Abstract: Methods and systems for measuring 5G communication quality are described herein. According to an implementation, a computer server may determine a plurality of subscribers where each corresponding user equipment carries at least one session on 5G new radio every first time period in a day. For each subscriber, the computer server may obtain data associated with second sessions carried on the corresponding user equipment during the first time period; extract, from the data, one or more features associated with the second sessions; and determine, using a machine learning model and based at least in part on the one or more features, a value being indicative of a likelihood that each subscriber is having a poor 5G network experience during the first time period. The computer server may further determine, based at least in part on the value, an overall 5G communication quality score of the plurality of subscribers.