During natural disasters, social media can provide real time or rapid disaster, perception information to help government managers carry out disaster response efforts efficiently. Therefore, it is of great significance to mine social media information accurately. In contrast to previous studies, this study proposes a multimodal data classification model for mining social media information. Using the model, the study employs Late Dirichlet Allocation (LDA) to identify subject information from multimodal data, then, the multimodal data is analyzed by bidirectional encoder representation from transformers (Bert) and visual geometry group 16 (Vgg-16). Text and image data are classified separately, resulting in real mining of topic information during disasters. This study uses Weibo data during the 2021 Henan heavy storm as the research object. Comparing the data with previous experiment results, this study proposes a model that can classify natural disaster topics more accurately. The accuracy of this study is 0.93. Compared with a topic-based event classification model KGE-MMSLDA, the accuracy of this study is improved by 12%. This study results in a real-time understanding of different themed natural disasters to help make informed decisions.

During natural disaster, social media can provide real-time or low delayed disaster perception information to help government managers carry out disaster response efforts efficiently,therefore, it is of great significance to mining social media information accurately. Distinguished from previous studies, this study proposed a multi-modal data classification model for mining social media information. This model conducted LDA(Late Dirichlet Allocation) to identify subject information from multi-modal data and then analyzed by Bert (bidirectional encoder representation from transformers) and Vgg-16 (visual geometry group 16), text data and image data were classified separately, resulting in real mining of topic information during disasters. This paper used Weibo data during the 2021 Henan heavy storm as the research object, comparing with previous data experiment results, this study proposed a model that can make natural disaster related topic classification of social media data more accurately, resulting in a real-time understanding of different themed natural disasters, it was helpful in making disastrous decisions.