Open AccessPrediction Medical Problem of Elderly People by Using Machine Learning Technique
Author(s) -
A. Tongkaw,
Sasalak Tongkaw
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1529/3/032083
Subject(s) - hobby , python (programming language) , computer science , machine learning , decision tree , graph , artificial intelligence , artificial neural network , id3 algorithm , data mining , decision tree learning , theoretical computer science , incremental decision tree , political science , law , operating system
This paper describes the machine learning technique for classification and prediction of the possibility of medical problem occur of elder people by using sklearn in Python. The research chose 16 attributes out of 23 attributes. Overall 353 samples can be classified and divided into category 0, no medical condition, 166 samples, and category 1, have medical condition, 187 samples. The root node is level of education. It can be classified with number of cigarettes per day and have exercise or not. An accuracy value after remove attribute hobby is about 67.80%, considered as good accuracy with the real data. The graph for represent the decision tree results draw by using pydotplus. This work can improve for future work by find out more parameters or other algorithm such as neural network for improving the accuracy.