An Event Early Warning Method Based on Knowledge Base and Semantic Fingerprint

The prevalence of the Internet and social networks has led to the low marginal cost of information replication, the widespread existence of duplicate content and the problem of “information disorientation”. Early detection of events helps relevant departments intervene and make decisions as soon as possible, thereby reducing the social harm cause by the aforementioned issues. Therefore, this paper proposes an event early warning method based on knowledge base and semantic fingerprint. Firstly, based on the idea of data mining, a classified event feature knowledge base is constructed. Next, using Simhash algorithm, a text deduplication method based on semantic fingerprints is proposed, which realizes the noise reduction and simplification of public opinion, and obtains the document feature value after deduplication. Finally, the document features are matched with the features in the knowledge base, realizing event classification and event early warning, facilitating relevant departments to take corresponding measures and reduce social impact.