The authenticity assessment of edible oils is crucial to reassure consumers of product compliance. In this study, a new approach was taken to combining untargeted profiling by using differential scanning calorimetry (DSC) with chemometric methods in order to distinguish cold-pressed oils (flaxseed, camelina, hempseed) from refined oils (rapeseed, sunflower, soybean). The whole spectrum of DSC melting profiles was considered as a fingerprint of each oil. Flaxseed and hempseed oils exhibited four endothermic peaks, while three peaks with one exothermic event were detected for camelina seed oil. In the case of refined oils, two endothermic peaks were detected for rapeseed oil, three for sunflower oil and four for soybean oil. Thermodynamic parameters, such as peak temperature, peak heat flow and enthalpy, differed for each type of oil. Principal component analysis (PCA) and orthogonal partial least squares-discriminant analysis (OPLS-DA) were used for processing data consisting of the whole spectrum of heat flow variables of melting phase transition. PCA demonstrated a clear separation between refined and cold-pressed oils as well as six individual oils. The OPLS-DA showed a distinct classification in six classes according to the types of oils. High OPLS-DA coefficients including R2X(cum)=0.971, R2(cum)=0.916 and Q2X(cum)=0.887 indicated good fitness of the model for oil discrimination. Variables influence on projection (VIP) plot indicated the most significant variables of the heat flow values detected at temperatures around −29°C, −32°C, −14°C, −10°C, −24°C and −41°C for the differentiation of oils. The study ultimately demonstrated great potential of the untargeted approach of using the whole melting DSC profile with chemometrics for the discrimination of cold-pressed and refined oils.