60 lines
2.2 KiB
Markdown
60 lines
2.2 KiB
Markdown
---
|
|
title: Spracovanie prirodzeného jazyka
|
|
published: true
|
|
taxonomy:
|
|
category: [project]
|
|
tag: [nlp]
|
|
author: Daniel Hladek
|
|
---
|
|
|
|
# Spracovanie prirodzeného jazyka
|
|
|
|
Odporúčané témy:
|
|
|
|
- [identifikácia pomenovaných entít](./named-entity) (named entity recognition)
|
|
- [dialógové systémy](./chatbot) (chatbot)
|
|
- strojový preklad (machine translation)
|
|
- oprava preklepov (spelling error correction, diacritics restoration)
|
|
- získavanie informácií (vyhľadávanie v texte)
|
|
- identifikácia viacslovných výrazov (chunking)
|
|
- anotácia textových dát
|
|
|
|
## Knihy
|
|
|
|
- [Krátky úvod do NLP](./mrakt) (prezentácia z predmetu Moderné rečové a komunikačné technológie 2019)
|
|
- Lane, Howard, Hopke: Natural language processing in action
|
|
- [Spacy Tutoriál](https://nlpforhackers.io/complete-guide-to-spacy/)
|
|
- [Dive into Python](https://diveintopython3.problemsolving.io/) [(česky)](http://diveintopython3.py.cz/index.html)
|
|
- [Natural Language Processing in Action](https://www.manning.com/books/natural-language-processing-in-action) [(git)](https://github.com/totalgood/nlpia)
|
|
- [Python Data Science Handbook](https://github.com/jakevdp/PythonDataScienceHandbook)
|
|
- [Speech and Language Processing 2rd edition](https://github.com/rain1024/slp2-pdf)
|
|
- [Speech and Language Processing 3rd edition](https://web.stanford.edu/~jurafsky/slp3/)
|
|
|
|
|
|
## Odporúčané nástroje
|
|
|
|
- Prostredie [Anaconda](https://www.anaconda.com/distribution/)
|
|
- Chatbot [RASA](https://rasa.com/)
|
|
- Knižnica [Spacy](https://spacy.io/)
|
|
- Anotácie [Prodigy](https://prodi.gy/)
|
|
- Získavanie informácií [Elasticsearch](https://www.elastic.co/products/elasticsearch)
|
|
- seq2seq Neurónové siete [Fairseq](https://github.com/pytorch/fairseq)
|
|
- webové aplikácie a REST [Flask](https://www.fullstackpython.com/flask.html)
|
|
- [Flair](https://github.com/zalandoresearch/flair)
|
|
- [Spacy](https://spacy.io/)
|
|
- [GloVe](https://nlp.stanford.edu/projects/glove/) word embedding
|
|
- [FastText](https://fasttext.cc/) word embedding
|
|
- [GenSim](https://radimrehurek.com/gensim/) (topic modelling)
|
|
|
|
|
|
Postup:
|
|
|
|
- Nainštaluj si Anacondu
|
|
- Prečítaj si materiály
|
|
- Vyber si zaujímavú úlohu a nástroj
|
|
- Vyskúšaj nástroj
|
|
- Napíš správu o riešení úlohy
|
|
|
|
|
|
|