search

NTMC-Community / MatchZoo

Facilitating the design, comparison and sharing of deep text matching models.
Apache License 2.0
3.85k stars 900 forks source link

What kind of model and feature is suitable for this scenery? (deal with synonym) #113

Closed guotong1988 closed 6 years ago

guotong1988 commented 6 years ago

Train data: Who is Michael Jordan? and Who is Air Jordan? are one right match. The test data is Who is Jumpman Jordan? and Who is Air Jordan?

What feature should I add for training? Which model can capture the match info for this case?

Thank you！！！ @pl8787 @faneshion @yangliuy @HouJP @JiafengGuo

bwanglzu commented 6 years ago

@guotong1988 Can you clarify your question?

yangliuy commented 6 years ago

@guotong1988 I think your question is on training/testing of paraphrase identification. You test data should also contain a question pair, correct ? You can refer to the QuoraQP data in the readme file. PI is supported by MatchZoo.

guotong1988 commented 6 years ago

Thank you. I have edit the question~

bwanglzu commented 6 years ago

@guotong1988 @yangliuy 's answer make sense for you?

bwanglzu commented 6 years ago

Thanks for your question, I'll close it for now. If you have further questions, please feel free to re-open it or create a new issue. Best.