Batch-based active learning: Application to social media data for crisis management
作者：
Highlights：
• Boundary-based selection criterion to identify uncertain data instances.
• Uncertain data instances will be queried to enhance the accuracy of the classifier.
• Different selection strategies to find the instances to be labelled.
• Extensive experimental evaluation using different datasets.
摘要
•Boundary-based selection criterion to identify uncertain data instances.•Uncertain data instances will be queried to enhance the accuracy of the classifier.•Different selection strategies to find the instances to be labelled.•Extensive experimental evaluation using different datasets.
论文关键词：Online learning,Active learning,Classification,Social media,Crisis management
论文评审过程：Received 25 April 2017, Revised 21 September 2017, Accepted 11 October 2017, Available online 13 October 2017, Version of Record 21 October 2017.
论文官网地址：https://doi.org/10.1016/j.eswa.2017.10.026
