Table 2.
Glossary of Technical Terms
| Model | Definition |
| Natural language processing | Often abbreviated as NLP, this is a branch of artificial intelligence that focuses on enabling computers to understand, interpret, and generate human language in a valuable and meaningful way |
| Deep learning | An advanced form of machine learning that involves training neural networks to recognize complex patterns within data through layers of interconnected nodes, where each layer extracts progressively more abstract features |
| Large language model | An advanced artificial intelligence tool that, having learned from analyzing massive amounts of text data, can generate human-like text based on the context provided |
| Transformer | This modeling architecture, which was first designed for text data, understands and generates language by comprehending multiple parts of text simultaneously, thereby improving language task performance |
| Attention | The attention mechanism is a component of a neural network that allows the model to focus on certain parts of the input data more than others, enhancing its ability to understand context and nuances in complex data-like language |
| Lemmatization and stemming | These techniques are used in natural language processing. Stemming is a method where words are reduced to their base or root form, often leading to grammatically incorrect roots, while lemmatization transforms words to their dictionary form, ensuring linguistic correctness |
| Autoregression | This is a concept in statistics where current values of a time series are predicted using previous values, serving as a fundamental approach for time-dependent data analysis |
| Feed-forward network | This is a type of neural network in which information passes from one layer (see: Deep Learning) to a subsequent layer. This contrasts with different types of models, some of which incorporate “loops” where data from subsequent layers is used as input to earlier layers |
| Reinforcement learning | A type of machine learning where an agent learns to make decisions by taking actions in an environment to maximize a reward signal, progressively improving its behavior based on trial and error |