Natural language processing (NLP)
Technology that enables digital systems to understand and produce human language in written and oral forms.
Natural language processing (NLP) has two related subfields: natural language understanding (NLU) and natural language generation (NLG).
Natural language processing comes from computer science, artificial intelligence, linguistics, and data science. In practice, machine learning and deep learning techniques are used. These techniques allow NLP to learn and improve its performance.
NLU analyzes the context of audio and text to understand the meaning of a sentence. This helps systems such as search engines determine your intent based on what you type or say.
NLG allows computers to generate meaningful words. Together, these technologies are used in chatbots and digital assistants. (The/Hus)
neural network
In computer science and AI, neural networks are often thought of as similar to deep learning technologies, depending on which history of AI you follow.
Similar to deep learning, a neural network (or net) is a type of AI model inspired by the human brain. They use large datasets to learn skills, analyze information, and develop decision-making power.
An AI neural network is a network of thousands or even millions of interconnected nodes (points in the network through which data flows and connects with other data).
Neural networks are often trained using manually labeled examples. For example, it is “fed” images of cars labeled as cars and learns what cars look like by detecting patterns or car-like features in those images.
A similar thing happens when you sign up for a new website. You will be asked to select all traffic lights, buses, bridges, crosswalks, and motorcycles in the series of images before proceeding. The software reCAPTCHA wants to verify that you're a human and stop bots from creating a ton of fake accounts. But each time we prove we're human, we're also training the bots to be more human-like, so the software has had to become more and more sophisticated. That's ironic.
Artificial neural networks are inspired by the brain, but they are very different from biological nervous systems. They have different architectures, learning mechanisms, energy efficiency, and the amount of information that can be integrated into a small network. According to many philosophers, artificial neural networks lack human reasoning, consciousness, and emotional capabilities. (fs/za)
coming soon:
no-code machine learning
source:
Machine Learning Glossary (Google) https://developers.google.com/machine-learning/glossary (accessed July 24, 2023)
NLP vs. NLU vs. NLG: The differences between three natural language processing concepts (IBM) https://www.ibm.com/blog/nlp-vs-nlu-vs-nlg-the-differences-between-three-natural – language -processing-concepts/ (accessed July 25, 2023)
Read the rest of DW's AI Dictionary.
A | B | C | D | E | Fu | G | H | Me | J | K | L | M | N | Oh | P | Q | R | S | Ta | U | V | W | Z
We look forward to hearing from you. Submit your comments and suggest entries. Also, if you think we missed something or got it wrong, please let us know if our AI lexicon helped you understand the technology better.
Written and edited by: Zulfikar Abany (za), Fred Schwaller (fs)
