About

Sentiment analysis applications are usually classified according to the granularity level they tackle: the document, sentence and aspect levels. In this work, we focus on aspect-based sentiment analysis, more specifically, on the aspect identification and clustering tasks, which is essential to perform richer text analyses and produce better results in the area.

In such context, we explore varied knowledge sources for aspect identification and clustering and propose the Oplcuster-PT, which significantly outperforms the other lexicon-based approaches. For more information, see Francielle A. Vargas MSc thesis.

The Opcluster-Pt is a tool for aspect-based sentiment analysis. The tool allows the automatic identification of opinion explicit and implicit aspects in product reviews. The proposed tool is also language and domain independent, nevertheless, it is currently customized for Portuguese. The system is available in the web and desktop version.

Developed by

Check similar projects for the Portuguese at OPINANDO