App: The Crystal Ball

Hi! Welcome to this very simple App called “The Crystal Ball”.

The Crystal Ball uses Natural Language Processing (NLP) techniques to predict the next word of a sentence.

Data: Twitter, Blogs, and News

The prediction model used in The Crystal Ball uses text from Twitter, Blogs, and News.

This way, the model can accurately predict the next word of sentences written in different contexts.

# https://d396qusza40orc.cloudfront.net/dsscapstone/dataset/Coursera-SwiftKey.zip
twitter <- readLines(file("en_US.twitter.txt", "r"))
blogs <- readLines(file("en_US.blogs.txt", "r"))
news <- readLines(file("en_US.news.txt", "r"))

Prediction model: Bi-, Tri-, Quad- grams

N-grams are defined as “a contiguous sequence of n items from a given sample of text or speech.” (Wikipedia)

The Crystal Ball uses Bi-, Tri-, and Quad- grams in an attempt to make the best prediction for you.

# Excerpt of the function for Quad-grams (3 words + 1 predicted word)
... if (length(cw) >= 3) {cw <- tail(cw,3) #Select last 3 words of customer's sentence
... head(quadgram[quadgram$unigram == cw[1] & quadgram$bigram == cw[2] & quadgram$trigram == cw[3], 4], 1) #Display the 3-word sentence

Instructions: The Crystal Ball

Now, go to The Crystal Ball and enjoy!

Type a short sentence and see what it predicts.

For example, try typing “How are”, and it will predict “How are you”.

I hope you enjoy!

Thank you!