SLEEPING CELL DETECTION

An apparatus, method and computer program is described, comprising: obtaining initial training labels for first performance management data of a mobile communication network, wherein the initial training labels are generated by a first classification model to identify potentially sleeping cells of the mobile communication network; training a machine-learning model based on the initial training labels, wherein the model is configured to identify potentially sleeping cells within the mobile communication network; identifying potentially sleeping cells using the trained machine-learning model based on second performance management data of the mobile communication network; receiving feedback based on an impact of corrective action taken based on the identification, by the machine-learning model, of said potentially sleeping cells; updating said training labels based on said received feedback; and retraining the machine-learning model based on the updated training labels.