Epilepsy is one of the most common neurological disorders worldwide. Its manifestations, the seizures, are due to a group of neurons' abnormal and synchronous activity. The unpredictable nature of these events hinders the quality of life of those affected. In particular, focal seizures show a localized onset of the abnormal activity and are the most common ones. Correct detection of the episodes can help clinicians to give the best medical treatments. This research project arises from the need to have automatic algorithms for seizure detection with a high number of correctly detected seizures for low false alarm rates. Recent studies have shown disorganization in how brain areas interact with each other before and during a seizure. We decided to model this change in connectivity patterns by inferring graphs from EEG recordings of epileptic patients. We work with seventeen subjects suffering from focal epilepsy, and we build, for each of them, a graph of the activity preceding (preictal) and during (ictal) a seizure. After that, we exploit techniques from graph signal processing to build a detector for seizures. Last, we analyze the density of connections of the inferred graphs to indicate the seizure onsets. The obtained results are unsuitable for real-life applications, but they are a starting point for further research. Furthermore, we find that most the proposed ictal or preictal graphs show less connections in the nodes involved with the seizure onset.