The analysis of EEG signals plays an important role in a wide range of applications, such as psychotropic drug research, sleep studies, seizure detection and hypnosis PROCESSING. From years ago hypnosis was known as a method to help the patients in different fields such as reduction of stress, leaving bad habits, pain control and etc. EEG signals during pure hypnosis would differ from those recorded in the normal no hypnotic conditions. There are several methods for analyzing the EEG signal and similarity index method is one of the famous methods in this branch. In this paper the features of EEG signal of three groups of people with different hypnotizability during hypnosis (Fractal, Wavelet Entropy, Frequency Bands) from left-right and frontal-back lobes were extracted and analyzed using Fuzzy Similarity Index Method to find whether there are any significant relations between the function of these HEMISPHERES and hypnotizability degree. Finally after detecting the significancy, we used the selected features were used to classify the subjects into three groups of hypnotizability. The best classification accuracy was obtained 94% (for two classes of features 1. entropy, Higuchi, high frequency, 2. energy and entropy) and the lowest was 87.5% (for entropy, Higuchi and low frequency features).