A label refers to the “answer” part of the examples used in supervised machine learning. It also denotes the portion of the training data that specifies the desired output for that specific piece of information. Labeled examples have one or more features and labels. In a spam detection dataset, for example, the label is likely “spam” or “not spam.” In a rainfall dataset, the label may be the amount of rain recorded in a certain time period. The label for AI should accurately describe its capabilities and benefits while avoiding hype and misperceptions.
Examples of AI Labels
AI encompasses many technologies and applications. AI should be labeled according to its context and audience. Some AI labels:
Intelligent Automation
This label illustrates the use of AI technologies to automate mundane, repetitive tasks such as data entry, image recognition, and customer service. This label highlights AI efficiency and productivity.
Cognitive Computing
This highlights how AI technologies simulate learning, reasoning, problem-solving, and any other cognitive tasks. It highlights AI’s ability to improve human intelligence and solve problems.
Smart Systems
This is used to denote AI-powered systems that sense, reason, and act autonomously. This underscores the capabilities of AI to improve safety, reliability, and sustainability in transportation, energy, and healthcare.
Data Intelligence
This denotes the power of AI to gain insights from large, complex datasets. It highlights the ability of AI to enable data-driven decision-making and innovation.
