Challenge on Fine-Grained Sentiment Analysis within ESWC2016
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
README.md
task5First50.xml
task5VerbsAnnotated.txt

README.md

SSA2016

Challenge on Fine-Grained Sentiment Analysis within ESWC2016

14 June 2016 - Publication of results
Task #1

  1. Efstratios Sygkounas, Xianglei Li, Giuseppe Rizzo and Raphaël Troncy. The SentiME System at the SSA Challenge Task 1 - Prec: 0.85686 - Rec: 0.90541 - F-meas: 0.88046
  2. Emanuele Di Rosa and Alberto Durante. App2Check extension for Sentiment Analysis of Amazon Products Reviews - Prec: 0.82777 - Rec: 0.90789 - F-meas: 0.87142
  3. Giulio Petrucci and Mauro Dragoni. An Information Retrieval-based System For Multi-Domain Sentiment Analysis - Prec: 0.81837 - Rec: 0.89198 - F-meas: 0.85359
  4. Andi Rexha, Mark Kröll, Mauro Dragoni and Roman Kern. Exploiting Propositions for Opinion Mining - Prec: 0.50494 - Rec: 0.81665 - F-meas: 0.62403

Task #2

  1. Soufian Jebbara and Philipp Cimiano - Aspect-Based Sentiment Analysis Using a Two-Step Neural Network Architecture - P: 0.41471 - R: 0.45196 - F: 0.43253 - A: 0.87356
  2. Marco Federici and Mauro Dragoni - A Knowledge-based Approach For Aspect-Based Opinion Mining - P: 0.34820 - R: 0.35745 - F: 0.35276 - A: 0.84925
  3. Andi Rexha, Mark Kröll, Mauro Dragoni and Roman Kern - Exploiting Propositions for Opinion Mining - N/A

Most Innovative Appoach

  1. The system "Aspect-Based Sentiment Analysis Using a Two-Step Neural Network Architecture", contribution of Soufian Jebbara and Philipp Cimiano, is the winner of the most innovative approach task. The system combines, for the first time, the deep learning paradigm and semantic resources (SenticNet and WordNet) for extracting aspects from sentences and for inferring the polarity associated with each aspect. The polarity has been inferred by aggregating the values of opinion words associated with each aspect.

May 2016 - Test set released. They can be downloaded from this link

8 February 2016 - Training Set for Task #1 released. Please download it from this link or check the sentiment analysis initiative for more details

16 February 2016 - Training Set for Task #2 released. Please download it from this link or check the sentiment analysis initiative for more details

16 February 2016 - Training Set and list of verbs (about 30) for Task #5 released. Please download the training set from this link and the annotated verbs from here or check the sentiment analysis initiative for more details

Check the wiki page to see more details

Also, check the ESWC2016 Semantic Sentiment Analysis Workshop that is connected to the Challenge.