Abstract: Urinary Tract Infection (UTI) is a serious health problem affecting millions of people each year and it is significant to identify the causal agent prior to treatment. The bacteria typically associated with UTI include shape Eschericha coli, shape Klebsiella, shape Proteus mirabilis, shape Citrobacter freundii and shape Enterococcus sp. In recent years, a number of spectroscopic methods such as Fourier transform infrared (FT-IR) spectroscopy have been used to analyse the bacteria associated with UTI which are generally described as rapid whole organism fingerprinting. FT-IR typically takes only 10 sec per sample and generates holistic biochemical profiles from biological materials. In the past, multivariate analysis and artificial neural networks have been used to analyse and interpret the information rich data. In this study, The Support Vector Machine (SVM) applied to the FT-IR data for the automatic identification of UTI bacteria. Cross-validation test results indicate that the generalization performance of the SVM was over 98% to identify the UTI bacteria, compared to neural network's accuracy of 81%. Among the various multi-class SVM schemes tested, the Directed Acyclic Graph (DAG) method gives the best classification results. A Principal Component Analysis (PCA) based dimension-reduction could accelerate the training/testing time to a great extent, without deteriorating the identification performance.