DidaxTo implements an unsupervised approach for discovering patterns, that will extract a domain-specific dictionary from product reviews. The approach utilizes opinion modifiers, sentiment consistency theories, polarity assignment graphs and pattern similarity metrics. Apart from extracting the dictionary of opinion words, the application allows the evaluation of the dictionary by means of a sentiment classification task on product reviews.
The DidaxTo approach is fully presented in the following paper publiced by the Knowledge and Information Systems Journal (KAIS).
Pantelis Agathangelou, Ioannis Katakis, Ioannis Koutoulakis, Fotios Kokkoras and Dimitrios Gunopulos. “Learning Patterns for Discovering Domain Oriented Opinion Words“, Knowledge and Information Systems Journal (Springer), 6, pp 1-33, 2017. DOI: 10.1007/s10115-017-1072-y
- DidaxTo (v.2.0.2) – 8.6MB zip
Includes the application, user’s guide and demo datasets. You must have the Java Runtime Environment (JRE) installed in order to run.
DidaxTo – Version History
- v.2.0.2 – 2016-12-10
[+] Evaluation Metrics Added (Precision, Recall, F-Score)
- v.2.0.1 – 2016-05-22: (First Public Release)
[+] Activated Sentiment Inheritance in Opinion Targets.
[+] Α stemmer was added for use in the pattern exploitation process.
- v.1.6.0: GUI and Operational improvements.
[+] “Pattern Utility Info” in the “Lexicon Extraction & Pattern Evaluation” tab.
[+] “Pattern Extraction Info” in the “Lexicon Extraction & Pattern Evaluation” tab.
Improvements in the “Domain Lexicon Evaluation” tab.
- v.1.0.0: Initial internal release.