This repository contains a Java-based implementation of Hutto and Gilbert's Python-based VADER: A Parsimonious Rule-based Model for Sentiment Analysis of Social Media Text, from:

VADER: A Parsimonious Rule-based Model for Sentiment Analysis of Social Media Text
(by C.J. Hutto and Eric Gilbert)
Eighth International Conference on Weblogs and Social Media (ICWSM-14). Ann Arbor, MI, June 2014.

Hutto and Gilbert's version can be found on GitHub here

The SentimentIntensityAnalyzer class has a static method, getDefaultAnalyzer(), that returns an instance of SentimentIntensityAnalyzer using the default VADER lexicon. You can use this object and its polarity_scores() method to get the polarity score for a given string.

NOTE: This code does not provide sentence identification or any other natural language utilities.

The polarity_scores() function returns a Java Map object with the positive, negative, neutral, and compound polarity scores.

The following example shows how to use this package in Scala, which is my primary language right now.

import edu.umd.hcil.vader.SentimentIntensityAnalyzer

val analyzer = SentimentIntensityAnalyzer.getDefaultAnalyzer

analyzer.polarity_scores("VADER is smart, handsome, and funny.").get("compound")

Here's a brief example on using this code to get the compound value for a file full of tweets

val source = scala.io.Source.fromFile("tweets.txt")
for ( line <- source.getLines ) {
    val score = analyzer.polarity_scores(line).get("compound")
    
    System.out.println(String.format("%s\t%s\n", line, score.toString))
}

I also relied heavily on Chris Humphreys's p2j Python-to-Java translation toolkit to handle a lot of the mundane translations.