ESTIMATION OF EMOTIONAL AND SOCIAL LONELINESS IN ELDERS WITH THE DEVELOPED ARTIFICIAL NEURAL NETWORKS AND MULTIPLE LINEAR REGRESSION MODELS

In recent years, with the increase in the amount of data, the development of the technology required for the analysis of this data has made it easier for artificial intelligence to enter all areas. In this study, "Loneliness Scale for the Elderly" was used to measure loneliness level as a dependent variable, and the predictability of emotional and social loneliness parameters obtained was investigated with artificial intelligence and statistical techniques. For this reason, various scales were used to examine Emotional Loneliness (EL), and Social Loneliness (SL) and various input parameters were used in the scales. In this study, we designed an expert system which uses the Artificial Neural Network (ANN) - Machine Learning Algorithm and Multiple Linear Regression (MLR) statistical methods to estimate the SL and EL values by feeding with input values. Root Mean Squared Error (RMSE) and Correlation Coefficient (R) parameters were used to evaluate the predictive performance of the expert system. When the performance criteria were analyzed, it was found that ANN was the best predictor of SL and EL values. Social or emotional loneliness of individuals can be estimated by entering the questionnaire responses that are not included in the sample through the expert system developed in this study.