Prediction of the clinical and naming status after anterior temporal lobe resection in patients with epilepsy

•We report state-of-the-art statistical approaches to model cognitive data.•Machine learning and graph theory are complementary methods to capture cognitive phenotypes.•We assess validity of cognitive indices in peri-surgical context of refractory temporal lobe epilepsy.•Neuropsychological scores robustly predict postoperative outcomes.•We provide statistical tools bringing a renewed focus of cognitive scores.