven activities (minigames) ingame, as well as their correspondence with the variables collected during a pilot neuropsychological intervention. Seven students participated in the study and played the minigames 1528 times during the 3-month intervention. Each of the minigames had a difficulty curve computed with the goal of keeping the players in the state of Flow. The curves were designed in cycles which grow throughout levels (Normal Level) to a peak (Peak Level), followed by a rest period (Rest Level). The pilot study encompassed three different analyses: (1) Exploratory performance analysis with Spearman correlation, which indicated a positive and significant general correlation between performance and level difficulty; (2) Success exploratory analysis, which showed that as the stages progressed, the success rate increased, even if the level difficulty also increased; (3) Analysis of the factors which influenced performance, through Mixed Effects Logistic Regression and the Backward method. This analysis demonstrated that the odds ratio for overcoming challenges between Normal levels was 0.71 [0.59;0.86] times lower than Rest Level ( p -value = 0.000), whereas in Peak levels it was 0.62 [0.47;0.83] times lower than Rest level values ( p -value = 0.001). These data confirm the overall planned behavior of the difficulty curves., (Copyright © 2019 Mossmann, Cerqueira, Barbosa, Fonseca and Reategui.)

This paper presents a new approach for integrating case-based reasoning (CBR) with a neural network (NN) in diagnostic systems. When solving a new problem, the neural network is used to make hypotheses and to guide the CBR module in the search for a similar previous case that supports one of the hypotheses. The knowledge acquired by the network is interpreted and mapped into symbolic diagnosis descriptors, which are kept and used by the system to determine whether a final answer is credible, and to build explanations for the reasoning carried out. The NN-CBR model has been used in the development of a system for the diagnosis of congenital heart diseases (CHD). The system has been evaluated using two cardiological databases with a total of 214 CHD cases. Three other well-known databases have been used to evaluate the NN-CBR approach further. The hybrid system manages to solve problems that cannot be solved by the neural network with a good level of accuracy. Additionally, the hybrid system suggests some solutions for common CBR problems, such as indexing and retrieval, as well as for neural network problems, such as the interpretation of the knowledge stored in a neural network and the explanation of reasoning.