Claim Missing Document
Check
Articles

Found 1 Documents
Search

Identifikasi Halitosis Berdasarkan Tingkatan Berbasis Sensor Gas Menggunakan Metode Learning Vector Quantization Dodon Yendri; Anisa Irviana; Andrizal Andrizal
JITCE (Journal of Information Technology and Computer Engineering) Vol 1 No 01 (2017): Journal of Information Technology and Computer Engineering
Publisher : Universitas Andalas

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (1227.274 KB) | DOI: 10.25077/jitce.1.01.35-47.2017

Abstract

Diabetes mellitus and gastric infections can be detected through bad breath bad breath (halitosis). Halitosis is a condition where the smell of bad breath occurs when a person exhales (usually smells when talking). This study aims to create an oral health identification and classification system (halitosis). TGS-2602 gas sensor will detect gas levels in the mouth of the patient, and send data in the form of an analog signal to the ATmega 328 microcontroller. By programming the data read on the Raspberry Pi, the data from the microcontroller is stored in a file and then the data is processed using the Fast Fourier Transform method. (FFT) so that the desired data pattern is obtained. The data pattern of the Fast Fourier Transform (FFT) output will be used as input data on the Learning Vector Quantization (LVQ) neural network method. System testing is done to people with halitosis and not halitosis bad breath. The results showed that the percentage success rate of sensor responses to mild halitosis samples was 25%, moderate halitosis samples were 50%, acute Halitosis samples were 50% and non-halitosis samples were 100%.