EDITS is a software package aimed at recognizing entailment relations between two portions of text, termed as T and H. The system is based on edit distance algorithms, and computes the T-H distance as the cost of the edit operations (i.e. insertion, deletion and substitution) that are necessary to transform T into H. EDITS requires that three modules are defined: an edit distance algorithm, a cost scheme for the three edit operations, and a set of rules expressing either entailment or contradiction. Each module can be easily configured by the user as well as the system parameters. EDITS can work at different levels of complexity, depending on the linguistic analysis carried on over T and H. Both linguistic processors and semantic resources that are available to the user can be integrated within EDITS, resulting in a flexible, modular and extensible approach to Textual Entailment.
Version used for the FBK-irst participation in the TAC-RTE6 evaluation campaign. To replicate our results you can find the obtained models in the share/models folder or download them from:
Functionality extended to test. Note that the does not maintain the original order of the pairs.
The user can now use our implementation of the Rouge summarization evaluation algorithms described in:
Chin-Yew Lin. ROUGE: A Package For Automatic Evaluation Of Summaries Workshop On Text Summarization Branches Out , ACL 2004
All algorithm that process tokens (the only one that works on trees is tree edit distance) from version 2.1 are adapted to work with the edges of the syntactic tree.