AIOps is (the usage of) Artificial Intelligence for IT Operations. It combines big data and machine learning to automate IT operations processes, including event correlation, anomaly detection and causality determination.

• https://www.gartner.com/en/information-technology/glossary/aiops-artificial-intelligence-operations

Adaptive Machine Learning

Adaptive machine learning builds on traditional machine learning to create a more advanced solution to real-time environments with variable data. As its name suggests, adaptive machine learning can adapt to rapidly changing data sets, making it more applicable to real-world situations.

• Reference(s) or Source:

• https://www.

• encora.com/insights/

• machine-learning-what-is-adaptive-ml

Adversarial AI/ML

A practice concerned with the design of ML algorithms that can resist security challenges, the study of the capabilities of attackers, and the understanding of attack consequences.

• Reference(s) or Source: “The Language of Trustworthy AI: An In-Depth Glossary of Terms (updated August 4, 2024)”  (https://docs.google.com/spreadsheets/d/e/2PACX-1vTRBYglcOtgaMrdF11aFxfEY3EmB31zslYI4q2_7ZZ8z_1lKm7OHtF0t4xIsckuogNZ3hRZAaDQuv_K/pubhtml)  NIST(Reznik,_Leon)

AI Accuracy

Closeness of computations or estimates to the exact or true values that the statistics were intended to measure.

• Reference(s) or Source:“The Language of Trustworthy AI: An In-Depth Glossary of Terms (Updated August4, 2024)”  (https://docs.google.com/spreadsheets/d/e/2PACX-1vTRBYglcOtgaMrdF11aFxfEY3EmB31zslYI4q2_7ZZ8z_1lKm7OHtF0t4xIsckuogNZ3hRZAaDQuv_K/pubhtml)  
   (https://nvlpubs.nist.gov/nistpubs/ai/NIST.AI.100-3.pdf) 

AI Agent

An artificial intelligence (AI) agent is a software program that can interact with its environment, collect data, and use the data to perform self-determined tasks to meet predetermined goals. Unlike traditional automation agents, which follow static, predefined rules, AI agents can learn from their environment, adapt their behaviour, and make autonomous decisions based on real-time data, making them more flexible and capable of handling dynamic situations. 

• Reference(s) or Source: https://aws.amazon.com/what-is/ai-agents/ 

AI as a Service

Artificial Intelligence as a Service (AIaaS) is a cloud-based service offering artificial intelligence (AI) outsourcing. AIaaS enables individuals and businesses to experiment with AI, and even take AI to production for large-scale use cases.

• Reference(s) or Source: https://www.run.ai/guides/machine-learning-in-the-cloud/ai-as-a-service

AI Deployment Flexibility

Flexibility to deploy the same system in multiple scenarios without any modifications to the AI models. It goes hand in hand with generalisability.

• Reference(s) or Source:  https://hexa-x.eu/wp-content/uploads/2023/07/Hexa-X-D1.4-Final.pdf

AI Policy Enforcer

AI functionality to implement a recommended policy.

• Reference(s) or Source: https://hexa-x.eu/wp-content/uploads/2023/07/Hexa-X-D1.4-Final.pdf

An AI-powered Virtual Agent is an animated virtual character, more complex than a chatbot, that makes use of technologies like machine learning and natural language processing (NLP). This allows it to actively participate in a conversation, acting more like a human.

• Reference(s): based on https://www.ringcentral.com/virtual-agent.html and TM Forum AI Fundamentals course [TMF_AIF] and TM Forum “AI and its pivotal role in transforming operations” report and webinar [TMF_AI]

A logical function in NWDAF, which performs inference, derives analytics information (i.e. derived statistics and/or predictions based on Analytics Consumer Request) and exposes analytics service.

• Reference(s) or Source:https://www.tech-invite.com/3m23/toc/tinv-3gpp-23-288_c.html 
  

API (Application Programming Interface)

An API is a set of commands, functions, protocols, and objects that programmers can use to create software or interact with an external system. Any data can be shared with an application program interface.

• based on https://techterms.com/definition/api and https://searchapparchitecture.techtarget.com/definition/appli cation-program-interface-API

Architecture component

An architecture component is a nontrivial, nearly independent, and replaceable part of a system that fulfills a clear function in the context of a well-defined architecture.

• TM Forum Reference, TMF071 ODA Terminology, TMF071, Release 19.0.1, October 2019

Architecture principles

Architecture principles define the underlying general rules and guidelines for the use and deployment of all IT resources and assets across the organisation. They reflect a level of consensus among the various elements of the enterprise, and form the basis for making future IT decisions.

• based on https://pubs.opengroup.org/architecture/togaf8-doc/arch/chap29.html

Human-like intelligence, which can be applied widely as opposed to narrow AI, which can only be applied to one particular problem or task. Also called 'strong' AI as opposed to 'weak' AI.

• Reference(s) or Source: “The Language of Trustworthy AI: An In-Depth Glossary of Terms (updated August 4, 2024)” (https://docs.google.com/spreadsheets/d/e/2PACX-1vTRBYglcOtgaMrdF11aFxfEY3EmB31zslYI4q2_7ZZ8z_1lKm7OHtF0t4xIsckuogNZ3hRZAaDQuv_K/pubhtml)  NIST (AI_Ethics_Mark_Coeckelbergh)
  

Artificial intelligence (AI) is the ability of a digital computer or computer-controlled robot to perform tasks commonly associated with intelligent beings. It is the system’s ability to correctly interpret external data, to learn from such data, and to use that learning to achieve specific goals and tasks through flexible adaptation.

• based on https://www.britannica.com/technology/artificial-intelligence and Kaplan, A., & Haenlein, M. “Siri, Siri, in my hand: Who’s the fairest in the land? On the interpretations, illustrations, and implications of artificial intelligence”. Business Horizons. 2019; 62:15–25 (https://www.sciencedirect.com/science/article/abs/pii/S0007681318301393)

Automated root cause analysis

Automated RCA is the process of using automation to investigate incident root causes in real time using AI/ML.

• Reference(s) or Source: https://www.bigpanda.io/blog/why-automated-root-cause-analysis-matters/ 
  

Automated service provisioning

Automated service provisioning is the ability to deploy an information technology or telecommunications service by using pre-defined procedures that are carried out electronically without requiring human intervention.

• multiple sources including US government documents, e.g. Financial Services and General Government Appropriations for 2016 p.201 (https://books.google.de/books?id=h4SVIm3XaUsC&printsec=frontcover&hl=de&source=gbs_ge_summary_r&cad=0#v=onepage&q=201&f=false)

Automation

Processing tasks in a repeatable manner to yield the same result every time without human intervention.

• internal definition

Autonomy (autonomous AI system)

AI-enabled Autonomy is the capability of machines (either platforms or computer software) to operate independent of direct human intervention, but within constraints, to achieve a goal or solve a problem.

• Reference(s) or Source: https://www.baesystems.com/en-us/definition/what-is-ai-enabled-autonomy 
  

Autoscale allows you to automatically scale your applications or resources based on demand.

• Reference(s) or Source: https://learn.microsoft.com/en-us/azure/azure-monitor/autoscale/autoscale-get-started  
  

A systematic error that occurs in the machine learning model itself due to incorrect assumptions in the ML process. Technically, bias is the error between average model prediction and the ground truth. Unwanted bias may place privileged groups at systematic advantage and unprivileged groups at systematic disadvantage.

• Reference(s) or Source: https://www.bmc.com/blogs/bias-variance-machine-learning/
  

Bidirectional Encoder Representations from Transformers (BERT) is a deep learning strategy for natural language processing (NLP) that helps artificial intelligence (AI) programs understand the context of ambiguous words in text.

• Reference(s) or Source: https://www.techopedia.com/definition/34116/bidirectional-encoder-representations-from-transformers-bert 
  

Big data reflects extremely large or complex datasets that may be analysed computationally, rather than by traditional data-processing application software, to reveal patterns, trends and associations, especially relating to human behaviours and interactions.

• based on https://link.springer.com/article/10.1057/s41272-019-00191-9 and https://en.wikipedia.org/wiki/Big_data

A type of future network framework that collects big data from networks and applications, and generates big data intelligence based on that data; it then provides big data intelligence to facilitate smarter and autonomous network management, operation, control, optimisation and security, etc.

• ITU Recommendation Y.3652 “Big data driven networking – requirements” (06/20) (https://www.itu.int/rec/T-REC-Y.3652-202006-I/en)

Blockchain

A blockchain is an expanding list of cryptographically signed, irrevocable transactional records shared by all participants in a network.

• reduced; from TM Forum Reference, TMF071 ODA Terminology, TMF071, Release 19.0.1, October 2019

Cgroups are linux kernel mechanisms to restrict and measure resource allocations to each process group. Using cgroups, you can allocate resources such as CPU time, network, and memory.

• https://subscription.packtpub.com/book/application_development/9781785883057/1/ch01lvl1sec15/namespaces-and-cgroups

A computer program that simulates and processes human conversation (either written or spoken), allowing humans to interact with digital devices, systems and platforms as if they were communicating with a real person.

• https://www.oracle.com/chatbots/what-is-a-chatbot/ 

A software that allows a user to ask it questions using conversational, or natural, language. It is a language model developed by OpenAI, and is based on the GPT (Generative Pre-training Transformer) architecture, which is a type of neural network designed for natural language processing tasks.

• Reference(s) or Source: https://www.britannica.com/technology/ChatGPT 
  

A classifier is the algorithm itself—the rules used by machines to classify data. A classification model, on the other hand, is the end result of your classifier's machine learning. The model is trained using the classifier, so that the model, ultimately, classifies your data.

• Reference(s) or Source: https://monkeylearn.com/blog/what-is-a-classifier/ 
  

Closed-loop processes

An automatic control system in which an operation, process, or mechanism is regulated by feedback.

• Reference(s) or Source: https://www.merriam-webster.com/dictionary/closed%20loop
  

Cloud native application

Cloud Native Application (CNA) refers to a type of computer software that natively utilises services and infrastructure provided by cloud computing providers.

• reduced; from TM Forum Reference, TMF071 ODA Terminology, TMF071, Release 19.0.1, October 2019

Component

A component is a functionally independent part of any system. It performs some function and may require some input or produce some output.

• https://www.techopedia.com/definition/3217/component

Composite service

A composite service is an assembly of one or more elements into an end to end service. It may be recursive so a composite service may become a component of yet another service.

• based on TM Forum Reference, TMF071 ODA Terminology, TMF071, Release 19.0.1, October 2019 and TR274 DSRA Guide R17.5 Reference R02

Container

A container is a standard unit of software that packages up code and all its dependencies so the application runs quickly and reliably from one computing environment to another.

• Reference(s) or Source: https://www.docker.com/resources/what-container/

Control plane

The control plane is responsible for processing a number of different control protocols that may affect the forwarding table, depending on the configuration and type of network device. These control protocols are jointly responsible for managing the active topology of the network.

• Software Defined Networks, A Comprehensive Approach, Paul Göransson, Chuck Black Morgan Kaufmann, 2014

Conversational agents/conversational AI (chatbots)

A conversational agent is any dialogue system that conducts natural language processing (NLP) and responds automatically using human language. Conversational agents represent the practical implementation of computational linguistics, and are usually deployed as chatbots and virtual or AI assistants.

• Reference(s) or Source: https://www.dashbot.io/blog/conversational-agent  

Conversational AI

Conversational AI (or conversational artificial intelligence), refers to technologies that enable machines to understand, process, and respond to human language naturally. These include chatbots and virtual assistants which can perform tasks or provide information based on voice or text inputs.

• Reference(s) or Source: https://boost.ai/blog/what-is-conversational-ai/ 
  

Convolutional neural network (CNN)

A convolutional neural network (CNN) is a type of artificial neural network used primarily for image recognition and processing. Due to its ability to recognize patterns in images, a CNN is a powerful tool but requires millions of labelled data points for training.

• Reference(s) or Source: https://www.arm.com/glossary/convolutional-neural-network

Cortex XSOAR

A platform for security orchestration, automation, and response (SOAR), enhanced with ChatGPT for user-friendly incident analysis and response.

• Reference(s) or Source: https://www.paloaltonetworks.com/blog/security-operations/using-chatgpt-in-cortex-xsoar/

Cortex XSOAR Playbook

A set of automated workflows in Cortex XSOAR, designed to handle security incidents efficiently.

• Reference(s) or Source: https://www.paloaltonetworks.com/blog/security-operations/using-chatgpt-in-cortex-xsoar/

Cross-domain data services

Data services that are delivered across multiple administrative, information or technological domains that allow data sharing among authorized consumers in different domains.

• internal definition based on ETSI GS ZSM 007 V1.1.1 (2019-08): Zero-touch network and Service Management (ZSM); Terminology for concepts in ZSM (https://www.etsi.org/deliver/etsi_gs/ZSM/001_099/007/01.01.01_60/gs_ZSM007v010101p.pdf)

Customer Facing Services (CFS)

A logical capability that is packaged as part of a product offering by service providers to their customers, which is directly purchased, leased, visible to and/or otherwise directly usable by those customers. The logical functionality can be derived from underlying network or information technology (i.e., a dedicated contact number or tailored web-based access to operational support for a specific customer) or may be delivered or supplied by staff or contractors employed by the service provider (i.e., dedicated service team or help desk for a specific customer).

• Reference(s) or Source: https://www.itu.int/rec/T-REC-E.480-200609-I 
  

Data center interconnect (DCI) is a segment of the networking market that focuses on the technology used to link two or more data centers so the facilities can share resources.

• https://searchnetworking.techtarget.com/definition/data-center-interconnect

Data Governance

Data governance is the process of managing the availability, usability, integrity, and security of the data in enterprise systems, based on internal data standards and policies that also control data usage. It ensures that data is consistent, trustworthy, and doesn't get misused, facilitating effective decision-making. It also means setting internal standards – data policies – that apply to how data is gathered, stored, processed, and disposed of. It governs who can access what kinds of data and what kinds of data are under governance.

• Reference(s) or Source: https://cloud.google.com/learn/what-is-data-governance 
  

Data ingestion is the process of transporting data from one or more sources to a target site, system or platform for further processing and analysis. This data can originate from a range of sources, including data lakes, IoT devices, on-premises databases, and SaaS apps, and end up in different target environments, such as cloud data warehouses or data marts.

• https://www.striim.com/what-is-data-ingestion-and-why-this-technology-matters/

• https://www.techtarget.com/searchdatamanagement/definition/data-lake

• https://en.wikipedia.org/wiki/Data_model

Data plane

The data plane (sometimes known as the user plane, forwarding plane, carrier plane or bearer plane) is the part of a network that carries user traffic from one interface to another.

• based on https://searchnetworking.techtarget.com/definition/data-plane-DP

Data Poisoning

Data Poisoning is an adversarial attack that tries to manipulate the training dataset in order to control the prediction behaviour of a trained model such that the model will label malicious examples into desired classes (e.g., labelling spam e-mails as safe).

• Reference(s) or Source: https://paperswithcode.com/task/data-poisoning

Data Quality

Data quality measures how well a dataset meets criteria for accuracy, completeness, validity, consistency, uniqueness, timeliness, and fitness for purpose, and it is critical to all data governance initiatives within an organization.

• Reference(s) or Source: https://www.ibm.com/topics/data-quality

A decision management engine is a customisable solution that represents the logic, often in the form of a rules flow or decision tree, that can be operationalised to automate a decision. […] A decision management engine articulates how smaller decisions branch off to bigger and more complex decisions and ultimately end with a final outcome. This logic can be codified, documented, and often executed in an automated fashion.

• based on https://www.fico.com/en/glossary/decision-engine 

Decoupling

Building approach (in electronics, software, etc.) where the constituent components of a system can be produced, sourced and interchanged independently of the other.

• based on TOGAF 9.2 Reference R16

Deep learning is an iterative approach to artificial intelligence (AI) that stacks machine learning algorithms in a hierarchy of increasing complexity and abstraction. Each deep learning level is created with knowledge gained from the preceding layer of the hierarchy.

• https://www.techopedia.com/definition/30325/deep-learning 

Domain

A collection of network infrastructure under the administrative control of the same organisation.

• internal definition based on ITU-T Y.110 TELECOMMUNICATION STANDARDIZATION SECTOR OF ITU (06/98); SERIES Y: GLOBAL INFORMATION INFRASTRUCTURE; General Global Information Infrastructure principles and framework architecture (https://www.itu.int/rec/dologin_pub.asp?lang=e&id=T-REC-Y.110-199806-I!!PDF-E&type=items)

Dynamic Function Placement (DPS)

The act of dynamically placing network functions. This is done by deploying intelligent algorithms to optimally orchestrate differentiated services across multiple sites and clouds based on diverse intents and dynamic environments' policy constraints.

• Reference(s) or Source:  https://hexa-x.eu/wp-content/uploads/2023/07/Hexa-X-D1.4-Final.pdf

Edge computing refers to a distributed computing paradigm that brings computation and data storage closer to the location where it is needed to improve response times and save bandwidth. Instead of relying on a centralised cloud data centre, edge computing performs these processes at or near the physical location of the user or data source.

• Reference(s) or Source: https://www.ibm.com/cloud/what-is-edge-computing

Evasion attacks

Evasion attacks (a.k.a. adversarial examples) consist of carefully perturbing the input samples at test time to have them misclassified.

• Reference(s) or Source: https://secml.readthedocs.io/en/stable/tutorials/03-Evasion.html

Expert system

An expert system uses artificial intelligence (AI) technologies to simulate the judgement and behaviour of a human expert based on “knowledge” programmed into it by humans, and only following predetermined rules.

• Reference(s) or Source: https://www.techtarget.com/searchenterpriseai/definition/expert-system and TM Forum “AI Fundamentals” course [TMF_AIF]

The data processing technique that engineers use to extract data from different sources, transform the data into a usable and trusted resource, and load that data into the systems end users can access and use downstream to solve business problems.

• https://databricks.com/glossary/extract-transform-load 

A learning model that addresses the problem of data governance and privacy by training algorithms collaboratively without transferring the data to another location.

• Reference(s) or Source: “The Language of Trustworthy AI: An In-Depth Glossary of Terms (updated August 4, 2024)” (https://docs.google.com/spreadsheets/d/e/2PACX-1vTRBYglcOtgaMrdF11aFxfEY3EmB31zslYI4q2_7ZZ8z_1lKm7OHtF0t4xIsckuogNZ3hRZAaDQuv_K/pubhtml),  NIST(Public_Health_and_Informatics_MIE_2021)
  

Federated orchestration

Service orchestration performed by multiple autonomous management domains, to effectively allow services to span across several providers.

• internal definition based on https://e-archivo.uc3m.es/bitstream/handle/10016/27125/service_WCNCW_2018_ps.pdf?sequence=1, ETSI GS ZSM 007 V1.1.1 (2019-08): Zero-touch network and Service Management (ZSM); Terminology for concepts in ZSM (https://www.etsi.org/deliver/etsi_gs/ZSM/001_099/007/01.01.01_60/gs_ZSM007v010101p.pdf) and https://www.researchgate.net/publication/318473608_Orchestration_of_Network_Services_across_multiple_operators_The_5G_Exchange_prototype
  

Foundation model

An AI model that is trained on broad data at scale, is designed for generality of output, and can be adapted to a wide range of distinctive tasks.

• Reference(s) or Source: “Proposal for a regulation of the European Parliament and of the Council on harmonised rules on Artificial Intelligence (Artificial Intelligence Act) and amending certain Union Legislative Acts”  (https://www.europarl.europa.eu/meetdocs/2014_2019/plmrep/COMMITTEES/CJ40/DV/2023/05-11/ConsolidatedCA_IMCOLIBE_AI_ACT_EN.pdf)

Functional block

Self contained unit in an overall system that performs a specific function or task.

• based on TM Forum Reference, TMF071 ODA Terminology, TMF071, Release 19.0.1, October 2019 and ETSI Network Functions Virtualization (NFV); Infrastructure; Methodology to describe Interfaces and Abstractions Reference R08

An approach to training AI models useful for applications like data synthesis, augmentation, and compression where two neural networks are trained in tandem: one is designed to be a generative network (the forger) and the other a discriminative network (the forgery detector). The objective is for each network to train and better itself off the other, reducing the need for big, labeled training data.

• Reference(s) or Source: “The Language of Trustworthy AI: An In-Depth Glossary of Terms (updated August 4, 2024)” (https://docs.google.com/spreadsheets/d/e/2PACX-1vTRBYglcOtgaMrdF11aFxfEY3EmB31zslYI4q2_7ZZ8z_1lKm7OHtF0t4xIsckuogNZ3hRZAaDQuv_K/pubhtml), NIST(NSCAI)

Generative AI

Foundation models used in AI systems specifically intended to generate, with varying levels of autonomy, content such as complex text, images, audio, or video.

• Reference(s) or Source: “Proposal for a regulation of the European Parliament and of the Council on harmonised rules on Artificial Intelligence (Artificial Intelligence Act) and amending certain Union Legislative Acts”  (https://www.europarl.europa.eu/meetdocs/2014_2019/plmrep/COMMITTEES/CJ40/DV/2023/05-11/ConsolidatedCA_IMCOLIBE_AI_ACT_EN.pdf),  P. 42

Generative Pre-trained Transformer

GPT, or Generative Pre-trained Transformer, is a state-of-the-art language model developed by OpenAI. It uses deep learning techniques to generate natural language text, such as articles, stories, or even conversations, that closely resemble human-written text.

• Reference(s) or Source: https://encord.com/glossary/gpt-definition/

Orchestration decomposed into one or more hierarchical interactions where parts of the service are delegated to a subordinate orchestrator.

• ETSI GS ZSM 007 V1.1.1 (2019-08): Zero-touch network and Service Management (ZSM); Terminology for concepts in ZSM (https://www.etsi.org/deliver/etsi_gs/ZSM/001_099/007/01.01.01_60/gs_ZSM007v010101p.pdf)

Holistic Anomaly Detection (e.g., via multi-vector AI/ML-based behavioural analytics)

Anomaly detection, or outlier detection, is the identification of observations, events or data points that deviate from what is usual, standard or expected, making them inconsistent with the rest of a data set. Holistic anomaly detection takes an overall approach to anomaly detection using a variety of methods.

Holistic anomaly detection is a comprehensive approach to identifying unusual patterns or behaviors within data. Rather than relying on a single method, it combines multiple techniques—such as statistical analysis, machine learning models, and rule-based algorithms—to capture a wider range of anomalies. This approach is valuable because it considers the data from multiple perspectives, enhancing the ability to detect different types of anomalies, including subtle or complex ones that might be missed by a single-method approach.

• Reference(s) or Source: https://www.ibm.com/topics/anomaly-detection
  

Horizontal Scaling

Horizontal scaling (or scaling out) means that you scale by adding more machines into your pool of resources.

• Reference(s) or Source: https://ibm.github.io/data-science-best-practices/scaling.html 
  

Human-centric AI

Human-Centered AI (HCAI) is an emerging discipline intent on creating AI systems that amplify and augment rather than displace human abilities. HCAI seeks to preserve human control in a way that ensures artificial intelligence meets our needs while also operating transparently, delivering equitable outcomes, and respecting privacy.

• Reference(s) or Source: https://research.ibm.com/blog/what-is-human-centered-ai 
  

Technology incorporating artificial intelligence (AI) and machine learning to automate administrative tasks across a network.

• based on TM Forum Reference, TMF071 ODA Terminology, TMF071, Release 19.0.1, October 2019

An architectural concept for the support, maintenance, operation and provision of new services which is characterised by: information processing, efficient management, control and use of network resources and standardised communication between physical resources, network functions and services.

• based on International Telegraph and Telephone Consultative Committee (CCITT) Recommendation I.312 / Q.1201 (10/92) Principles of Intelligent Network Architecture (https://www.itu.int/rec/dologin_pub.asp?lang=e&id=T-REC-I.312-199210-I!!PDF-E&type=items) 

Intent-based Networking

A software-enabled automation process that uses high levels of intelligence, analytics, and orchestration to improve network operations and uptime.

• Reference(s) or Source: https://www.juniper.net/us/en/research-topics/what-is-intent-based-networking.html

Intent-based policy / network

Technology incorporating artificial intelligence (AI) and machine learning to automate administrative tasks across a network.

• Reference(s) or Source: based on TM Forum Reference Document, “TMF071 ODA Terminology”, Release 19.0.1, October 2019 [TMF071]
  

The Internet of Things, or IoT, is a system of interrelated networking computing devices, mechanical and digital machines aimed  at objects, animals or people and provided with unique identifiers (UIDs) and the ability to transfer data over a network without requiring human-to-human or human-to-computer interaction.

• based on https://en.wikipedia.org/wiki/Internet_of_things and https://www.techtarget.com/iotagenda/definition/Internet-of-Things-IoT 

Kubernetes is an open-source platform used to automate the deployment, scaling, and management of containerized applications. It orchestrates computing, networking, and storage infrastructure on behalf of user workloads, providing a resilient environment for running distributed systems. Kubernetes allows for self-healing, scaling, and service discovery, making it a vital tool for managing containerized applications at scale.

• Reference(s) or Source: https://kubernetes.io/docs/concepts/overview/what-is-kubernetes/

A machine-learning model designed to represent the language domain.

• Reference(s) or Source: https://www.deepset.ai/blog/what-is-a-language-model

Large Language Model

A class of language models that use deep-learning algorithms and are trained on extremely large textual datasets that can be multiple terabytes in size. LLMs can be classed into two types: generative or discriminatory. Generative LLMs are models that output text, such as the answer to a question or even writing an essay on a specific topic. They are typically unsupervised or semi-supervised learning models that predict what the response is for a given task. Discriminatory LLMs are supervised learning models that usually focus on classifying text, such as determining whether a text was made by a human or AI.

• Reference(s) or Source: “The Language of Trustworthy AI: An In-Depth Glossary of Terms (updated August 4, 2024)” (https://docs.google.com/spreadsheets/d/e/2PACX-1vTRBYglcOtgaMrdF11aFxfEY3EmB31zslYI4q2_7ZZ8z_1lKm7OHtF0t4xIsckuogNZ3hRZAaDQuv_K/pubhtml), NIST (AI_Assurance_2022)

Processes that enable computational systems to “understand” data and gain “knowledge” from it without necessarily being explicitly programmed. (Supervised machine learning and unsupervised machine learning are two examples of machine learning.)

• based on ETSI GR ENI 004 V2.1.1 (2019-10), Experiential Networked Intelligence (ENI); Terminology for Main Concepts in ENI (https://www.etsi.org/deliver/etsi_gr/ENI/001_099/004/02.01.01_60/gr_eni004v020101p.pdf) and Telecommunication Standardisation Sector of ITU (ITU-T) Recommendation Y.3177 (02/2021) Architectural framework for artificial intelligence-based network automation for resource and fault management in future networks including IMT-2020 (https://www.itu.int/rec/dologin_pub.asp?lang=s&id=T-REC-Y.3177-202102-I!!PDF-E&type=items)

Management

The processes aiming at fulfilment, assurance, and billing of services, network functions, and resources in both physical and virtual infrastructure including compute, storage, and network resources.

• based on ITU-T Y.3100 TELECOMMUNICATION STANDARDIZATION SECTOR OF ITU (09/2017); SERIES Y: GLOBAL INFORMATION INFRASTRUCTURE, INTERNET PROTOCOL ASPECTS, NEXT-GENERATION NETWORKS, INTERNET OF THINGS AND SMART CITIES; Future networks: Terms and definitions for IMT-2020 network (https://www.itu.int/rec/dologin_pub.asp?lang=e&id=T-REC-Y.3100-201709-I!!PDF-E&type=items)

Management API

A Management API allows a service requestor to perform all management operations before, during and after the use of a service.

• based on TM Forum Reference, TMF071 ODA Terminology, TMF071, Release 19.0.1, October 2019

Management domain

A collection of physical or functional elements under the control of an entity, aiming at fulfilment, assurance, and billing of services, network functions, and resources in both physical and virtual infrastructure.

• internal definition based on ITU-T Y.3100 TELECOMMUNICATION STANDARDIZATION SECTOR OF ITU (09/2017); SERIES Y: GLOBAL INFORMATION INFRASTRUCTURE, INTERNET PROTOCOL ASPECTS, NEXT-GENERATION NETWORKS, INTERNET OF THINGS AND SMART CITIES; Future networks: Terms and definitions for IMT-2020 network (https://www.itu.int/rec/dologin_pub.asp?lang=e&id=T-REC-Y.3100-201709-I!!PDF-E&type=items) and ITU-T Y.110 TELECOMMUNICATION STANDARDIZATION SECTOR OF ITU (06/98); SERIES Y: GLOBAL INFORMATION INFRASTRUCTURE; General Global Information Infrastructure principles and framework architecture (https://www.itu.int/rec/dologin_pub.asp?lang=e&id=T-REC-Y.110-199806-I!!PDF-E&type=items)

Maturity level

A maturity level is a defined evolutionary plateau for organisational process improvement. Each maturity level matures an important subset of the organisation’s processes, preparing it to move to the next maturity level. The maturity levels are measured by the achievement of the specific and generic goals associated with each predefined set of process areas.

• https://www.megatronicstech.com/maturity-level-of-technology/

A maturity model is an instrument that evaluates the current position of certain capabilities of an organisation and provides indications of how it can transform to improve.

• based on https://www.bmc.com/blogs/maturity-model-itsm/ and the TM Forum AI Fundamentals course [TMF_AIF]

Microservices

Microservices is an approach to software architecture that builds a large, complex application from multiple small components that each perform a single function, such as authentication, notification, or payment processing. Each microservice is a distinct unit within the software development project, with its own code base, infrastructure, and database. The microservices work together, communicating through web APIs or messaging queues to respond to incoming events.

• https://www.nginx.com/learn/microservices/

Modelling Abstractions

Model abstraction is a way of simplifying an underlying conceptual model on which a simulation is based while maintaining the validity of the simulation results with respect to the question being addressed by the simulation.

• Reference(s) or Source: https://www.sciencedirect.com/book/9780123850850/model-based-engineering-for-complex-electronic-systems

Natural language generation (NLG) is the use of artificial intelligence (AI) programming to produce written or spoken narratives from a data set.

• Reference(s) or Source: https://www.qualtrics.com/uk/experience-management/customer/natural-language-generation/?rid=ip&prevsite=en&newsite=uk&geo=RO&geomatch=uk

Natural language processing (NLP) refers to the branch of AI concerned with giving computers the ability to understand text and spoken words in much the same way human beings can.

• https://www.ibm.com/cloud/learn/natural-language-processing

Network automation

Network automation is the process of automating the configuration, management, testing, deployment, and operations of physical and virtual devices within a network.

• https://www.juniper.net/us/en/products-services/whatis/network-automation/
• https://www.cisco.com/c/en/us/solutions/automation/networkautomation.html
• https://www.netsync.com/practices/service-provider/networkautomation/

Network controller

Functional block that centralizes some or all of the control and management functionality of a network domain and may provide an abstract view of its domain to other functional blocks via well-defined interfaces.

• ETSI GS NFV 003 V1.4.1 (2018-08), Network Functions Virtualisation (NFV); Terminology for Main Concepts in NFV

• (https://www.etsi.org/deliver/etsi_gs/

• NFV/001_099/

• 003/01.

• 04.01_60/gs_

• nfv003v010401p.pdf)

Network function (NF)

Network Function (NF) – a functional building block within a network infrastructure, which has well-defined external interfaces and a well-defined functional behaviour.

• ETSI GS ZSM 007 V1.1.1 (2019-08): Zero-touch network and Service Management (ZSM); Terminology for concepts in ZSM (https://www.etsi.org/deliver/etsi_gs/ZSM/001_099/007/01.01.01_60/gs_ZSM007v010101p.pdf)

Network function disaggregation (NFD)

Defines the evolution of switching and routing appliances from proprietary, closed hardware and software sourced from a single vendor, towards totally decoupled, open components which are combined to form a complete switching and routing device.

• https://drivenets.com/blog/network-disaggregation-101/

A three-level application of automation capabilities (i.e., full automated infrastructure management, data centre infrastructure management and traceable/intelligent patch cords), including those enabled by integrating artificial intelligence techniques in the network.

• Telecommunication Standardisation Sector of ITU (ITU-T) Recommendation Y.3173 (02/2020) Series Y: Global Information Infrastructure, Internet Protocol Aspects, Next-Generation Networks, Internet of Things and Smart Cities – Future networks: Framework for evaluating intelligence levels of fufure networks including IMT-2020 network

• (https://www.itu.int/rec/dologin_pub.asp?lang=e&id=T-REC-

• Y.

• 3173-

• 202002-I!!PDF-E&type=items)

Network namespaces

Network namespaces is a virtualization mechanism (a virtualised networking stack) which provides abstraction and virtualisation of network protocol services and interfaces. Each network namespace has its own network device instances that can be configured with individual network addresses.

• internal definition

• based on https://

• subscription.

• packtpub.

• com/

• book/

• application_development/9781785883057/1/ch01lvl1sec15/namespaces-and-cgroups

Network orchestration

Network orchestration is the execution of the operational and functional processes involved in designing, creating, and delivering an end-to-end service. For example, it uses network automation to provide services through the use of applications that drive the network. An orchestrator functions to arrange and organise the various components involved in delivering a network service.

• internal definition based on: Ciena,

• https://www.

• ciena.

• com/

• insights/what-is/what-is-service-orchestration.html
  

Network resource

Physical or logical network component of hardware, software or data in the data, control or management planes within an organization's infrastructure.

• internal definition

Network service

A collection of network functions with a well specified behavior (i.e. content delivery networks (CDNs) and IP multimedia subsystem (IMS)).

• internal definition based on ITU-T - REC-Y.3515-201707: SERIES Y: GLOBAL INFORMATION INFRASTRUCTURE, INTERNET PROTOCOL ASPECTS, NEXT-GENERATION NETWORKS, INTERNET OF THINGS AND SMART CITIES, Cloud Computing – Functional architecture of Network as a Service;  (https://www.itu.int/rec/dologin_pub.asp?lang=

• e&id=T-REC-Y.

• 3515-

• 201707-I!!PDF-E&type=items)

Network Service Meshes

A network service mesh is intended to support application-to-application and function-to-function communications in networks and scenarios through dynamic and automated virtual network services – to be allocated on-demand, based on application requirements. Additionally, a service mesh is a software layer that handles all communication between services in applications. This layer is composed of containerized microservices.

• Reference(s) or Source: https://aws.amazon.com/what-is/service-mesh/#:~:text=service%20mesh%20requirements%3F-,What%20is%20a%20service%20mesh%3F,the%20performance%20of%20the%20services

Network slice instance

A network slice instance is a set of network function instances and the required resources (e.g., compute, storage and networking resources) which form a deployed network slice.

• Reference(s) or Source: based on TM Forum Reference Document, “TMF071 ODA Terminology”

• , Release 19.0.1, October 2019

• [TMF071] and the 3rd Generation Partnership Project (3GPP) Technical Specification (TS) 23.501, System architecture for the 5G System (5GS) (https://portal.3gpp.org/desktopmodules/Specifications/SpecificationDetails.aspx?specificationId=3144)

Network slicing

Network slicing is a specific form of virtualisation that allows multiple logical networks to run on top of a shared physical network infrastructure. (..) The intent of network slicing is to be able to partition the physical network at an end-to-end level to allow optimum grouping of traffic, isolation from other tenants, and configuring of resources at a macro level.

• https://www.

• idginsiderpro.

• com/article/

• 3231244/what-is-the-difference-between-network-slicing-and-quality-of-service.html and https://www.samenacouncil.org/thought-leadership-read?id=151, https://www.

• idginsiderpro.

• com/article/

• 3231244/what-is-the-difference-between-network-slicing-and-quality-of-service.html

Neural Network

Neural networks, also known as artificial neural networks (ANNs) or simulated neural networks (SNNs), are a subset of machine learning and are at the heart of deep learning algorithms. Their name and structure are inspired by the human brain, mimicking the way that biological neurons signal to one another. Artificial neural networks (ANNs) consist of multiple layers: an input layer, one or more hidden layers, and an output layer, all organized within a node structure. Each node, or artificial neuron, connects to another and has an associated weight and threshold. If the output of any individual node is above the specified threshold value, that node is activated, sending data to the next layer of the network. Otherwise, no data is passed along to the next layer of the network.

• Reference(s) or Source:  https://www.ibm.com/topics/neural-networks

NFV

Network Function Virtualisation (NFV) is a network architecture concept that uses virtualization to classify entire classes of network node functions into building blocks that may connect, or chain together, to create communication services. More specifically, it is the deployment of software implementations of traditional network functions (e.g. load balancers, firewalls, office switches/routers) on virtualized infrastructure rather than on function-specific specialized hardware devices.

• based on Dijiang Huang, Huijun Wu, in Mobile Cloud Computing, 2018, (https://www.sciencedirect.com/topics/computer-science/network-function-virtualization)

NFV-MANO

(Network Functions Virtualisation Management and Orchestration)

Management and orchestration (MANO) is a key element of the ETSI network functions virtualization (NFV) architecture. MANO is an architectural framework that coordinates network resources for cloud-based applications and the lifecycle management of virtual network functions (VNFs) and network services. As such, it is crucial for ensuring rapid, reliable NFV deployments at scale. MANO includes the following components: the NFV orchestrator (NFVO), the VNF manager (VNFM), and the virtual infrastructure manager (VIM).

• https://www.

• adva.com/en/

• products/

• technology/what-is-nfv-mano

NFV-MANO Architectural Framework

(Network Functions Virtualisation Management and Orchestration Architectural Framework)

Collection of all functional blocks (including those in NFV-MANO category as well as others that interwork with NFV-MANO), data repositories used by these functional blocks, and reference points and interfaces through which these functional blocks exchange information for the purpose of managing and orchestrating NFV.

• ETSI GS NFV 003 V1.4.1 (2018-08), Network Functions Virtualisation (NFV); Terminology for Main Concepts in NFV (

• https://www.

• etsi.org/deliver/etsi_gs/NFV/001_099/003/01.04.01_60/gs_nfv003v010401p.pdf)

NFVO

(Network Functions Virtualisation Orchestrator)

Functional block that

manages the Network Service (NS) lifecycle and coordinates the management of NS lifecycle, VNF lifecycle (supported by the VNFM) and NFVI resources (supported by the VIM) to ensure an optimized allocation of the necessary resources and connectivity.

• ETSI GS NFV 003 V1.4.1 (2018-08), Network Functions Virtualisation (NFV); Terminology for Main Concepts in

• NFV (https://www.etsi.org/deliver/etsi_gs/NFV/001_099/003/01.04.01_60/gs_nfv003v010401p.pdf)

Omnichannel capabilities is a term used in e-commerce and retail to describe if a business has the capabilities to implement a strategy that aims to provide a seamless shopping experience across all channels, including in-store, mobile, and online.

• Reference(s) or Source: https://www.

• techtarget.

• com/

• searchcustomerexperience/

• definition/omnichannel

OpenFlow protocol

OpenFlow protocol is a protocol defined by the OpenFlow Switch Specification that allows separation of the network control plane by providing access to the forwarding plane.

• internal definition based on: OpenFlow Switch Specification - Open Networking Foundation

• https://www.

• opennetworking.org/

• wp-content/uploads/2014/10/openflow-switch-v1.5.1.pdf and https://www.opennetworking.org/sdn-definition/?nab=1

OpenFlow (standard)

OpenFlow is an open standard that enables you to control traffic and run experimental protocols in an existing network by using a remote controller. The OpenFlow components consist of a controller, an OpenFlow or OpenFlow-enabled switch, and the OpenFlow protocol.

• https://www.

• juniper.net/documentation/en_US/junos/topics/concept/junos-sdn-openflow-support-overview.html

OpenStack

Open source software for creating private and public clouds. OpenStack software controls large pools of compute, storage, and networking resources throughout a data center, managed through a dashboard or via the OpenStack API.

• https://www.openstack.org/

Open virtual network (OVN)

Open Virtual Network (OVN) is an Open vSwitch-based software-defined networking (SDN) solution for supplying network services to instances.

• https://access.redhat.com/documentation/en-us/red_hat_openstack_platform/13/html/networking_with_open_virtual_network/open_virtual_network_ovn

Open vSwitch (OVS)

Open source multilayer virtual switch that supports standard interfaces and protocols.

• based on https://www.openvswitch.org/

Operational domain

Scope of management delineated by an administrative and technological boundary.

• based on TM Forum Reference, TMF071 ODA Terminology, TMF071, Release 19.0.1, October 2019 

Orchestration (ONAP)

The arrangement, sequencing and automated implementation of tasks, rules and policies to coordinate logical and physical resources in order to meet a customer or on-demand request to create, modify or remove network or service resources.

• from: TM Forum Technical Specification, Terminology for Zero-touch Orchestration, Operations and Management, TMF071, Release 17.0.1, November 2017, version 0.4.1, IPR Mode RAND 
• (synonyms for the system performing the function: manager, coordinator)

Process automation refers to the usage of technology to automate complex processes. It typically has three functions: automating processes, centralising information, and reducing the requirement for input from people. It is designed to remove bottlenecks and reduce errors and data loss, all while increasing transparency, communication across departments, and processing speed.

• https://www.tibco.com/reference-center/what-is-process-automation 

In the context of machine learning, a 'raw model' typically refers to a model that has been trained on data without much preprocessing or feature engineering. It is a basic model without any fine-tuning or optimisation.

• Reference(s) or Source: https://www.sciencedirect.com/science/article/pii/S0952197622003049

Rectification Activation Function

Rectification is the process of using a rectifier activation function (also referred to as a Rectified Linear Unit or ReLU): Rectified linear units, allow for faster and effective training of deep neural architectures on large and complex datasets compared to sigmoid function or similar activation functions.

• Reference(s) or Source: https://www.ml-science.com/rectifier-activation-function

Recurrent Neural Network

RNN stands for Recurrent Neural Network. This is a type of artificial neural network that can process sequential data, recognise patterns, and predict the final output. This type of neural network is called recurrent because it can repeatedly perform the same task or operation on a sequence of inputs.

• Reference(s) or Source: https://hackernoon.com/what-is-an-rnn-recurrent-neural-network-in-deep-learning

Reinforcement learning, in the context of machine learning and artificial intelligence (AI), is a type of dynamic programming that trains algorithms using a system of reward and punishment.

• based on https://www.techopedia.com/definition/32055/reinforcement-learning-rl 

Resource Facing Services (RFS)

A logical capability that is packaged as part of a product offering by service providers to their customers, but which is not directly visible to and/or usable by those customers. The logical functionality can be derived from underlying network or information technology (i.e., MPLS capabilities provided as part of a router), or may be delivered or supplied by staff or contractors employed by the service provider.

• Reference(s) or Source: https://www.itu.int/rec/T-REC-E.480-200609-I

Resource slice

A grouping of physical or virtual (network, compute, storage) resources. A resource slice could be one of the components of Network Slice, however on its own does not represent fully a Network Slice.

• based on https://tools.ietf.org/id/draft-geng-netslices-architecture-01.html#rfc.section.1.2

Robotic Process Automation (RPA) is a type of AI; it is a software technology that allows people to configure robots to perform rules-based tasks. It can be particularly useful for processes with predictable and frequent interactions with multiple applications.

• based on TM Forum AI Fundamentals course [TMF_AIF] and TM Forum “AI and its pivotal role in transforming operations” report and webinar [TMF_AI] 

Security operations, also known as SecOps, refers to a business combining internal information security and IT operations practices to improve collaboration and reduce risks.

• Reference(s) or Source: https://www.cyberark.com/what-is/security-operations/

Self-learning AI

Self-learning models are AI models that, once deployed, can be optimised by training them on data that becomes more available over time. This process prevents engineers from having to begin building new AI models from scratch every single time they collect more data.

• Reference(s) or Source: https://www.monolithai.com/blog/what-is-a-self-learning-model

Software-defined networking (SDN)

A programmable network approach that supports the separation of control and forwarding planes via standardized interfaces. 

• IRTF, RFC 7426: Software-Defined Networking (SDN): Layers and Architecture Terminology, January 2015, https://tools.ietf.org/html/rfc7426

Self-configuration

A process by which computer systems or networks automatically adapt their own configuration of components without human direct intervention.

• based on https://www.igi-global.com/dictionary/aspect-oriented-self-configuring-p2p/26200 (retrieved Nov. 11, 2019) and ETSI GS ZSM 007 V1.1.1 (2019-08), “Zero-touch network and Service Management (ZSM); Terminology for concepts in ZSM” [ETSI_ZSM_007]

The term self-organising network comes from the mobile radio network industry and refers to automated planning, configuration, management, optimisation and healing of a network.

• based on https://www.celona.io/network-architecture/self-organizing-network 

Service access point

A Service Access Point is a kind of Resource Function (RF) that handles access into and out of another RF, such as an application RF or virtualized appliance RF.

• TM Forum Reference, TMF071 ODA Terminology, TMF071, Release 19.0.1, October 2019

Serverless Architecture

Serverless architecture is a cloud-computing execution model where the cloud provider dynamically manages the allocation of machine resources. Pricing is based on the actual amount of resources consumed by an application, rather than pre-purchased units of capacity. This architecture allows developers to build and run applications without managing the underlying infrastructure.

• Reference(s) or Source: https://aws.amazon.com/serverless/

Service chaining (NFV)

Network service chaining, also known as service function chaining (SFC) is a capability that uses software-defined networking (SDN) capabilities to create a service chain of connected network services (such as L4-7 like firewalls, network address translation [NAT], intrusion protection) and connects them in a virtual chain.  This capability can be used by network operators to set up suites or catalogs of connected services that enable the use of a single network connection for many services, with different characteristics.

• https://www.sdxcentral.com/networking/virtualization/definitions/what-is-network-service-chaining/

Single Source of Truth

A single source of truth can be defined as a centralized and authoritative data repository that serves as the definitive reference for all relevant information within an organization.

• Reference(s) or Source: https://www.kohezion.com/technology-excellence/single-source-of-truth

Software Engineering Governance or Software Governance is the set of structures, processes and policies by which the software development and deployment function within an organisation is directed and controlled to yield business values and to mitigate risk.

• https://www.cs.uoregon.edu/events/icse2009/images/postConf/TB-Governance-ICSE09.pdf

Software defined exchanges (SDX)

Software Defined IXP (SDX) is an internet exchange that utilizes SDN to do interdomain routing. In addition, SDX design incorporates high levels of programmability, open APIs, shared resources across multiple domains, dynamic provisioning, resource discovery, quick resource integration and configuration, and granulated control of resources.

• internal definition based on https://sdx.cs.princeton.edu/ and J. Mambretti, J. Chen, F. Yeh, Software-Defined Network Exchanges (SDXs): Architecture, services, capabilities, and foundation technologies, 2014 26th International Teletraffic Congress (ITC), DOI: 10.1109/ITC.2014.6932970.

Supervised learning, also known as supervised machine learning, is an approach to creating artificial intelligence (AI), where a computer algorithm is trained on input data that has been labelled for a particular output. The model is trained until it can detect the underlying patterns and relationships between the input and output labels, enabling it to yield accurate labelling results when presented with never-before-seen data.

Also: “Note 2 – Supervised machine learning and unsupervised machine learning are two examples of machine learning types.” From ITU Recommendation Y.3172 (06/19).

• based on https://www.techtarget.com/searchenterpriseai/definition/supervised-learning and https://www.ibm.com/cloud/learn/supervised-learning

Switch abstraction interface (SAI)

Definition of the API to provide a vendor-independent way of controlling forwarding elements, such as a switching ASIC, an NPU or a software switch in a uniform manner.

• Open Compute Project github page, https://github.com/opencomputeproject/SAI

Architecture of generic services and functions that provides a foundation on which more specific architectures and architectural components can be built.

• https://pubs.opengroup.org/architecture/togaf8-doc/arch/chap19.html (TOGAF™ stands for The Open Group Architecture Framework)

The Network Data Analytics Function (NWDAF)

A network function that collects data from various network functions, application functions, as well as operations, administration, and management (OAM) systems, and operational support systems.

Note: This term is frequently used in 5G architecture.

• Reference(s) or Source: https://www.linkedin.com/pulse/network-data-analytics-function-nwdaf-5g-mintu-kumar-chetry/

Training Data

The data that are used to try to fit the best combination of weights and biases to a machine learning algorithm to minimize a loss function over the prediction range.

• Reference(s) or Source: Based on “The Language of Trustworthy AI: An In-Depth Glossary of Terms (updated August 4, 2024)”  ((https://docs.google.com/spreadsheets/d/e/2PACX-1vTRBYglcOtgaMrdF11aFxfEY3EmB31zslYI4q2_7ZZ8z_1lKm7OHtF0t4xIsckuogNZ3hRZAaDQuv_K/pubhtml),  NIST(C3.ai_Model_Training)

Training model

A machine learning training model is a process in which a machine learning (ML) algorithm is fed with sufficient training data to learn from. ML models can be trained to benefit manufacturing processes in several ways. The result of the process is a trained model.

• Reference(s) or Source: https://oden.io/glossary/model-training/

Transfer Learning

A technique in machine learning in which an algorithm learns to perform one task, such as recognising cars, and then is used as the starting point for a second, different task such as recognising cats. By using the knowledge from the first task the model can learn more quickly and effectively on the second task.

• Reference(s) or Source: based on: https://www.geeksforgeeks.org/ml-introduction-to-transfer-learning/

Transformers

A procedure that modifies a dataset.

• Reference(s) or Source: “The Language of Trustworthy AI: An In-Depth Glossary of Terms (updated August 4, 2024)” (https://docs.google.com/spreadsheets/d/e/2PACX-1vTRBYglcOtgaMrdF11aFxfEY3EmB31zslYI4q2_7ZZ8z_1lKm7OHtF0t4xIsckuogNZ3hRZAaDQuv_K/pubhtml) NIST (AI_Fairness_360)

Unsupervised learning, also known as unsupervised machine learning, uses machine learning algorithms to analyse and cluster unlabelled datasets. These algorithms discover hidden patterns or data groupings without human intervention. Its ability to discover similarities and differences in information makes it the ideal solution for exploratory data analysis, cross-selling strategies for offering different products to customers, customer segmentation, and image recognition.

• https://www.ibm.com/cloud/learn/unsupervised-learning

User interface orchestration

User Interface Orchestration defines, formats and structures the sequence of user interfaces (UIs) needed for a process. For example, the orchestration of UI during a service request from customers.

• based on TM Forum Reference, TMF071 ODA Terminology, TMF071, Release 19.0.1, October 2019 and IG1167 R18.0 "ODA Functional Architecture" Reference R21

‘Validation data’ means data used for providing an evaluation of the trained AI system and for tuning its non-learnable parameters and its learning process, among other things, in order to prevent underfitting or overfitting; whereas the validation dataset is a separate dataset or part of the training dataset, either as a fixed or variable split.

• Reference(s) or Source: “Proposal for a regulation of the European Parliament and of the Council on harmonised rules on Artificial Intelligence (Artificial Intelligence Act) and amending certain Union Legislative Acts” (https://www.europarl.europa.eu/meetdocs/2014_2019/plmrep/COMMITTEES/CJ40/DV/2023/05-11/ConsolidatedCA_IMCOLIBE_AI_ACT_EN.pdf)

Vertical scaling

Vertical scaling (or scaling up) means that you scale by adding more power (CPU, RAM) to an existing machine.

• Reference(s) or Source: https://ibm.github.io/data-science-best-practices/scaling.html

Virtual content delivery network

A content delivery network using virtualisation technology that enables the allocation of virtual storage, virtual machines, and network resources according to providers' requirements in a dynamic and scalable manner.

• based on ITU-T F.743.4 TELECOMMUNICATION STANDARDIZATION SECTOR OF ITU (03/2017) SERIES F: NON-TELEPHONE TELECOMMUNICATION SERVICES Multimedia services, Functional requirements for virtual content delivery networks (https://www.itu.int/rec/dologin_pub.asp?lang=e&id=T-REC-F.743.4-201703-I!!PDF-E&type=items)

Virtual eXtensible Local Area Network (VXLAN)

Virtual eXtensible Local Area Network (VXLAN) enables the encapsulation of Ethernet frames inside UDP packets with a designated UDP destination port (4789). VXLAN allows users to overlay L2 networks on top of existing L3 networks. In the data center, it is commonly used to stretch an L2 network across multiple racks.

• https://github.com/Mellanox/mlxsw/wiki/Virtual-eXtensible-Local-Area-Network-(VXLAN)
• https://tools.ietf.org/html/rfc7348

Virtual routing and forwarding (VRF)

Virtual Routing and Forwarding is a layer 3 abstraction, which provides a separate routing table for each instance, usually this is done by adding some sort of VRFID to the routing table lookup.

• internal definition
• https://en.wikipedia.org/wiki/Virtual_routing_and_forwarding

Virtualisation

Abstraction of network or service objects to make them appear generic, i.e. disassociated from the underlying hardware implementation specifics.

• internal definition

Virtualised network function (VNF) - virtual network function

Virtual Network Function (VNF) is a network task written as software that can be provided in a virtualised manner (i.e. firewall, router, switch).

• internal definition based on https://www.sdxcentral.com/networking/nfv/definitions/virtual-network-function/
• https://www.webopedia.com/TERM/V/virtualized-network-function.html

The sequence of steps through which a piece of work passes

from initiation to completion.

• based on https://www.merriam-webster.com/dictionary/workflow

Workflow management

Workflow management (WFM) is a technology supporting the re-engineering of business and information processes. It involves: Defining workflows, (...) and providing for fast (re)design and (re)implementation of the processes as business needs and information systems change.

• D. Georgakopoulos, M. Hornick, A. Sheth, An Overview of Workflow Management: From Process Modeling to Workflow Automation Infrastructure, Distributed and Parallel Databases, 3, 119-153 (1995), http

• ://www.

• workflowpatterns.com/documentation/

• documents/workflow95.pdf.

Zero-touch provisioning (ZTP), or zero-touch enrolment, is the process of remotely provisioning large numbers of network devices such as switches, routers and mobile devices without having to manually program each one individually.

• Reference(s) or Source: https://en.wikipedia.org/wiki/Zero-touch_provisioning and https://www.techtarget.com/searchitoperations/definition/zero-touch-provisioning-ZTP