Gain insights into processing text data, creating word embeddings, and using LSTMs for semantic analysis and machine translation. Explore industry-relevant NLP techniques with Python and TensorFlow.

nlp_docker.tar.gz

mnist-fashion

In this course you'll learn techniques for processing text data, creating word embeddings, and using long short-term memory networks (LSTM) for tasks such as semantic analysis and machine translation. After completing this course, you will be able to solve the important day-to-day NLP problems faced in industry, which is incredibly useful given the prevalence of text data.

The code for this course is built around the TensorFlow framework, one of the premier frameworks for industry machine learning, and the Python pandas library for data analysis. Knowledge of Python and TensorFlow are prerequisites. 

This course was created by AdaptiLab, a company specializing in evaluating, sourcing, and upskilling enterprise machine learning talent. It is built in collaboration with industry machine learning experts from Google, Microsoft, Amazon, and Apple.

Natural Language Processing with Machine Learning

<h2 class="section-title">Chapter Goals:</h2>
<ul>
	<li>Learn about K-nearest neighbors in terms of word similarity</li>
	<li>Create a function that computes the K-nearest neighbors for a given word</li>
</ul>


<h2>A. Similar words</h2>

<div id="section1_l8" class="collapse show">
<p>When comparing cosine similarities for word embeddings, a common procedure is to find the <i>K-nearest neighbors</i> for a given word. This means that for a given a word <i>w</i> and an integer <i>K</i>, we can find the <i>K</i> vocabulary words whose embedding vectors have the highest cosine similarity to the embedding vector for <i>w</i>.</p>
<p>Using K-nearest neighbors can help us evaluate our embedding model and make sure it was trained properly. For example, if we notice that the K-nearest neighbors for any given word is always the same <i>K</i> words, then something may have gone wrong with our embedding training. Likewise, if the K-nearest neighbors are completely different from what we would expect to see (e.g. if the 3 nearest neighbors for the word "computer" are "waterfall", "ocean", and "soda"), then we may want to take a closer look at our model's embedding matrix.</p>
<p>When the embedding model is trained well, the K-nearest neighbors metric can provide some useful insights about the text corpus, particularly for specialized text corpora. For example, we normally expect the word "code" to be related to terms about computer programming or software. However, if our training corpus were related to military operations, the K-nearest neighbors for "code" might include words like "signal", "transmission", or "decode".</p>
</div>


<h2>Chapter Goals:</h2>
<ul>
	<li>Learn about K-nearest neighbors in terms of word similarity</li>
	<li>Create a function that computes the K-nearest neighbors for a given word</li>
</ul>


<h2>A. Similar words</h2>

<div id="section1_l8">
<p>When comparing cosine similarities for word embeddings, a common procedure is to find the <i>K-nearest neighbors</i> for a given word. This means that for a given a word <i>w</i> and an integer <i>K</i>, we can find the <i>K</i> vocabulary words whose embedding vectors have the highest cosine similarity to the embedding vector for <i>w</i>.</p>
<p>Using K-nearest neighbors can help us evaluate our embedding model and make sure it was trained properly. For example, if we notice that the K-nearest neighbors for any given word is always the same <i>K</i> words, then something may have gone wrong with our embedding training. Likewise, if the K-nearest neighbors are completely different from what we would expect to see (e.g. if the 3 nearest neighbors for the word "computer" are "waterfall", "ocean", and "soda"), then we may want to take a closer look at our model's embedding matrix.</p>
<p>When the embedding model is trained well, the K-nearest neighbors metric can provide some useful insights about the text corpus, particularly for specialized text corpora. For example, we normally expect the word "code" to be related to terms about computer programming or software. However, if our training corpus were related to military operations, the K-nearest neighbors for "code" might include words like "signal", "transmission", or "decode".</p>
</div>


In this chapter, we will be learning how to calculate K-nearest neighbors based on cosine similarity. 


What you'll learn from this course

Word Embeddings

Language Model

Text Classification

Seq2Seq Model

K-Nearest Neighbors

Chapter Goals:

A. Similar words