Linguistics

Philosophy

Trigram

Support vector machine

Economics

Computer science

Natural language processing

Task (project management)

Mathematics

Management

Feature vector

Feature (linguistics)

SemEval

Artificial intelligence

Geometry

Programming language

Sentiment analysis

Word (group theory)

Negation

Machine learning

Bigram

This paper describes University of Leipzig’s approach to SemEval-2013 task 2B on Sentiment Analysis in Twitter: message polarity classification. Our system is designed to function as a baseline, to see what we can accomplish with well-understood and purely data-driven lexical features, simple generalizations as well as standard machine learning techniques: We use one-against-one Support Vector Machines with asymmetric cost factors and linear “kernels” as classifiers, word uni- and bigrams as features and additionally model negation of word uni- and bigrams in word n-gram feature space. We consider generalizations of URLs, user names, hash tags, repeated characters and expressions of laughter. Our method ranks 23 out of all 48 participating systems, achieving an averaged (positive, negative) F-Score of 0.5456 and an averaged (positive, negative, neutral) F-Score of 0.595, which is above median and average.

ASVUniOfLeipzig: Sentiment Analysis in Twitter using Data-driven Machine Learning Techniques