Bays cross model-based government affair approval intelligent screening method

Abstract

The invention provides a bays cross model-based government affair approval intelligent screening method. The method comprises the following steps of S1, performing word segmentation on approval item information; S2, filtering stop word lists of a word segmentation result; S3, generating an approval item word bag model; S4, performing cross multiplication on the approval item word bag vector and the historical approval category word segmentation probability matrix of the same business domain item to obtain a probability list that the item belongs to the approval category; The historical examination and approval category word segmentation probability matrix executes the historical examination and approval item word bag model obtained in the step S1-step S3 through historical examination andapproval item information, and the historical examination and approval item word bag model and an examination and approval result category word bag are trained through a naive Bayesian model and a cross model and are obtained through a convergence result. The efficiency can be greatly improved, and the labor cost can be greatly reduced.

Description

technical field [0001] The invention relates to the fields of information science, big data processing, and Internet technology, and in particular to an intelligent screening method for government approval based on a bayes cross model, and an online screening algorithm based on approval content on a Paas service platform of a government form center. Background technique [0002] At present, in government services, most application scenarios have realized the electronic process of matters, and the business information and content to be approved are archived and stored in the government application database in the form of electronic forms. As for the examination and approval screening of accepted matters, it is still up to the case-handling personnel to analyze the specific circumstances of the matter based on the laws and regulations and approval experience to determine the approval result. [0003] The major drawbacks of the prior art are: [0004] For the examination and a...

AI Technical Summary

Problems solved by technology

[0006] In view of the large number of items and cases accumulated in historical approval, there is no effective plan to form an effective auxiliary app

Images