By far the most useful recent advancement in Natural Language Processing (NLP) has been: word vector embeddings (also known as word vectors, people also reference it as word2vec which is one of the most popular implementations of word vector embeddings).
Word vectors allow you to do very nifty things like compare the underlying meaning of words like "car" and "boat", and "data" and "information". Sounds like magic? Come along and we'll explain how they work and how to use them to build a simple "You may also want to read ..." system that finds related articles using word vectors!
• We will start with a short talk about word vectors
• We will then show a live walkthrough of using them
• Then it's all hands on with the code
• Then Pizza!
We aim to get everyone to the level where they understand a bit about what word vectors are and how they can be used.
For this lesson, we will mainly be using Kaggle kernels so make sure you have already created a Kaggle account.
You can also use Jupyter with Annaconda Navigator if you prefer (https://www.anaconda.com/download/) installed with Python 3.
As usual, this meetup will be very relaxed and we encourage people of all skills in python to come along and learn from each other. We'll work on the problem for a few hours then have some pizza and drinks.
* Laptops, Charging Cords, an Inquisitive Spirit
HOW TO FIND IT
Please head to Norfolk place (W2 1PE - see photos), and it's at the faculty of medicine. The entrance (see photos) is opposite the Cambridge wing and might be behind the mobile MRI scanner (it moves so it might not!). On the day go in and speak to the person on the desk and tell them that you are here for the data science meetup, they will hopefully let you in and explain where to go.
Claim the event and start manage its content.I am the organizer