Event identification method and event identification device

Abstract

The application proposes an event identification method and an event identification device. The method comprises the following steps: extracting the features of a to-be-identified event based on preset feature extraction rules; determining the similarity between historical events and the to-be-identified event by comparing the features of the historical events with the features of the to-be-identified event; and selecting a predetermined number of historical events according to the similarity, and determining the type of the to-be-identified event based on the types of the predetermined number of historical events selected. Automatic identification of event type is realized. The method and the device can deal with a large number of various types of events, and can be used to quickly and accurately identify the type of a to-be-identified event. There is no need for too much dependency on labor. The efficiency and accuracy are improved.

Description

technical field [0001] The embodiments of the present application relate to the field of communication technologies, and in particular, to an event identification method and device. Background technique [0002] In the existing risk event trial process, judgments are basically made manually based on certain experience and summaries, and the resulting problems are: [0003] (1) The efficiency of the trial is low. You can only judge whether there is a risk in the current event based on your past experience and ability. It is necessary to manually obtain detailed data from multiple corresponding information systems according to keywords, and go through a series of manual discrimination, correlation or calculation. Only then can the current event be judged [0004] (2) The accuracy rate of trial and identification is not high or the difference is large, especially for case trial personnel who are inexperienced and have not been practicing for a long time. [0005] (3) The tria...

AI Technical Summary

Problems solved by technology

[0003] (1) The efficiency of the trial is low. We can only judge whether there is a risk in the current event based on personal past experience and ability. It is necessary to manually obtain detailed data based on keywords from multiple corresponding information systems, and go through a series of manual discrimination, correlation or calculation. Only then can the current event be judged

[0004] (2) The accuracy of trial and identification is not high or the difference is large, especially for case trial personnel who are inexperienced and have not been practicing for a long time

[0005] (3) The trial method is too artificial and not intelligent. When judging an event, it is necessary to manually retrieve key information from multiple systems and manually compare the information one by one to make a judgment

[0006] (4) Similar events cannot be quickly matched, and cannot be quickly and effectively identified based on historical events and current events when dealing with a single case

[0007] It can be seen that in the prior art, events need to be manually judged, but the number of events is large and the forms of events are diverse, resulting in the judgment process being too dependent on manual labor, and the efficiency and accuracy are not high

Images