치매에 대한 예측 머신러닝데이터 관점에서 = Prediction of Dementia from Machine Learning Data

다국어 초록 (Multilingual Abstract)

The main purpose of the study is to predict mental health status, that is, Alzheimer's diagnosis, by analyzing factors tailored to each data set using machine learning models. This study aims to find more relevant factors by analyzing unique factors existing in each data set. To this end, this study used decision tree models, random forest models, KNN models, SVM models, artificial neural network models, naive Bayesian models, logistic regression analysis, and XG boost models among the developed machine learning models. In the process of training the model using medical data, we went through trial and error, such as increasing variable values to increase the model performance index value that determines the degree of learning. In addition, we did not end this study by comparing the performance of the models but ended the study by finding out which variables are closely related to dementia prediction and their weights. These results can provide a foundation for what approach is needed when processing medical data. In addition, it will be helpful for research that predicts results through medical data and finds out which variables are closely related.