Making Watson-speak elementary: A glossary for users of AI in HR

when discussing AI in HR circles.

Artificial intelligence is the broad term for simulating human intelligence in machines, especially computers. Most of the other terms are AI methods or technologies.

Cognitive computing is a close cousin of artificial intelligence, and it's the variant used by IBM's Watson.

Text mining (sometimes called text analytics) is analysis of the information in natural language text. It's one of the techniques developers of AI in HR software use to glean talent insights from unstructured data sources such as social media and résumés.

Machine learning is a type of AI that gives computers the ability to learn and adapt when exposed to new data. It's the most common approach to AI in HR. Machine learning is similar to data mining in that both attempt to identify patterns in data, but it goes a step further by using new information to change its own behavior and what it communicates to users. The Amazon and Netflix recommendation engines are well-known examples. Likewise, machine learning AI in HR applications can, for example, recommend courses based on employee search histories or tag candidates whose profiles match patterns in internal performance management data.

Natural language processing (NLP) defines a computer's ability to understand human speech. Making sense of often ambiguous speech -- as opposed to precise, highly structured programming languages -- presents a huge challenge in algorithm design, processing and bandwidth. The availability of scalable resources in the cloud is what makes voice recognition technology such as Amazon's Alexa and Apple's Siri possible. Applications of this method of AI in HR talent management processes include analysis of video interviews and communication with chatbots.

Neural networks are computer hardware or software systems meant to mimic the neurons in the brain. They support NLP and Facebook's facial recognition feature.