The number of multimedia services in modern fixed and mobile networks is growing day by day. Quality of service assessment is an important indicator for both the service provider and the end user. The article describes and implements a new machine learning model that allows modeling of video quality perceived by the user, which uses the values of metrics for objective video quality assessment as input. Quality models built so far based on single classic metrics, such as Peak Signal-to-Noise Ratio or Structural Similarity Index, show very limited correlation values between the objective assessment and the subjective assessment of users. The developed model, using a combination of many objective metrics, allows us to obtain a higher correlation value between QoS and QoE than the classic analytical approach.