# What is BERT? Definition & Technical Mechanics

> **Source:** [SPIDITS AI AI Glossary](https://spidits.com/ai-glossary)  
> **Category:** Natural Language Processing  
> **Canonical Citation:** [https://spidits.com/ai-glossary/bert](https://spidits.com/ai-glossary/bert)

## Definition
BERT (Bidirectional Encoder Representations from Transformers) is a language model developed by Google in 2018. Unlike autoregressive models, BERT is bidirectional, looking at the words before and after a target word to understand its context.

## Architectural Mechanics & Deep Dive
BERT (Bidirectional Encoder Representations from Transformers) is a landmark NLP model open-sourced by Google. Unlike autoregressive, left-to-right models, BERT uses a bidirectional architecture to read text in both directions simultaneously. By pre-training on a Masked Language Modeling (MLM) task, BERT develops a deep, context-aware understanding of word semantics, making it highly effective for search retrieval, question answering, and sentiment analysis.

## Industry Applications & Real-World Use Cases
Search engine understanding, text classification, and entity extraction

## Frequently Asked Questions

### What is the main difference between BERT and GPT?
BERT is an encoder-only model designed to understand context bidirectionally. GPT is a decoder-only model designed to generate text autoregressively.

### What is masked language modeling in BERT?
A training method where words in a sentence are hidden (masked) and the model is trained to predict the hidden words based on surrounding context.

## Related Terminology
- [llm](https://spidits.com/ai-glossary/llm)
- [transformer](https://spidits.com/ai-glossary/transformer)
- [nlp](https://spidits.com/ai-glossary/nlp)

---
*From the SPIDITS AI Knowledge Base (336+ terms indexed). Explore full technical definitions: [https://spidits.com/ai-glossary](https://spidits.com/ai-glossary)*