Leveraging Wikipedia knowledge to classify multilingual biomedical documents
作者：
Highlights：
• We apply a Wikipedia bag-of-concepts representation to classify multilingual biomedical documents.
• We propose a technique to convert Wikipedia concepts to one language to another.
• We create a multilingual corpus about biomedical topics, named ML-UVigoMED.
• We evaluate our approach by conducting experiments in ML-UVigoMED corpus.
• We improve the performance of multilingual biomedical document's classifiers.
摘要
•We apply a Wikipedia bag-of-concepts representation to classify multilingual biomedical documents.•We propose a technique to convert Wikipedia concepts to one language to another.•We create a multilingual corpus about biomedical topics, named ML-UVigoMED.•We evaluate our approach by conducting experiments in ML-UVigoMED corpus.•We improve the performance of multilingual biomedical document's classifiers.
论文关键词：Biomedical document classification,Hybrid word-concept document representation,Multilingual text classification,Wikipedia-based bag of concepts document representation,Wikipedia Miner semantic annotator
论文评审过程：Received 11 August 2017, Revised 6 April 2018, Accepted 23 April 2018, Available online 3 May 2018, Version of Record 7 June 2018.
论文官网地址：https://doi.org/10.1016/j.artmed.2018.04.007
