Tweester at SemEval-2017 Task 4: Fusion of Semantic-Affective and pairwise classification models for sentiment analysis in Twitter.
Athanasia KolovouFilippos KokkinosAris FergadisPinelopi PapalampidiElias IosifNikolaos MalandrakisElisavet PalogiannidiHaris PapageorgiouShrikanth S. NarayananAlexandros PotamianosPublished in: SemEval@ACL (2017)
Keyphrases
- sentiment analysis
- classification models
- pairwise
- social media content
- sentiment lexicon
- public opinion
- opinion mining
- micro blog
- sentiment classification
- text classification
- training data
- sentence level
- decision trees
- social media
- feature selection
- feature set
- semi supervised
- text mining
- semantic information
- social networks
- micro blogging
- product reviews
- real world
- online reviews
- natural language
- product features
- neural network
- feature subset
- user generated
- semantic similarity
- user generated content
- cross lingual
- online social networks
- transfer learning
- co occurrence
- classification accuracy
- similarity measure