作者: Xiangyang Luo , Yaqiong Qiao , Chenliang Li , Jiangtao Ma , Yimin Liu
DOI: 10.1016/J.IPM.2020.102375
关键词:
摘要: Abstract Geographical locations of microblog users are essential for user profiling, event localization and target advertising, to name a few. Automatic identification from the corresponding microblogs has been explored decades. This paper presents survey existing studies on problem geolocalisation. Specifically, we present general framework geolocalisation users. In this framework, location-related information is presented along with overview microblogging platforms. Furthermore, main steps introduced in detail. The state-of-the-art methods categorized into three classes: text-based methods, network-based multiview-based methods. Moreover, advantages limitations analyzed theoretically experimentally. Survey findings conclude that superior as well Besides that, cannot capture user's home location change, resulting misjudged results. Also, how locate multiple social platforms relatively unaddressed. Hence, geolocalization across one problems deserve further research.