Decision Tree Algorithm for the Classification of Dental Caries Severity via Saliva

Dental caries is one of the most prevalent chronic diseases. Early detection prominent to avoid tooth weakening or worst loss. UV absorption spectroscopy a non-invasive techniqueused for salivary alpha-amylase which are increasing in presence caries. Spectrum acquired from patient at Faculty dentistry, UKM showed significant peak around 260-300nm correspond amino acid found alpha-amylase.The spectra arepreprocesses using autoscale and multiplicative scatter correction (MSC) optimize signal. Decision tree algorithm was implemented on spectra. The best model decision obtained when preprocessing method. accuracy, precision, sensitivity specificity validation data were 0.70, 1.00, 0.14and 1.00 respectively. requiresmore tuning robustness future application.