Paraphrasing (computational linguistics): Difference between revisions

Revision as of 05:00, 30 January 2018

Revision as of 05:04, 30 January 2018

Line 42:

There are multiple methods that can be used to evaluate paraphrases. Since paraphrase recognition is simply a classification problem, most standard evaluations metrics such as [[accuracy]], [[f1 score]], or an [[receiver operating characteristic|ROC curve]] will do.

ParaphraseThe generation,evaluation similarlyof toparaphrase machine translation,generation has multiplesimilar factorsdifficulties thatas canthe affectevaluation itsof evaluation[[machine translation]]. Often the quality of a paraphrase is dependent upon its context, whether it is being used as a summary, and how it is generated among other factors. Additionally, a good paraphrase usually is lexically dissimilar from its source phrase.

The simplest method used to evaluate paraphrase generation would be through the use of human judges. Unfortunately, evaluation through human judges tends to be time consuming. Automated approaches to evaluation prove to be challenging as it is essentially a problem as difficult as paraphrase recognition.<ref name=needed>{{Citation needed}}</ref>