We propse an autoencoder-based technique for identifying anomalies in rating-scale data.
Coming soon!