The evaluation of edible oil by molecular spectroscopy increased over the years. In this work, cotton oil samples were evaluated by ultraviolet-visible (UV-Vis) and near infrared (NIR) spectroscopy from room temperature up to sixty frying cycles, with a spectrum collect at every five cycles. A correlation analysis between the absorbance results at a different wavelength on UV-Vis region (200 – 800 nm) and NIR (900 – 1680 m) was carried out using Matlab R2007b software and PLS-Toolbox 5.2 tools. A linear regression was performed for each absorbance value in the UV–Vis region against each absorbance values in the NIR region. The correlation coefficient was calculated in each case and, with the objective of facilitating the visualization of results, correlation maps were built. The characteristic region of tocopherol in the UV–Vis region (325 nm) showed the correlation with the NIR spectroscopy around 1090 nm and 1420 nm. This NIR region also presented correlation with the UV–Vis region attributed to the oxidation products (350 –550 nm). The chlorophyll absorption on UV-Vis (670 nm) exhibited correlation with NIR region around 1100 and 1310 nm. The results suggest that NIR spectroscopy is able to evaluate edible oils and can be an alternative to the characterization of oils subjected to frying cycles since it is more selective when compared to UV-Vis spectroscopy.