Locality-adapted kernel densities of term co-occurrences for location prediction of tweets
作者：
Highlights：
• We estimate tweet locations using spatial co-occurrence patterns and kernel density estimators.
• Our method combines probability distributions of texts based on locality-adapted KDE settings.
• Our method does not use any language-specific setting, parameter tuning or data other than tweet text.
• We compare our method with neural networks, generative models and discriminative models.
• Evaluations using three different tweet sets indicate significant improvement in accuracy.
摘要
•We estimate tweet locations using spatial co-occurrence patterns and kernel density estimators.•Our method combines probability distributions of texts based on locality-adapted KDE settings.•Our method does not use any language-specific setting, parameter tuning or data other than tweet text.•We compare our method with neural networks, generative models and discriminative models.•Evaluations using three different tweet sets indicate significant improvement in accuracy.
论文关键词：Location prediction,Twitter,Kernel density estimation,Spatial point patterns
论文评审过程：Received 22 October 2018, Revised 29 January 2019, Accepted 15 February 2019, Available online 22 March 2019, Version of Record 22 March 2019.
论文官网地址：https://doi.org/10.1016/j.ipm.2019.02.013
