Aspect-Based Sentiment Analysis

Overview

This software package performs aspect-based sentiment analysis. It can analyze documents and identify aspect targets, their aspect category and their relevance. For usage in live systems, it features a relevance classifier to filter irrelevant documents. JavaDoc documentation is available on the documentation page.

How to cite

The system description can be obtained from the original paper. To cite it, use the following reference:

Ruppert E., Kumar A., Biemann C. (2017): LT-ABSA: An extensible open-source system for document-level and aspect-based sentiment analysis. In: Proceedings of the GSCL GermEval Shared Task on Aspect-based Sentiment in Social Media Customer Feedback, pages 55–60, Berlin, Germany

Quickstart

• create a new project
• add the jitpack Maven dependency

	<repositories>
		<repository>
		    <id>jitpack.io</id>
		    <url>https://jitpack.io</url>
		</repository>
	</repositories>

• add the AB-Sentiment dependency

	<dependency>
	    <groupId>com.github.uhh-lt</groupId>
	    <artifactId>LT-ABSA</artifactId>
	    <version>-SNAPSHOT</version>
	</dependency>

• copy models into your project home
• adapt the configuration file with the correct paths
• create a Java class for analysis:

import uhh_lt.ABSA.ABSentiment.type.Result;
import uhh_lt.ABSA.ABSentiment.AbSentiment;

public class MyClass {

    public static void main(String[] args) {
        AbSentiment analyzer = new AbSentiment("congiguration.txt");

        Result result = analyzer.analyzeText("This is the input string");

        // get Sentiment of text
        System.out.println(result.getSentiment());
        System.out.println(result.getSentimentScore());

    }
}

• analyze aspects and sentiment :)

Licence

This software is published under the Apache Software Licence 2.0

Third party license

This software uses the jblas library. The copyright link is : https://github.com/mikiobraun/jblas/blob/e1de8249b28137fa94a79558ee90ff037fd7c47d/COPYING