Projects per year
Chemical patents are an important resource for chemical information. However, few chemical Named Entity Recognition (NER) systems have been evaluated on patent documents, due in part to their structural and linguistic complexity. In this paper, we explore the NER performance of a BiLSTM-CRF model utilising pre-trained word embeddings, character level word representations and contextualized ELMo word representations for chemical patents. We compare word embeddings pre-trained on biomedical and chemical patent corpora. The effect of tokenizers optimized for the chemical domain on NER performance in chemical patents is also explored. The results on two patent corpora show that contextualized word representations generated from ELMo substantially improve chemical NER performance w.r.t. the current state-of-the-art. We also show that domain-speciﬁc resources such as word embeddings trained on chemical patents and chemical-speciﬁc tokenizers have a positive impact on NER performance.
|Original language||American English|
|Title of host publication||Improving Chemical Named Entity Recognition in Patents with Contextualized Word Embeddings|
|State||Published - 2019|
|Event||18th ACL Workshop on Biomedical Natural Language Processing: BioNLP @ ACL 2019 - |
Duration: Aug 1 2019 → …
|Workshop||18th ACL Workshop on Biomedical Natural Language Processing|
|Period||08/1/19 → …|
FingerprintDive into the research topics of 'Improving Chemical Named Entity Recognition in Patents with Contextualized Word Embeddings'. Together they form a unique fingerprint.
Information Extraction for Chemical Reactions with University of Melbourne
Verspoor, K., Akhondi, S., Tsatsaronis, G., Drukenbrodt, C., Thorne, C., Hoessel, R., Cohn, T., Nguyen, D. Q., Zhai, Z., Fang, B., Yohsikawa, H. & Doornenbal, M.
06/1/18 → 12/31/22
ChEMU: Cheminformatics Elsevier Melbourne University lab
05/15/20 → 09/30/20