NLP /

OctWorkshop

Hub

https://colab.research.google.com/drive/1aKhgZrB27ZGU3qvUWBRFTDzdeoPkjS0K

General

https://towardsdatascience.com/turbo-charge-your-spacy-nlp-pipeline-551435b664ad

https://www.machinelearningplus.com/spacy-tutorial-nlp/

Topic modelling

https://github.com/mariask2/topics2themes

https://medium.com/mlreview/topic-modeling-with-scikit-learn-e80d33668730

https://www.machinelearningplus.com/nlp/topic-modeling-visualization-how-to-present-results-lda-models/

Sentiment analysis,POS/NER/Sequence labeling

https://notebook.community/nealcaren/workshop_2014/notebooks/3_Sentiment

https://sst5-explainer.herokuapp.com/result

(http://nlp.stanford.edu:8080/sentiment/rntnDemo.html)

https://stackoverflow.com/questions/34102420/pos-tagging-using-spacy

Word vectors

https://lamyiowce.github.io/word2viz/

Exercises

https://www.machinelearningplus.com/nlp/nlp-exercises/

Setup

Colab (Preferred, web-based) https://colab.research.google.com/notebooks/intro.ipynb (Google account required)

or

Anaconda (local install) https://www.anaconda.com/products/individual

Schedule

General

  • Intro round
  • Intro general NLP
Korp, CLARIN, Swe-Clarin
  • Intro Python and Spacy
wikipedia package
Advanced NLP with Spacy: <https://course.spacy.io/en>
  • Python/Spacy practice

Topic modelling

  • Intro topic modelling
  • Topic modelling vis
  • Topic modelling advanced
  • Topic modelling trends
  • Topic modelling command line
  • Topic modelling practice

Sentiment analysis,POS/NER/Sequence labeling

  • Intro Sentiment analysis,POS/NER/Sequence labeling
  • SPNS practice
  • Corpora creation/Labeling

Word vectors

  • Intro word vectors
  • Word vectors practice
  • Project discussion

Dear participant,

Here are some setup instructions for the workshop 28-29 Oct.

We will use Python in the workshop, particularly in the form of notebooks. There are different ways of running these.

Colab

The first is through Google colab. This is a web based python notebook, so you don't have to install anything. This requires a google account. Get one and head over to <https://colab.research.google.com/notebooks/intro.ipynb> to learn more about running and editing python code in them.

Anaconda

The other way is by downloading and installing a python environment. I prefer Anaconda <https://anaconda.org/>. You will then have to download the notebooks provided and run them through a local notebook server. There is excellent documentation on how to do this online, so I will not repeat it here.

You can also use the Spyder editor that comes with it. Spyder can not run notebooks so then you'll have to copy the code from them.

Preparatory exercise

If you have no previous experience at all of Python, notebooks and/or colab, it could be useful to run through this preparatory notebook:

https://colab.research.google.com/drive/1z403uzu0Xf0hcy6cgfcLII3jSMBISmq4

It will teach you the basics.

Kind regards and see you on Wed (I will send a Zoom link separately)