Information processing method and device based on deep learning and computer storage medium

Abstract

The embodiment of the invention provides an information processing method and device based on deep learning and a computer storage medium. The method comprises: obtaining keyword entity information ofto-be-searched information according to an information extraction model; determining the relevancy between the information to be searched and a target search term according to the relative position relationship between the keyword entity information of the information to be searched and the target search term in a multi-dimensional vector space; determining the influence degree of the to-be-searched information on the target search term according to an emotion model; sorting the to-be-searched information according to the relevancy and the influence degree to obtain the to-be-searched information with the maximum relevancy and influence degree corresponding to the target search term. According to the method, massive information can be quickly, accurately and reliably filtered, picked, understood and matched.

Description

technical field [0001] The present application relates to the field of deep learning, in particular to an information processing method, device and computer storage medium based on deep learning. Background technique [0002] With the popularization and development of the Internet and mobile Internet, more and more information is disclosed and fermented on the Internet. It is particularly important to obtain valuable information for enterprises by detecting the massive amount of data on the Internet. For example, through this information, enterprises can find Credit data and corporate public opinion. [0003] For Internet data information monitoring, the current solution adopted in the existing technology is to obtain effective data through manual analysis or keyword search after a large amount of Internet data is captured, such as the current mainstream information monitoring platforms such as Sina Public Opinion and Enterprise Early Warning Generally, these platforms can ...

AI Technical Summary

Problems solved by technology

[0005] 1. Filtering based on keywords, low accuracy and poor coverage

Chinese is extensive and profound, and it is difficult to cover all the content that needs to be collected by setting keywords. There are endless keyword descriptions for the target content, and the role of keywords in the sentence will cause ambiguity. For example, the target keyword is "Jingdong", the actual content It may be that "in Dongcheng District of Beijing" it can be seen that this situation will reduc

Images