Why pay more? A simple and efficient named entity recognition system for tweets
作者：
Highlights：
• This paper investigates the problem of named entity recognition from tweets.
• We have combined deep learning models with traditional hand-crafted features.
• Output of bidirectional long short term memory is combined with extracted features.
• Finally, the conditional random field is applied to perform the task.
• Results claim, our system greatly outdo the performance of the multi-modal system.
摘要
•This paper investigates the problem of named entity recognition from tweets.•We have combined deep learning models with traditional hand-crafted features.•Output of bidirectional long short term memory is combined with extracted features.•Finally, the conditional random field is applied to perform the task.•Results claim, our system greatly outdo the performance of the multi-modal system.
论文关键词：Conditional random field,Hand-crafted feature,Long short term memory,Named entity recognition,Sequence labelling
论文评审过程：Received 15 March 2020, Revised 3 September 2020, Accepted 4 October 2020, Available online 15 October 2020, Version of Record 10 February 2021.
论文官网地址：https://doi.org/10.1016/j.eswa.2020.114101
