In this section, we'll look at some of the artificial intelligence use cases that comm service providers are looking at and some of the initiatives that are ongoing in the industry today. So across the whole sphere of the telco business model and the operational model, there's a wide range of different use cases from everything from the network core services associated with areas such as revenue assurance into the OSS/BSS and fraud detection, from managing customers churn prediction, quality of experience analysis for example, you know, the service assurance and maintaining the SLAs that those service providers and customers are receiving and contracted to. Areas including the MANO integration, the predictive maintenance aspects, you know, detecting faults before they actually occur and it becomes service affecting within a network. And then as I mentioned, the Holy Grail of the covenant networking, the self optimizing network, the self-healing network within the operators operational domain. And then if you're looking for the field around the Core and the Edge aspects of the comm service providers, everything from natural language processing, speech recognition, text and customer communications, chat bots virtual agents, for example. Cybersecurity, looking at how security instance within the network can be identified and rectified before they become service affecting. And then down to the IoT, the Edge Analytics, the Visual Analytics, the whole area of media analytics, identifying images, analyzing video streams as well, for various reasons both for within the comm service provider domain, but also for enabling over-the-top services for the comm service providers and customers. So if we look at some of the use cases appropriate in the network side of things, a wide range of very detailed use cases which are covering everything from capacity planning and optimization, network planning and optimization in terms of your forward planning of networks, and how they're going to plan and address future growth of the networks. And then also with the wider adoption of SDN and NFE within the operating networks, how you're actually going to orchestrate and provision the various levels of various software components within them. And then looking further ahead in terms of 5G, network slicing, as well as areas such as the 5G run radio access network, how the access network is going to be planned and managed going forward. So if we look at some of the examples where artificial intelligence is being looked at within the network and investigated from a com services provider angle, this is an initiative that's being coordinated by Intel and run by Intel in terms of looking at the platform service assurance aspects, everything from hardware, the software levels, upwards in terms of gathering data, telemetry data, performance data, and looking at how that could be used for specific use cases, such as workload placement within the network, looking at how to manage noisy neighbors, how to manage and predict hardware and faults before they actually, again, become service affecting to the end customer services. On a security aspect, there's the community driven Apache Spot initiative very much built from the ground up, providing advanced analytics to the IT security telemetry data altogether on an open scalable platform. And really what Apache Spot does, it expedites the threat detection investigation and remediation for modeling of the data using machine learning techniques, gathering multiple sources of data and using this open data model, which can be expansible for managing security within the IT security domain. Acumos is a open-source AI platform. It's very much a framework driven approach, making it easy to build, scale, and deploy artificial intelligence applications. It's really a ranking of standardizing the infrastructure and stack components that are required for deploying artificial intelligence applications. And really, it enables the data scientists and data engineers to focus on what they're actually good at, which is building those data science and artificial intelligence applications without having to worry about the actual underlying infrastructure in which it's going to operate on. And then finally within this section, you have ETSI. We're looking at this experiential network intelligence approach, very much looking at defining an architecture that uses artificial intelligence techniques to drive context where metadata driven policies drive a number of different use cases, which they've identified covering the key aspects of the operators operational site such as the Network Operations Network Assurance Service Orchestration, as well as the Infrastructure Management.
