基于位置微博的地震灾情提取

Earthquake disaster information extraction based on location microblog

震后微博具有内容丰富,更新速度快等特点。研究了一种基于位置微博的地震应急灾情提取方法。介绍了位置微博的概念,论述了基于位置微博的地震灾情提取流程,包括:地震灾情位置微博抓取、处理、分析和展示等。研究了基于位置API的源位置微博抓取方法,阐述了基于粗格网与细格网的位置参数优化方法。通过构建位置微博地震灾情分类映射表,实现了位置微博灾情的归一化处理。在此基础上建立了位置微博地震灾情分级表,将位置微博灾情内容数值化,并通过应用反距离空间差值法,实现了点集位置微博灾情向面状灾情信息的转化。最后,以云南永善地震为实例,展示了位置微博在地震灾情提取中的应用。实验结果表明,位置微博在地震灾情提取与分析中有好的应用前景。

Post-earthquake microblogs have the characteristics with rich contents and high update speed. In the pa- per, an effective earthquake disaster information acquisition method based on location microblogs was presented. The concept of location microblog was introduced, and the earthquake disaster information extraction process was expounded based on it, which includes earthquake disaster condition location microblogs grasping, process, analy- sis and display. First, location microblog extraction method based on location API of microblog platform was stud- ied, and a location parameter optimization method based on coarse grid and fine grid was described. Then, an earthquake disaster information classification and mapping table for location microblogs was built, and the normali- zation process of location microblog disaster information was realized. A location microblog disaster information grading table was established to digitalize the microblog contents. By applying IDW （inverse distance weighted） method, the location microblog disaster information was converted from point set to polygon. Finally, Yongshan earthquake in Yunnan Province was selected as the case to display the application of location microblogs in earth- quake disaster extraction. Experimental results show that location microblog has a good application prospect in earthquake disaster information extraction and analysis.