Traffic emergency identification method and system based on social network platform information

Abstract

The invention discloses a traffic emergency identification method and system based on social network platform information. The method comprises the following steps: crawling, primarily screening and digitalizing social network platform information texts; segmenting the processed information into a training text and a test text; constructing word vector matrixes of the training text and the test text through Boolean weighting; and sending the constructed test text word vector matrix into an SVM classifier, then performing classification by using a classification model to obtain a binarization classification result, and further screening out traffic emergency information. The system comprises a data preprocessing module, a feature word selection module, a word vector matrix construction module and an event classification and recognition module. According to the invention, through mining of social network platform text information and processing and classification processing of a naturallanguage technology (NLP), a new information acquisition mode can be provided for traffic emergency identification so as to assist traffic operation and management related units and departments to realize traffic event monitoring.

Description

technical field [0001] The invention belongs to the field of intelligent transportation, and in particular relates to a method for identifying traffic emergencies based on social network platform information. Background technique [0002] The era of mobile Internet has given birth to a number of social networking platforms used by hundreds of millions of users, such as QQ, WeChat, Weibo, Facebook, Twitter and so on. Users share what they see and hear in real time through social networking platforms. It also contains information related to traffic emergencies. Although traditional traffic detection technologies, such as coils, microwave radar, and video, have more mature applications in traffic incident monitoring, with the continuous expansion of the expressway network, the detection range limitations of traditional technologies have given rise to the The need for updates and innovations in traffic incident monitoring technology. The massive amount of information shared b...

AI Technical Summary

Problems solved by technology

[0004] In the current monitoring methods of various traffic emergencies, the microblog information released by the public has not been fully exploited and utilized.

Images