CN-121278610-B - Application classification method, device, electronic equipment and storage medium

CN121278610BCN 121278610 BCN121278610 BCN 121278610BCN-121278610-B

Abstract

The embodiment of the specification provides an application classification method, an application classification device, electronic equipment and a storage medium, wherein the method comprises the following steps: the method comprises the steps of obtaining a classification evaluation request aiming at a target application, generating a search instruction corresponding to the classification evaluation request based on a planning model, collecting relevant data of the target application from a public channel based on the search instruction, classifying and evaluating the target application based on the relevant data by using a classification decision model, outputting a classification evaluation result of the target application, verifying the classification evaluation result by using verification models to obtain a classification verification result of the target application, and determining a classification result of a final target application according to the classification verification result of each verification model.

Inventors

WANG GUANNAN
FENG ZIHAO
CHENG WEIHONG
WANG YUQIONG

Assignees

钱塘征信有限公司

Dates

Publication Date: 20260505
Application Date: 20251205

Claims (10)

1. An application classification method, the method comprising: Acquiring a classification evaluation request aiming at a target application, and generating a search instruction corresponding to the classification evaluation request based on a planning model; collecting related data of the target application according to the search instruction based on a search model; Classifying and evaluating the target application based on the related data by adopting a classifying decision model, and outputting a classifying and evaluating result of the target application; Acquiring a first classification verification result generated by each verification model aiming at the classification evaluation result; Respectively counting the first quantity of the first classification verification results of each type; Determining a first classification verification result, in which the ratio of the first number to the total number of verification models exceeds a set threshold, as a classification result of the target application; If the ratio of the first quantity to the total number of verification models is smaller than the set threshold value, acquiring the supplementary data of the target application based on the search model; generating a second classification verification result of the classification evaluation result based on the supplemental data using the verification model; respectively counting the second quantity of the second classification verification results of each type; And determining a second classification verification result of which the ratio of the second quantity to the total number of verification models exceeds the set threshold value as a classification result of the target application.
2. The method of claim 1, the obtaining a classification evaluation request for a target application, generating search instructions corresponding to the classification evaluation request based on a planning model, comprising: acquiring a classification evaluation request aiming at a target application, and generating a search instruction set corresponding to the classification evaluation request based on depth-first search, wherein the classification evaluation request comprises a task description text aiming at the target application; Screening and obtaining a target search instruction from the search instruction set based on a predefined rule, and taking the target search instruction as a search instruction corresponding to the classification evaluation request.
3. The method of claim 1, wherein the acquiring the relevant data of the target application according to the search instruction based on the search model comprises: collecting related data of the target application according to the search instruction based on a search model; Acquiring text evaluation parameters of the related data, wherein the text evaluation parameters comprise text information entropy, text length, keyword number and whether advertisement content is contained or not; And screening the related data based on the text evaluation parameters to obtain screened related data.
4. The method of claim 1, wherein the classifying the target application based on the related data using a classification decision model, outputting a classification evaluation result of the target application, comprises: converting the related data into vector data based on a vector generation model; and carrying out classification evaluation on the target application based on the vector data by adopting a classification decision model, and outputting a classification evaluation result of the target application.
5. The method of claim 4, wherein the classifying the target application based on the vector data using a classification decision model, outputting a classification evaluation result of the target application, comprises: determining the similarity between the vector data and each preset class vector by adopting a classification decision model; and determining a target preset category similar to the target application based on the similarity, and taking the target preset category as a classification evaluation result of the target application.
6. The method of claim 1, wherein the classification evaluation request is a risk classification request, the classifying the target application based on the relevant data using a classification decision model, outputting a classification evaluation result of the target application, comprising: determining fund transaction characteristics, privacy characteristics and functional risk characteristics of the target application from the related data by adopting a classification decision model; Determining a risk score for the target application based on the funds transaction feature, the privacy feature, and the functional risk feature; And determining a classification evaluation result of the target application based on the risk score and a score threshold.
7. An application classification apparatus comprising: the planning unit is used for acquiring a classification evaluation request aiming at the target application and generating a search instruction corresponding to the classification evaluation request based on a planning model; the searching unit is used for collecting relevant data of the target application according to the searching instruction based on a searching model; The decision unit is used for carrying out classification evaluation on the target application based on the related data by adopting a classification decision model and outputting a classification evaluation result of the target application; The system comprises a verification unit, a first classification verification unit, a second classification verification unit and a second classification verification unit, wherein the verification unit is used for acquiring first classification verification results generated by each verification model aiming at the classification evaluation results, counting first quantity of the first classification verification results of each type respectively, determining the first classification verification results of which the ratio of the first quantity to the total number of the verification models exceeds a set threshold as classification results of the target application, acquiring supplementary data of the target application based on the search model if the ratio of the first quantity to the total number of the verification models is smaller than the set threshold, generating second classification verification results of the classification evaluation results based on the supplementary data by adopting the verification models, counting second quantity of the second classification verification results of each type respectively, and determining the second classification verification results of which the ratio of the second quantity to the total number of the verification models exceeds the set threshold as classification results of the target application.
8. An electronic device comprising a processor and a memory, wherein the memory stores a computer program adapted to be loaded by the processor and to perform the steps of the method according to any of claims 1 to 6.
9. A storage medium storing a computer program which, when executed by a processor, implements the steps of the method according to any one of claims 1 to 6.
10. A computer program product comprising a computer program which, when executed by a processor of an electronic device, causes the processor to perform the steps of the method according to any one of claims 1 to 6.

Description

Application classification method, device, electronic equipment and storage medium Technical Field The present disclosure relates to the field of APP classification detection technology, and in particular, to an application classification method, apparatus, electronic device, and storage medium. Background As the number of application software is increased in a burst type, the types are increasingly complicated, and the necessity of classification of the application software is gradually increased. For users, the target types (such as health management and office cooperation) can be rapidly positioned from mass applications, information overload is avoided, acquisition efficiency is greatly improved, for enterprises, financial institutions can strengthen wind control by classifying and identifying risk APP (such as fraud and illegal lending), and marketers can push according to category orientation, so that passenger acquisition cost is reduced. Disclosure of Invention The main purpose of the present specification is to provide an application classification method, apparatus, electronic device and storage medium, which aim to achieve accurate classification of applications. The technical scheme is as follows: in a first aspect, embodiments of the present disclosure provide an application classification method, including: Acquiring a classification evaluation request aiming at a target application, and generating a search instruction corresponding to the classification evaluation request based on a planning model; collecting related data of the target application according to the search instruction based on a search model; Classifying and evaluating the target application based on the related data by adopting a classifying decision model, and outputting a classifying and evaluating result of the target application; And generating a classification verification result of the classification evaluation result based on a verification model, and determining a classification result of the target application based on the classification verification result. In a second aspect, embodiments of the present disclosure provide an application classification apparatus, including: the planning unit is used for acquiring a classification evaluation request aiming at the target application and generating a search instruction corresponding to the classification evaluation request based on a planning model; the searching unit is used for collecting relevant data of the target application according to the searching instruction based on a searching model; The decision unit is used for carrying out classification evaluation on the target application based on the related data by adopting a classification decision model and outputting a classification evaluation result of the target application; and the verification unit is used for generating a classification verification result of the classification evaluation result based on a verification model and determining a classification result of the target application based on the classification verification result. In a third aspect, the present description provides an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, which when executed by the processor, performs the steps of the method as described above. In a fourth aspect, embodiments of the present description provide a storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the method as described above. In a fifth aspect, embodiments of the present description provide a computer program product comprising a computer program which, when executed by a processor of an electronic device, causes the processor to at least implement a method as described in the first aspect. In the embodiment of the specification, a search instruction corresponding to a classification evaluation request is generated based on a planning model by acquiring the classification evaluation request for the target application, relevant data of the target application is acquired from a public channel based on the search instruction based on the search model, classification evaluation is performed on the target application based on the relevant data by using a classification decision model, a classification evaluation result of the target application is output, then a verification model is used for verifying the classification evaluation result to obtain a classification verification result of the target application, and a classification result of a final target application is determined according to the classification verification result of each verification model. By adopting the method, the automatic classification of the application is realized by combining a plurality of models, the public text data of the application is used as a classification basis, the collection of personal data in the application is avoided, the comp