Aims: Post-ischemic ventricular tachycardia (VT) frequently arises from scar-related re-entrant circuits in myocardial tissue. Ventricular arrhythmogenic substrates are characterized by abnormal potentials (VAPs) occurring in intracardiac electrograms. VAPs identification is a challenging issue, since they constitute the ablation targets in substrate-guided mapping and ablation procedures for VT treatment. Aiming to improve the identification of VAPs in ablation procedures, in this work, two approaches for the supervised classification of VAPs in bipolar intracardiac electrograms are evaluated and compared. Methods: To this aim, in a retrospective analysis, 954 bipolar electrograms were annotated by an expert cardiologist for the supervised classifiers training and test. Signals were acquired from six patients affected by post-ischemic VT by the CARTO3 system at the San Francesco Hospital (Nuoro, Italy) during routine procedures, during which usual clinical protocols were adopted for the subsequent radiofrequency ablation. The first classification approach was based on a support vector machine trained and tested on four different features, extracted from both the time and time-scale domain, to identify physiological and abnormal potentials. Conversely, in order to assess the significance of the first approach and the actual need of exploiting features to solve the classification problem, all the samples constituting a time-domain segment of each bipolar electrogram were given as input to a feed-forward artificial neural network. Results: Classification results showed high recognition accuracies, above 79%, suggesting the efficacy of the two proposed approaches, with some differences in terms of false positive and negative rates according to the chosen identification method. Conclusion: These findings, which underline the possibility of an automatic recognition of VAPs also without trying to identify some peculiar features characterizing abnormal and physiological potentials, are promising. Due to the limited number of patients, a validation on a larger dataset will be pursued.