Design Of A Predictive Model For TCM Pulse Diagnosis In Malaysia Using Machine Learning

Ong, Jia Ying (2020) Design Of A Predictive Model For TCM Pulse Diagnosis In Malaysia Using Machine Learning. Final Year Project, UTAR.

Preview

PDF Download (3553Kb) | Preview

Abstract

Pulse diagnosis is one of the main diagnosis methods used on patients in Traditional Chinese medicine (TCM). TCM pulse is a time series signal which can be sensed using fingers by TCM practitioners in traditional way. In this project, the TCM pulse signal is collected using a pulse-taking system that consists of amplify spontaneous emissions (ASE), fibre Bragg grating analyser (FBGA) and fibre optic sensor (FBG). In this project, Python is used to build the machine learning models to classify if a person is active in exercising or not through his/her TCM pulse. The machine learning algorithms applied in this project are k-nearest neighbors (KNN), naïve Bayes, random forest, gradient boosting and support vector machine (SVM). People that active in exercising tends to have a slower pulse rate and higher pulse’s height from left ‘Cun’ through the observation of the results in this project. SVM model has the best performance to classify the data set with 315 data samples.

Actions (login required)