RESUMEN
Software estimation involves meeting a huge number of different requirements, such as resource allocation, cost estimation, effort estimation, time estimation, and the changing demands of software product customers. Numerous estimation models try to solve these problems. In our experiment, a clustering method of input values to mitigate the heterogeneous nature of selected projects was used. Additionally, homogeneity of the data was achieved with the fuzzification method, and we proposed two different activation functions inside a hidden layer, during the construction of artificial neural networks (ANNs). In this research, we present an experiment that uses two different architectures of ANNs, based on Taguchi's orthogonal vector plans, to satisfy the set conditions, with additional methods and criteria for validation of the proposed model, in this approach. The aim of this paper is the comparative analysis of the obtained results of mean magnitude relative error (MMRE) values. At the same time, our goal is also to find a relatively simple architecture that minimizes the error value while covering a wide range of different software projects. For this purpose, six different datasets are divided into four chosen clusters. The obtained results show that the estimation of diverse projects by dividing them into clusters can contribute to an efficient, reliable, and accurate software product assessment. The contribution of this paper is in the discovered solution that enables the execution of a small number of iterations, which reduces the execution time and achieves the minimum error.
RESUMEN
Background and Objectives: Hyperinsulinemia and insulin resistance are not synonymous; if the risk of developing insulin resistance in adolescents is monitored, they do not necessarily have hyperinsulinemia. It is considered a condition of pre-diabetes and represents a condition of increased risk of developing DM (diabetes mellitus); it can exist for many years without people having the appropriate symptoms. This study aims to determine the risk of developing hyperinsulinemia at an early age in adolescents by examining which factors are crucial for its occurrence. Materials and Methods: The cross-sectional study lasting from 2019 to 2021 (2 years) was realized at the school children's department in the Valjevo Health Center, which included a total of 822 respondents (392 male and 430 female) children and adolescents aged 12 to 17. All respondents underwent a regular, systematic examination scheduled for school children. BMI is a criterion according to which respondents are divided into three groups. Results: After summary analyzes of OGTT test respondents and calculated values of HOMA-IR (homeostatic model assessment for insulin resistance), the study showed that a large percentage of respondents, a total of 12.7%, are at risk for hyperinsulinemia. The research described in this paper aimed to use the most popular AI (artificial intelligence) model, ANN (artificial neural network), to show that 13.1% of adolescents are at risk, i.e., the risk is higher by 0.4%, which was shown by statistical tests as a significant difference. Conclusions: It is estimated that a model using three different ANN architectures, based on Taguchi's orthogonal vector plans, gives more precise and accurate results with much less error. In addition to monitoring changes in each individual's risk, the risk assessment of the entire monitored group is updated without having to analyze all data.