Paraphrasing (computational linguistics): Difference between revisions

Revision as of 10:15, 20 August 2021

Revision as of 11:29, 21 August 2021

Line 38:

Since paraphrases carry the same semantic meaning between one another, they should have similar skip-thought vectors. Thus a simple [[logistic regression]] can be trained to a good performance with the absolute difference and component-wise product of two skip-thought vectors as input.

== Sample Paraphrasing Softwares ==

* Free example: Paraphrser: Reword your text<ref>{{cite web |title=Paraphraser free app |url=https://paraphraser.tech/ |website=Paraphraser: reword your text}}</ref>

* Paid example: QuillBot<ref>{{cite web |title=QuillBot paid services |url=https://quillbot.com/ |website=QuillBot}}</ref>

== Evaluation ==

Line 61 ⟶ 57:

* [https://www.microsoft.com/en-us/download/details.aspx?id=52398 Microsoft Research Paraphrase Corpus] - a dataset consisting of 5800 pairs of sentences extracted from news articles annotated to note whether a pair captures semantic equivalence

* [http://paraphrase.org/#/ Paraphrase Database (PPDB)] - A searchable database containing millions of paraphrases in 16 different languages

* [https://paraphraser.tech/ Paraphraser: reword your text] - An example of a free app to do the paraphrasing

[[Category:Computational linguistics]]