User's Comment Classifying Method Using Self Organizing Feature Map on Healthcare System for Diabetic

Mera Kazuya

Ichimura Takumi

Fourth International Workshop on Computational Intelligence & Applications Proceedings : IWCIA 2008

Diabetes is a metabolic disorder characterized by the elevation of blood glucose. Glysemic control can delay the onset and slow the progression of vascular complications. Lifestyle modification including weight reduction can contribute significantly to glycemic control. The Health Support Intelligent System for Diabetic Patients (HSISD) can provide guideline-based decision support for lifestyle modifications in the treatment of diabetes. HSISD also provides opportunities for telecounseling (TC) with the use of mobile devices and the Internet. The telecounseling phase inquires about the patient's condition and the patient answer in a questionnaire. In the questionnaire, there is a question like “Have you developed any symptoms of anxiety? If yes, tell me the details." The answer is described freely so the physician should read all of patient's answer. But it is hard for physicians to read all text carefully because a physician has a lot of patients. We propose a method to analyze text data from the patients and classify them into five anxiety types (mental problem, physical problem, diet, physical activity, and medicine) automatically. Related to the classified anxiety type, the method can analyze the patient's inner emotion to guess serious and emergency degree of the patient. In this method, Self organizing feature map is trained by the distribution of feature words (morphemes) in the input text and also classifies anxiety type and emotion type.

技術・工学 [ 500 ]

英語

IEEE SMC Hiroshima Chapter

(c) Copyright by IEEE SMC Hiroshima Chapter.

[ISSN] 1883-3977