Natural Language Processing

A presentation to accompany the demo of my unofficially named Twitter sentiment analysis project, Morgan. It's like Klout for events! Discusses the use of the naïve bayes classifier! If you want to see more visit me on github.com/meganspeir.

Transcript

1. NATURAL LANGUAGE PROCESSING
2. > > I’M > > ___ ___ _____ _____ ___ _ _ > | \/ || ___| __ \ / _ \ | \ | | > | . . || |__ | | \// /_\ \| \| | > | |\/| || __|| | __ | _ || . ` | > | | | || |___| |_\ \| | | || |\ | > \_| |_/\____/ \____/\_| |_/\_| \_/ > > _____ ______ _____ _____ ______ > / ___|| ___ \| ___|_ _|| ___ \ > \ `--.|| |_/ /| |__ | | || |_/ / > `--. \| __/ | __| | | || / > /\__/ /| | | |___ _| |_|| |\ \ > \____/ \_| \____/ \___/ \_| \_| > > > findme@meganspeir.com > > > > >
3. MORGAN
4. PROBLEM? What did people really think about my event? Discover the truth with Morgan’s Twitter Sentiment Analysis
5. NAÏVE BAYES CLASSIFIER A Naïve Bayesian classifier assumes probabilities being combined are independent of each other. Not true considering python (the snake) and python (the language) mean different things given the context.
6. To make up for its naïveté, calculate the probability of a tweet being in a specific category by multiplying together all the probabilities
   of the individual words in a tweet. Love Hate Excited Bored 0 20 40 60 80 Negative Positive
7. EXAMPLES Positive: #FBGGD Negative: #McDstories Neutral: @jack
8. THE STACK Python, Flask, PostgreSQL, nginx, uWSGI, Twitter API Bootstrap, CSS, Javascript, D3.js
9. CHALLENGES The Twitter API Humans
10. RESOURCES You can do it too! ! http.www.laurentluce.com/posts/
 twitter-sentiment-analysis-using-python-and-nltk
11. > > > > _____ _ _ ___ _ _ _ __ > |_ _| | | | / _ \ | \ | || | / / > | | | |_| |/ /_\ \| \| || |/ / > | | | _ || _ || . ` || \ > | | | | | || | | || |\ || |\ \ > \_/ \_| |_/\_| |_/\_| \_/\_| \_/ > > __ _______ _ _ > \ \ / / _ | | | | > \ V /| | | | | | | > \ / | | | | | | | > | | \ \_/ / |_| | > \_/ \___/ \___/ > > > findme@meganspeir.com > > > > Copyright © 2013 Megan Speir. All rights reserved. >