NLU — Natural Language Understanding

Sub-bidang NLP yang fokus pada pemahaman bahasa — bukan hanya menghasilkan. Meliputi intent recognition, entity extraction, sentiment, semantic parsing.

NLU: pemahaman bahasa, bukan generasi. Intent classification, NER, semantic parsing, QA. Modern NLU: BERT, T5, GPT-3.5-turbo (understanding vs generation).

Also known as: pemahaman bahasa alami
Print

NLU

Definisi

NLU (Natural Language Understanding) adalah sub-bidang NLP yang fokus pada pemahaman bahasa — interpretasi makna, bukan generasi teks.

Tugas

  • Intent classification — apa yang user mau?
  • Named Entity Recognition (NER) — siapa, di mana, kapan?
  • Sentiment analysis — positif, negatif, netral?
  • Semantic parsing — konversi ke formal representation
  • Question answering — ekstrak jawaban dari konteks
  • Coreference resolution — “dia” merujuk ke siapa?
  • Relation extraction — hubungan antar entitas

Era

  • Rule-based (1990-an) — hand-crafted patterns
  • Statistical (2000-2014) — CRF, SVM
  • Neural (2014-2017) — RNN, LSTM, attention
  • Transformer (2017+) — BERT, RoBERTa, T5
  • LLM (2020+) — zero/few-shot NLU

Aplikasi

  • Voice assistants (Siri, Alexa, Google Assistant)
  • Chatbots customer service
  • Search engines
  • Email (auto-classification, smart compose)
  • Healthcare (clinical NLU dari rekam medis)

Connected to

Not yet written

The following pages are referenced but don't exist yet — they'd make good future additions.

  • /concepts/nlp

References

  1. Wikipedia

Type at least 2 characters to search.

Press to navigate, to open, esc to close.