search

shangjingbo1226 / AutoNER

Learning Named Entity Tagger from Domain-Specific Dictionary
https://shangjingbo1226.github.io/AutoNER/
Apache License 2.0
483 stars 91 forks source link

How does Fuzzy CRF work during decoding? #45

Open unikcc opened 2 years ago

unikcc commented 2 years ago

Hi, I'm just curious about the decoding process of Fuzzy CRF. The paper said "For inference, we apply the Viterbi algorithm to maximize the score." But the original Viterbi algorithm only select one path with the highest score, In the multi-label classification problem, there may be multi valid path, so how to decide the number of the paths.. Is there any threshold or something other in the decoding process. Since unfamiliar with C++, I cant find the decoding process in the repository. Looking forward your answer, thank you!

hadifar commented 2 years ago

Any update on this?

unikcc commented 2 years ago

@hadifar Oh, I made a mistake. The author proposed Fuzzy-CRF but didn't aims to solve multi-label sequence labeling problem, the final output of the Fuzzy-CRF still have only one path. This answer on stackoverflow misled me. The so-called fuzzy-CRF can't handle multi-label sequence labeling and is not specifically proposed to solve this problem