Machine-learning (ML) and artificial-intelligence (AI) algorithms can be applied for networks performance evaluation, monitoring and troubleshooting techniques, and voice/video quality of service (QoS) and quality of experience (QoE) prediction models such as the ones developed by ITU-T.

A powerful technique, ML is inherently very complex and therefore prone to misusage and misinterpretation, which can consequently drastically impact the strengths and benefits of ML techniques. Therefore, it is required to carefully follow well defined guidelines when applying ML.

Thus, this Recommendation introduces general guidelines for applying ML within the context of ITU-T which are suitable to these techniques. In addition to the guidance provided by this Recommendation, it is suggested that the guidance given in [b-ITU-T P.Suppl. 28] also be taken into account.