SentiBench - a benchmark comparison of state-of-the-practice sentiment analysis methods

Filipe Nunes Ribeiro, Matheus Araújo, Pollyanna Gonçalves, Marcos André Gonçalves, Fabrí­cio Benevenuto · 2016

Read SentiBench - a benchmark comparison of state-of-the-practice sentiment analysis methods by Filipe Nunes Ribeiro, Matheus Araújo, Pollyanna Gonçalves, Marcos André Gonçalves, Fabrí­cio Benevenuto (2016) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/openalex-w2471350540.txt.

Description

Filipe Nunes Ribeiro, Matheus Araújo, and Pollyanna Gonçalves, with Marcos André Gonçalves and Fabrício Benevenuto, published SentiBench in EPJ Data Science in 2016. Their paper tackles a basic problem in sentiment analysis, the computer task of judging whether a message sounds positive or negative. Many methods claim success, but no fair head-to-head test showed which one works best.

The team compared twenty-four widely used methods on eighteen labeled datasets. The sets covered social network posts, movie and product reviews, and comments on news articles. Performance varied a lot from one dataset to another, so no single method won everywhere. The authors released the code and data in an open benchmark system. That gives later researchers a shared yardstick for testing new sentiment tools.

Catalog record

CREATORFilipe Nunes Ribeiro, Matheus Araújo, Pollyanna Gonçalves, Marcos André Gonçalves, Fabrí­cio Benevenuto
YEAR2016
COLLECTIONPAPERS
FORMATTXT
FILE SIZE103 KB
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEOpenAlex / EPJ Data Science

Source: https://doi.org/10.1140/epjds/s13688-016-0085-1

Browse the full library · About the Tower of Babel