Sentiment analysis

using word2vec

What you have here is a project for CS50. Any of the text you enter here stays with your browser and only you can see what is entered. This is strictly for fun!

Recommendations for input are:

I'm feeling quite happy today!
I'm starting to get angry
I woke up feeling quite solemn

What's going on here

Using the word2vec programs, a dataset of word vectors was created. This data set was used to determine a small vocabulary of cosine similarity between the words Sad, Mad and Glad. Indefinite articles have been removed from the vector data set.

The data was then loaded into a memcached memory object caching system for reference in the javascript programs on this page. These javascript programs then utilize the memcache PHP extension to query the memcache dataset.

My contact information long.kevin (a) gmail

Technology used

HTML
PHP
Javascript
JQuery
Bootstrap
Memcached
word2vec
perl
c
gcc
Wikipedia

Wishlist

Certainty level [0-100]%
Sentiment polarity [-1,0,+1]
Emotional state [angry, sad, happy]
Objective | Subjective

Further reading

Sentiment Analysis and Subjectivity
Wikipedia: Sentiment analysis
Word to vector
Efficient Estimation of Word Representations in Vector Space