Different categories of features have been extracted from the uterine Electrohysterography Signal (EHG), in order to find specific useful information that can be used to classify between labor and pregnancy contractions. The selection of pertinent features is very important to solve classification problem. In our study we selected 26 features (linear, nonlinear and features related to the EHG propagation) extracted from the literature. In this paper we present two feature selection methods. The first one is a filter method named F-score. The second one is a wrapper method named Genetic algorithm. The aim of this paper is to compare the results of feature selection methods and classification obtained by using monopolar and then bipolar EHG signals.