The purpose of this study was to provide basic information on the development of nutrition education programs to improve the mealtime behavior of children with autism spectrum disorder (ASD) by investigating the mealtime behavior and food preferences of children with ASD through the perception of special education teachers. Surveys were given to 108 special education teachers in special education schools in Korea regarding the demographic characteristics, nutrition education support needs, mealtime behavior, and food preferences of children with ASD. Most of the special education teachers responded that nutrition education in special schools had not been conducted properly and nutrition education for ASD children is necessary. Mealtime behavior analysis classified the behavior into three clusters: cluster 1, ‘less problematic mealtime behavior’; cluster 2, ‘general feature of autism’; cluster 3, ‘difficulty in self-directed diet’. The age, eating habits, and food preferences were different according to each mealtime behavior cluster. Therefore, it will be necessary to develop a nutrition education program based on the characteristics of mealtime behavior.

Support vector regression (SVR) is devised to solve the regression problem by utilizing the excellent predictive power of Support Vector Machine. In particular, the є-insensitive loss function, which is a loss function often used in SVR, is a function thatdoes not generate penalties if the difference between the actual value and the estimated regression curve is within є. In most studies, the є-insensitive loss function is used symmetrically, and it is of interest to determine the value of є. In SVQR (Support Vector Quantile Regression), the asymmetry of the width of є and the slope of the penalty was controlled using the parameter p. However, the slope of the penalty is fixed according to the p value that determines the asymmetry of є. In this study, a new ε-insensitive loss function with p1 and p2 parameters was proposed. A new asymmetric SVR called GSVQR (Generalized Support Vector Quantile Regression) based on the new ε-insensitive loss function can control the asymmetry of the width of є and the slope of the penalty using the parameters p1 and p2 , respectively. Moreover, the figures show that the asymmetry of the width of є and the slope of the penalty is controlled. Finally, through an experiment on a function, the accuracy of the existing symmetric Soft Margin, asymmetric SVQR, and asymmetric GSVQR was examined, and the characteristics of each were shown through figures.

This study examined the eating habits and preferences of foods according to the age of children with autism spectrum disorders and identified the nutritional needs of parents to support their healthy eating habits and provide basic data for the development of educational programs. A survey of 100 parents of children with autism spectrum disorders was conducted on their children’s snack intake, food preferences, and parents’ nutrition education needs. As a result, there were four times more boys than girls, and the body mass index differed according to age (p<0.001). Most children ate snacks more than once a day, and parents found that they wanted to develop nutritious snacks that could replace high-protein, low-calorie meals with the minimal use of additives. For the food preference, there was a significant difference in grains and starches, vegetables/seafood and fruits, milk and dairy products (p<0.05), fats and sugars, and Korean fermented pastes (p<0.01) between the two age groups. Regarding whether nutrition education was required, many parents had a high demand for nutrition education (4.24±0.70), and the intention to participate was 3.62±1.09. The most important content to be dealt with in parent education was ‘Inducing a change in bad eating habits and behaviors’, which was the highest (63%), and mostly face-to-face education was preferred. In conclusion, young children require diet education, and older children need to focus on weight control and obesity.