CN115496555B - Intelligent cross-border e-commerce website security quality assessment method and system - Google Patents
Intelligent cross-border e-commerce website security quality assessment method and system Download PDFInfo
- Publication number
- CN115496555B CN115496555B CN202211162298.7A CN202211162298A CN115496555B CN 115496555 B CN115496555 B CN 115496555B CN 202211162298 A CN202211162298 A CN 202211162298A CN 115496555 B CN115496555 B CN 115496555B
- Authority
- CN
- China
- Prior art keywords
- website
- border
- cross
- commerce
- information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0609—Buyer or seller confidence or verification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06395—Quality analysis or management
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Educational Administration (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Finance (AREA)
- Theoretical Computer Science (AREA)
- Accounting & Taxation (AREA)
- Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides an intelligent cross-border e-commerce website security quality assessment method and system, which relate to the technical field of artificial intelligence, and are characterized in that a cross-border e-commerce website set is acquired through big data and data crawling is carried out to obtain a cross-border e-commerce website database set, commodity release information sets of all e-commerce websites are obtained, an e-commerce website security quality assessment model and a website commodity security quality assessment model are constructed based on the information, the two are combined to generate a cross-border e-commerce website security quality assessment model, so that e-commerce website data information is assessed, and a mark reaching a website quality qualification threshold is a trusted e-commerce website. The method solves the technical problems that the quality judgment of the cross-border e-commerce website still depends on the judgment of manually opening the webpage click browsing, so that the quality judgment process is low in efficiency and poor in accuracy, realizes automatic data crawling and scoring of the cross-border e-commerce website, and further improves the efficiency and accuracy of the quality judgment of the cross-border e-commerce website.
Description
Technical Field
The invention relates to the technical field of artificial intelligence, in particular to an intelligent cross-border e-commerce website security quality assessment method and system.
Background
The cross-border electronic commerce is used as a new trade means in recent years, the related policy is changed from none to none and from passive to active, and in recent years, the cross-border electronic commerce is rapidly expanded in China under the influence of policy support, market environment improvement and other beneficial factors, and the over-border electronic commerce on-line trend of overseas retail lines is accelerated, so that more new electronic commerce consumers are emerging, and the trend of diversification and sustainability of purchasing goods is presented, and some merchants utilize the information of factories and cross-border terminal consumption groups to be asymmetric, so that the selling of false goods, the infringement of brands, low-price dumping and other unfair competition are carried out, and the cross-border electronic commerce website safety quality evaluation is particularly important. However, the method for evaluating the safety quality of the cross-border e-commerce website in the prior art has certain defects, and certain liftable space exists for the evaluation means.
In the prior art, the judgment of the quality of the cross-border e-commerce website still depends on the judgment of manually opening the webpage to click and browse, a great amount of time and energy are required to be consumed, and the quality of the website cannot be completely judged due to limited cognition of people, so that the efficiency of the quality judgment process is low and the judgment accuracy is low.
Disclosure of Invention
The application provides an intelligent cross-border e-commerce website safety quality assessment method and system, which are used for solving the technical problems that in the prior art, the quality judgment of the cross-border e-commerce website still depends on the judgment of manually opening a webpage to click and browse, a great amount of time and energy are required to be consumed, and the quality of the website cannot be completely judged due to limited cognition of people, so that the efficiency of the quality judgment process is low and the judgment accuracy is low, realizing automatic data crawling and scoring of the cross-border e-commerce website, and further improving the efficiency and accuracy of the quality judgment of the cross-border e-commerce website.
In view of the above problems, the application provides an intelligent cross-border e-commerce website security quality assessment method and system.
In a first aspect, the application provides an intelligent cross-border e-commerce website security quality assessment method, which comprises the following steps: acquiring a cross-border E-commerce website set through big data acquisition; the data crawling is carried out on the cross-border e-commerce website set, and a cross-border e-commerce website database set is obtained; acquiring commodity release information sets of all electronic commerce websites in the cross-border electronic commerce website sets; constructing an e-commerce website security quality assessment model based on the cross-border e-commerce website database set; constructing a website commodity security quality assessment model based on the commodity release information set; combining the E-commerce website security quality assessment model and the website commodity security quality assessment model to generate a cross-border E-commerce website security quality assessment model; inputting website data information of a cross-border e-commerce website to be evaluated into the cross-border e-commerce website security quality evaluation model for evaluation, and obtaining website security quality evaluation coefficients; setting a website quality qualification threshold, and marking the cross-border e-commerce website to be evaluated as a trusted e-commerce website when the website safety quality evaluation coefficient reaches the website quality qualification threshold.
In a second aspect, the present application provides an intelligent cross-border e-commerce website security quality assessment system, the system comprising: the cross-border e-commerce website acquisition module is used for acquiring a cross-border e-commerce website set through big data acquisition; the database acquisition module is used for crawling data of the cross-border e-commerce website set to obtain the cross-border e-commerce website database set; the commodity release information acquisition module is used for acquiring commodity release information sets of all electronic commerce websites in the cross-border electronic commerce website sets; the website evaluation model building module is used for building an e-commerce website safety quality evaluation model based on the cross-border e-commerce website database set; the commodity evaluation model building module is used for building a website commodity security quality evaluation model based on the commodity release information set; the cross-border e-commerce website evaluation model building module is used for combining the e-commerce website security quality evaluation model and the website commodity security quality evaluation model to generate a cross-border e-commerce website security quality evaluation model; the website evaluation coefficient acquisition module is used for inputting website data information of the cross-border e-commerce website to be evaluated into the cross-border e-commerce website security quality evaluation model for evaluation to obtain a website security quality evaluation coefficient; the website evaluation coefficient judgment module is used for setting a website quality qualification threshold, and marking the cross-border e-commerce website to be evaluated as a trusted e-commerce website when the website safety quality evaluation coefficient reaches the website quality qualification threshold.
One or more technical schemes provided by the application have at least the following technical effects or advantages:
the application provides an intelligent cross-border e-commerce website security quality assessment method, which relates to the technical field of artificial intelligence, and comprises the steps of acquiring a cross-border e-commerce website set through big data acquisition, performing data crawling, acquiring a cross-border e-commerce website database set, acquiring commodity release information sets of all e-commerce websites in the cross-border e-commerce website set, constructing an e-commerce website security quality assessment model based on the cross-border e-commerce website database set, constructing a website commodity security quality assessment model based on the commodity release information sets, combining the two to generate a cross-border e-commerce website security quality assessment model, inputting website data information of a cross-border e-commerce website to be assessed into the cross-border e-commerce website security quality assessment model for assessment, obtaining website security quality assessment coefficients, setting a website quality qualification threshold, and marking the cross-border e-commerce website to be assessed as a trusted e-commerce website when the website security quality assessment coefficients reach the website quality qualification threshold. According to the method, the cross-border e-commerce website database set and the commodity release information set are acquired in the cross-border e-commerce website security quality evaluation process, the cross-border e-commerce website security quality evaluation model is constructed, and whether the cross-border e-commerce website to be evaluated is credible or not can be effectively evaluated by detecting the website data information of the cross-border e-commerce website to be evaluated. The method and the device solve the technical problems that in the prior art, judgment on quality of a cross-border e-commerce website still depends on manual webpage opening click browsing judgment, a great amount of time and effort are required, and the quality of the website cannot be completely judged due to limited cognition of people, so that the quality judgment process is low in efficiency and low in judgment accuracy, realize automatic data crawling and scoring on the cross-border e-commerce website, and further achieve the technical effects of improving efficiency and accuracy of the quality judgment of the cross-border e-commerce website.
Drawings
FIG. 1 is a schematic flow chart of an intelligent cross-border E-commerce website security quality assessment method;
FIG. 2 is a schematic flow chart of a method for establishing an E-commerce website security quality assessment model in an intelligent cross-border E-commerce website security quality assessment method;
FIG. 3 is a schematic flow chart of an optimized evaluation model for obtaining the security quality of website commodity in the intelligent cross-border e-commerce website security quality evaluation method;
fig. 4 is a schematic structural diagram of an intelligent cross-border e-commerce website security quality assessment system provided by the application.
Reference numerals illustrate: the system comprises a cross-border e-commerce website acquisition module 1, a database acquisition module 2, a commodity release information acquisition module 3, a website evaluation model construction module 4, a commodity evaluation model construction module 5, a cross-border e-commerce website evaluation model construction module 6, a website evaluation coefficient acquisition module 7 and a website evaluation coefficient judgment module 8.
Detailed Description
The application provides an intelligent cross-border e-commerce website safety quality assessment method, which is used for solving the technical problems that in the prior art, the quality judgment of the cross-border e-commerce website still depends on the judgment of manually opening a webpage to click and browse, a great deal of time and energy are required to be consumed, and the quality of the website cannot be completely judged due to limited cognition of people, so that the efficiency of the quality judgment process is low and the judgment accuracy is low.
Example 1
As shown in fig. 1, the embodiment of the application provides an intelligent cross-border e-commerce website security quality assessment method, which comprises the following steps:
step S100: acquiring a cross-border E-commerce website set through big data acquisition;
specifically, the embodiment of the application provides an intelligent cross-border e-commerce website safety quality assessment method, which relates to the technical field of artificial intelligence.
Firstly, acquiring a cross-border e-commerce website set through big data acquisition, namely discriminating purchasing groups and facing areas of all e-commerce websites through big data information, determining the cross-border e-commerce websites and collecting information to obtain the cross-border e-commerce website set, wherein the difference of international stations and domestic stations for screening the alebab can be judged through registration modes, country-oriented and user names of all products or merchants used by customers and other information, the problem of judging whether the cross-border e-commerce website is the cross-border e-commerce is solved through acquisition of the cross-border e-commerce websites, and the effects of accurately confirming target websites and rapidly obtaining the cross-border e-commerce website set are achieved.
Step S200: data crawling is carried out on the cross-border e-commerce website set, and a cross-border e-commerce website database set is obtained;
specifically, through legal information collection means, data information such as commodity content, whether a commodity detail page is illegal, whether a website page is illegal, whether the whole website architecture is dangerous or not and the like are collected, for example, through data crawling software or running codes, keywords are classified and identified, products corresponding to the keywords are classified and identified, and the information is integrated to obtain a cross-border E-commerce website database set. By acquiring the cross-border e-commerce website database set, judgment on cross-border e-commerce website information is solved, the problem of website information integration is solved, and the effect of calling and using the website information at any time is achieved.
Step S300: acquiring commodity release information sets of all electronic commerce websites in the cross-border electronic commerce website sets;
specifically, the obtained cross-border e-commerce website collection information is screened to obtain commodity release information collection of each e-commerce website, the commodity release information collection comprises information such as website release product types, product properties, whether products are illegal or not, and the like, and by taking an Ariba International station as an example, whether the products are RTS products or ordered products, whether products which can be sold only after responding to qualification, whether the products are illegal or forbidden drugs, product delivery modes and the like can be distinguished, and the problems of judgment and integration of the product content of each e-commerce website are solved through the acquisition of the commodity release information collection, so that the effects of integration of the e-commerce website products and integration of commodity safety information are achieved.
Step S400: constructing an e-commerce website security quality assessment model based on the cross-border e-commerce website database set;
specifically, the obtained cross-border e-commerce website database set is subjected to website classification to obtain information service cross-border e-commerce website database information and online transaction cross-border e-commerce website database information, abnormal feature extraction is performed on the obtained database information to obtain the information service cross-border e-commerce website abnormal feature set and the online transaction cross-border e-commerce website abnormal feature set, neural network model training is performed respectively based on the information service cross-border e-commerce website abnormal feature set and the online transaction cross-border e-commerce website abnormal feature set, an information service website safety quality assessment model and an online transaction website safety quality assessment model are generated, and model parameters of the obtained assessment model are subjected to joint training to obtain an e-commerce website safety quality assessment model. The e-commerce website safety quality assessment model is a model for assessing the cross-border e-commerce website after integrating information such as whether the website architecture of the e-commerce website is complete and safe, whether the website page meets the regulations, whether the page quality meets the requirements, whether the released product meets the corresponding safety regulations and the like, and the effect of quickly identifying the non-compliant website and carrying out next screening on the reasonably compliant website is achieved in the process of monitoring and screening the cross-border e-commerce website through the construction of the e-commerce website safety quality assessment model.
Step S500: constructing a website commodity security quality assessment model based on the commodity release information set;
specifically, classifying the obtained data information of the products released by the cross-border e-commerce website to obtain product database information and product transaction database information, extracting abnormal characteristics of the obtained database information to obtain abnormal information characteristics of the released product database and abnormal information characteristics of the transaction product database, respectively training a neural network model based on the abnormal information characteristics and the abnormal information characteristics of the transaction product database, generating a released product information safety quality assessment model and a product transaction safety quality assessment model, jointly training model parameters of the obtained assessment model, and obtaining a safety quality assessment model of the products released by the e-commerce website. The commodity safety quality assessment model is a method for collecting and assessing safety quality information of product contents released by cross-border e-commerce websites, wherein the safety quality information of the product is information related to the product, such as whether the released product meets legal requirements, whether detail pages of the released product meet requirements and are complete, whether the quality of the released product meets market and consumer requirements, and the like, and the information is integrated and used for judging whether the product released by the e-commerce websites meets the requirements. By constructing the website commodity safety quality assessment model, the method solves the problem of identifying whether a large number of products meet the requirements of cross-border E-commerce websites, and achieves the effect of rapidly judging commodity information quality.
Step S600: combining the E-commerce website security quality assessment model and the website commodity security quality assessment model to generate a cross-border E-commerce website security quality assessment model;
specifically, the obtained e-commerce website security quality assessment model and the obtained website commodity security quality assessment model are combined, namely, a method for identifying website security information and a method for identifying product security information are combined to form a method capable of simultaneously processing the website security information and the product security information, so that a cross-border e-commerce website security quality assessment model is generated, the cross-border e-commerce website security quality assessment model is a method for simultaneously acquiring and screening the website information security of the cross-border e-commerce website and the product security information published by the cross-border e-commerce website, the cross-border e-commerce website security quality assessment model is a method for integrating the website security quality assessment model and the website commodity security quality assessment model, and qualified information features and abnormal information features collected by the website security quality assessment model and the product security information feature are further combined by an artificial intelligent system to learn, so that a set of the website security and the website product security information features is finally obtained. The cross-border e-commerce website safety quality assessment model solves the identification processing work of the whole important content of the website, and achieves the effect of rapidly and accurately obtaining whether the cross-border e-commerce website accords with the specified information.
Step S700: inputting the website data information of the cross-border e-commerce website to be evaluated into a cross-border e-commerce website security quality evaluation model for evaluation, and obtaining a website security quality evaluation coefficient.
Specifically, the cross-border e-commerce website safety quality assessment model comprises an input layer, a quality assessment logic layer and an output layer, wherein website data information of a cross-border e-commerce website to be assessed is used as the input layer and is input into the quality assessment logic layer, the website information and product information characteristics of the cross-border e-commerce website are judged, whether the characteristic information of the cross-border e-commerce website meets the regulations or not is obtained after the characteristic information of the cross-border e-commerce website is judged, a website safety quality assessment coefficient is output, the website safety quality assessment coefficient is output based on the output layer as a model output result, the website safety quality assessment coefficient is obtained, the website safety quality assessment coefficient is an assessment coefficient for judging whether the abnormal characteristic information quantity of the website safety quality assessment coefficient is in an illegal degree according to the information characteristics of the website safety quality assessment coefficient, and the website safety quality assessment coefficient is output based on the output layer as a model output result, and the website safety quality assessment coefficient is obtained. By acquiring the website security quality evaluation coefficients, the evaluation work of each cross-border e-commerce website is solved, the efficiency and accuracy of website evaluation are ensured, and the effects of high efficiency of website processing and timely judgment of abnormal factors of the website are achieved.
Step S800: setting a website quality qualification threshold, and marking the cross-border e-commerce website to be evaluated as a trusted e-commerce website when the website safety quality evaluation coefficient reaches the website quality qualification threshold.
Specifically, a website quality qualification threshold is set according to actual conditions, the website quality qualification threshold is a qualification standard set according to relevant laws and regulations for cross-border e-commerce website bodies and commodities offered on websites, the standard is converted into scores, the scores obtained after the cross-border e-commerce websites are evaluated are compared with the qualification threshold, when the e-commerce websites reach the standard or are higher than the standard, the websites can be determined to be trusted websites, if the obtained scores are lower than the standard, the websites are marked as untrusted e-commerce websites, through setting of the website quality qualification threshold, the evaluation and marking problems of the e-commerce websites are solved, the qualified and unqualified websites are timely classified, and a user can notice the problems to avoid unpredictable results, so that the effects of protecting the user and the consumer are achieved, and the cross-border e-commerce websites can be timely processed and supervised.
Further, as shown in fig. 2, step S400 of the present application further includes:
Step S410: classifying websites of the cross-border e-commerce website database collection to obtain information service cross-border e-commerce website database information and online transaction cross-border e-commerce website database information;
step S420: carrying out abnormal feature extraction on information service cross-border e-commerce website database information and online transaction cross-border e-commerce website database information to obtain an information service cross-border e-commerce website abnormal feature set and an online transaction cross-border e-commerce website abnormal feature set;
step S430: respectively training a neural network model based on the information service cross-border e-commerce website abnormal feature set and the online transaction cross-border e-commerce website abnormal feature set to generate an information service website safety quality assessment model and an online transaction website safety quality assessment model;
step S440: and carrying out joint training on model parameters of the information service website safety quality assessment model and the online transaction website safety quality assessment model to obtain the e-commerce website safety quality assessment model.
Specifically, the obtained cross-border e-commerce website database set is subjected to website classification to obtain information service cross-border e-commerce website database information and online transaction cross-border e-commerce website database information, abnormal feature extraction is carried out on the information service cross-border e-commerce website database information and the online transaction cross-border e-commerce website database information respectively, an e-commerce website security evaluation dimension set is constructed, the information service cross-border e-commerce website database information and the online transaction cross-border e-commerce website database information are evaluated according to the e-commerce website security evaluation dimension set to obtain information service website security evaluation information and online transaction website evaluation information, a website abnormal feature support vector machine is obtained, the information service website security evaluation information and the online transaction website evaluation information are input into the website abnormal feature support vector machine, and the information service cross-border e-commerce website abnormal feature set and the online transaction cross-border e-commerce website abnormal feature set are output respectively.
The method comprises the steps of training a neural network model of a website abnormal characteristic support vector machine, respectively generating an information service website security quality assessment model and an online transaction website security quality assessment model for information service cross-border e-commerce website database information and online transaction cross-border e-commerce website database information, performing joint training on model parameters of the information service website security quality assessment model and the online transaction website security quality assessment model, namely fusing the information service website security quality assessment model and the online transaction website security quality assessment model to form a method capable of simultaneously processing the information service website security quality and the online transaction website security quality, such as whether a website architecture of an e-commerce website is complete and secure, whether a website page meets the requirements, whether the page quality meets the requirements, whether the released product quality meets the corresponding security regulations and the like, so as to generate the e-commerce website security quality assessment model.
Through construction of the E-commerce website safety quality assessment model, in the process of monitoring and screening the cross-border E-commerce website, the effects of rapidly identifying the non-compliant website and carrying out next screening on the reasonably compliant website are achieved.
Further, step S420 of the present application further includes:
Step S421: constructing an E-commerce website security evaluation dimension set;
step S422: evaluating information service cross-border e-commerce website database information and online transaction cross-border e-commerce website database information according to the e-commerce website security evaluation dimension set to obtain information service website security evaluation information and online transaction website evaluation information;
step S423: obtaining a website abnormal characteristic support vector machine;
step S424: and inputting the information service website security evaluation information and the online transaction website evaluation information into a website abnormal feature support vector machine, and respectively outputting an information service cross-border e-commerce website abnormal feature set and an online transaction cross-border e-commerce website abnormal feature set.
Specifically, an e-commerce website security evaluation dimension set is constructed, and three dimensions are constructed, wherein the larger the number is, the higher the evaluation is, such as the complete security A of a website architecture of the e-commerce website 1 、A 2 、A 3 Published product quality B 1 、B 2 、B 3 Compliance C of the issued product with the corresponding safety regulations 1 、C 2 、C 3 According to the E-commerce website security evaluation dimensionThe degree set evaluates information service cross-border e-commerce website database information and online transaction cross-border e-commerce website database information, if a certain cross-border e-commerce website evaluation result is A 1 B 1 C 2 The cross-border e-commerce website is poor in evaluation, so that information service website safety evaluation information and online transaction website evaluation information are obtained.
The method comprises the steps of obtaining a website abnormal characteristic support vector machine, enabling the support vector machine to find an optimal scheme for classification problems, obtaining historical website safety evaluation information, wherein the historical website safety evaluation information comprises historical information service website safety evaluation information and historical online transaction website evaluation information, dividing the historical website safety evaluation information according to a preset proportion to obtain a website evaluation information training sample and a website evaluation information test sample, obtaining a training sample evaluation characteristic label and a test sample evaluation characteristic label according to the website evaluation information training sample and the website evaluation information test sample, taking the website evaluation information training sample and the training sample evaluation characteristic label as training data, and constructing the website abnormal characteristic support vector machine. Inputting the information service website security evaluation information and the online transaction website evaluation information into a website abnormal feature support vector machine to respectively obtain an information service cross-border e-commerce website abnormal feature set and an online transaction cross-border e-commerce website abnormal feature set.
The method has the advantages that through the acquisition of the information service cross-border e-commerce website abnormal feature set and the online transaction cross-border e-commerce website abnormal feature set, in the process of monitoring and screening the cross-border e-commerce website, the effect of rapidly identifying the non-compliant website and carrying out next screening on the reasonably compliant website is achieved.
Further, step S423 of the present application further includes:
step S4231: acquiring historical website security evaluation information, wherein the historical website security evaluation information comprises historical information service website security evaluation information and historical online transaction website evaluation information;
step S4232: dividing historical website security evaluation information according to a preset proportion to obtain a website evaluation information training sample and a website evaluation information test sample;
step S4233: according to the website evaluation information training sample and the website evaluation information test sample, obtaining a training sample evaluation characteristic label and a test sample evaluation characteristic label;
step S4234: and taking the website evaluation information training sample and the training sample evaluation feature label as training data to construct a website abnormal feature support vector machine.
Specifically, in order to train the website abnormal feature support vector machine, historical website safety evaluation information is obtained through modes such as website background query, wherein the historical website safety evaluation information comprises historical information service website safety evaluation information and historical online transaction website evaluation information, the historical website safety evaluation information is randomly divided into two parts, and the two parts are respectively defined as website evaluation information training samples and website evaluation information testing samples. Further, according to the website evaluation information training sample and the website evaluation information test sample, corresponding training sample evaluation feature labels and test sample evaluation feature labels are respectively obtained, wherein the evaluation feature labels comprise feature normal labels and feature abnormal labels, and further, the website evaluation information training sample and the training sample evaluation feature labels are used as input information of a website abnormal feature support vector machine and participate in training the website abnormal feature support vector machine.
The website abnormal characteristic support vector machine is obtained through training, so that the defects of low judging speed, low accuracy and the like of the information service website safety and the online transaction website safety are overcome, a quick and effective method is provided for judging the information service website safety and the online transaction website safety, and the technical effects of judging and screening the information service website safety and the online transaction website safety quickly and with high accuracy are achieved.
Further, step S423 of the present application further includes:
step S4235: inputting the website evaluation information test sample into an abnormal feature support vector machine to obtain feature results of each sample in the website evaluation information test sample;
step S4236: comparing the characteristic results of each sample with the characteristic labels of the test samples to obtain the accuracy of the characteristic labels of the samples;
step S4237: and taking the accuracy of the sample feature labels as the analysis accuracy of the website abnormal feature support vector machine.
Specifically, a website evaluation information test sample obtained after the historical website security evaluation information is divided is input into a trained website abnormal feature support vector machine, each sample feature result in the website evaluation information test sample is obtained, each sample feature result obtained through intelligent prediction of the website abnormal feature support vector machine is compared with a test sample evaluation feature label, when the output result of the website abnormal feature support vector machine is consistent with the test sample evaluation feature label, the evaluation of the website abnormal feature support vector machine is accurate, and when the output result of the website abnormal feature support vector machine is inconsistent with the test sample evaluation feature label, the evaluation accuracy of the website abnormal feature support vector machine is insufficient, and subsequent model optimization such as sample increasing training or algorithm optimization is needed. That is, based on all samples of the website evaluation information test samples evaluated by the website abnormal feature support vector machine, the number of samples for evaluating the normal features and the abnormal features is counted respectively, and then the evaluation accuracy of the website abnormal feature support vector machine can be obtained through calculation.
The target of analyzing the evaluation result of the abnormal characteristic support vector machine of the website is realized through the test sample of the evaluation information of the website, and the technical effect of objectively and effectively evaluating the evaluation quality of the abnormal characteristic support vector machine of the website is achieved based on the evaluation accuracy.
Further, as shown in fig. 3, the present application further includes:
step S910: popularity evaluation is carried out on the website searching times, and website level information of the E-commerce is obtained;
step S920: analyzing the satisfaction degree of the commodity purchase of the website to obtain the quality index of the purchase of the website of the electronic commerce;
step S930: generating a website quality influence factor according to the electronic commerce website level information and the electronic commerce website purchase quality index;
step S940: and carrying out optimization learning on the website commodity safety quality model based on the website quality influence factors to obtain a website commodity safety quality optimization assessment model.
Specifically, the website searching times are collected and counted through the website background to obtain the website level information X of the E-commerce 1 、X 2 、X 3 The larger the number, the more the number of searches for the web site, i.e. X 3 The method comprises the steps of carrying out a first treatment on the surface of the The satisfaction degree of the purchase of the website commodity, namely the sales volume, the good evaluation rate, the poor evaluation rate, the return rate and the like of the website commodity are analyzed to obtain the purchase quality index Y of the website of the electronic commerce 1 、Y 2 、Y 3 The larger the number, the higher the commodity purchasing index, for example, the higher the commodity sales, the higher the good score, the lower the return rate, the higher the commodity purchasing quality index of the website, which is Y 3 The method comprises the steps of carrying out a first treatment on the surface of the With E-commerce website level information X 1 、X 2 、X 3 And E-commerce website purchase quality index Y 1 、Y 2 、Y 3 Generating website quality influence factors, so as to optimally learn a website commodity safety quality model, for example, a website is evaluated by an e-commerce website safety quality evaluation model and a website commodity safety quality evaluation model to obtain a website with poor website architecture, poor website pages and less product categories, belonging to a poor website, wherein the website quality influence factors of the website are obtained by analyzing the website searching times and commodity purchasing satisfaction of the website 3 Y 3 Namely, the website has high searching times and high commodity purchasing satisfaction, and the quality of the website influences factor X 3 Y 3 And (3) performing optimization learning on the website commodity safety quality model of the website, and improving the evaluation of the website to obtain a more accurate evaluation effect so as to obtain a website commodity safety quality optimization evaluation model.
Further, step S700 of the present application includes:
step S710: the cross-border e-commerce website safety quality assessment model comprises an input layer, a quality assessment logic layer and an output layer;
Step S720: inputting website data information as an input layer into a quality evaluation logic layer, and outputting a website safety quality evaluation coefficient;
step S730: and outputting the website security quality assessment coefficient as a model output result based on the output layer.
Specifically, website data information of a cross-border e-commerce website to be evaluated is input into a cross-border e-commerce website security quality evaluation model for evaluation, wherein the cross-border e-commerce website security quality evaluation model comprises an input layer, a quality evaluation logic layer and an output layer, the input layer is used for inputting the cross-border e-commerce website into the model, the quality evaluation logic layer is used for judging website information and product information characteristics of the cross-border e-commerce website, the output layer is used for judging whether the characteristic information of the cross-border e-commerce website meets the regulations and submitting the results, the website data information of the cross-border e-commerce website to be evaluated is used as the input layer, the evaluation coefficient for judging whether the abnormal characteristic information quantity of the cross-border e-commerce website is in an illegal degree according to the information characteristics of the cross-border e-commerce website is input into the quality evaluation logic layer, and the website security quality evaluation coefficient is output as a model output result based on the output layer, so that the website security quality evaluation coefficient is output. The evaluation work of each cross-border e-commerce website is solved, the efficiency and accuracy of website evaluation are guaranteed, and the effects of high efficiency of website processing and timely judgment of abnormal factors of the website are achieved.
Example two
Based on the same inventive concept as the method for evaluating the security quality of an intelligent cross-border e-commerce website in the foregoing embodiment, as shown in fig. 4, the application provides an intelligent cross-border e-commerce website security quality evaluation system, which comprises:
the cross-border e-commerce website acquisition module 1 is used for acquiring a cross-border e-commerce website set through big data acquisition;
the database acquisition module 2 is used for crawling data of the cross-border e-commerce website set to obtain the cross-border e-commerce website database set;
the commodity release information acquisition module 3 is used for acquiring commodity release information sets of all the electronic commerce websites in the cross-border electronic commerce website sets;
the website evaluation model building module 4 is used for building an e-commerce website safety quality evaluation model based on the cross-border e-commerce website database set;
the commodity evaluation model building module 5 is used for building a website commodity security quality evaluation model based on the commodity release information set;
the cross-border e-commerce website evaluation model building module 6 is used for combining the e-commerce website security quality evaluation model and the website commodity security quality evaluation model to generate a cross-border e-commerce website security quality evaluation model;
The website evaluation coefficient acquisition module 7 is used for inputting website data information of the cross-border e-commerce website to be evaluated into the cross-border e-commerce website security quality evaluation model for evaluation to obtain a website security quality evaluation coefficient;
the website evaluation coefficient judgment module 8 is used for setting a website quality qualification threshold, and when the website safety quality evaluation coefficient reaches the website quality qualification threshold, the cross-border e-commerce website to be evaluated is marked as a trusted e-commerce website.
Further, the system further comprises:
the website classification module is used for classifying websites of the cross-border e-commerce website database collection to obtain information service cross-border e-commerce website database information and online transaction cross-border e-commerce website database information;
the abnormal feature extraction module is used for extracting abnormal features of information service cross-border e-commerce website database information and online transaction cross-border e-commerce website database information to obtain an information service cross-border e-commerce website abnormal feature set and an online transaction cross-border e-commerce website abnormal feature set;
the neural network model training module is used for respectively training the neural network model based on the information service cross-border e-commerce website abnormal feature set and the online transaction cross-border e-commerce website abnormal feature set to generate an information service website safety quality assessment model and an online transaction website safety quality assessment model;
The joint training module is used for joint training of model parameters of the information service website safety quality assessment model and the online transaction website safety quality assessment model to obtain the e-commerce website safety quality assessment model.
Further, the system further comprises:
the electronic commerce website safety evaluation dimension set construction module is used for constructing an electronic commerce website safety evaluation dimension set;
the database information evaluation module is used for evaluating information service cross-border e-commerce website database information and online transaction cross-border e-commerce website database information according to the e-commerce website security evaluation dimension set to obtain information service website security evaluation information and online transaction website evaluation information;
the website abnormal feature support vector machine acquisition module is used for acquiring a website abnormal feature support vector machine;
the abnormal feature set output module is used for inputting the information service website security evaluation information and the online transaction website evaluation information into a website abnormal feature support vector machine and respectively outputting an information service cross-border e-commerce website abnormal feature set and an online transaction cross-border e-commerce website abnormal feature set.
Further, the system further comprises:
the historical website safety evaluation information acquisition module is used for acquiring historical website safety evaluation information, wherein the historical website safety evaluation information comprises historical information service website safety evaluation information and historical online transaction website evaluation information;
the historical website safety evaluation information dividing module is used for dividing the historical website safety evaluation information according to a preset proportion to obtain a website evaluation information training sample and a website evaluation information test sample;
the sample evaluation feature tag acquisition module is used for acquiring training sample evaluation feature tags and test sample evaluation feature tags according to the website evaluation information training samples and the website evaluation information test samples;
and the training data processing module takes the website evaluation information training sample and the training sample evaluation feature label as training data to construct a website abnormal feature support vector machine.
Further, the system further comprises:
the sample feature result acquisition module is used for inputting the website evaluation information test sample into the abnormal feature support vector machine to obtain each sample feature result in the website evaluation information test sample;
The sample characteristic label accuracy rate acquisition module is used for comparing each sample characteristic result with the test sample evaluation characteristic label to obtain sample characteristic label accuracy rate;
and the sample feature tag accuracy processing module is used for taking the sample feature tag accuracy as the analysis accuracy of the website abnormal feature support vector machine.
Further, the system further comprises:
the popularity evaluation module is used for performing popularity evaluation on the search times of the websites to obtain website level information of the electronic commerce;
the satisfaction analysis module is used for analyzing the purchase satisfaction of the website commodity and obtaining the purchase quality index of the website of the electronic commerce;
the website quality influence factor generation module is used for generating website quality influence factors according to the e-commerce website level information and the e-commerce website purchase quality index;
and the optimization learning module is used for performing optimization learning on the website commodity safety quality model based on the website quality influence factors to obtain a website commodity safety quality optimization assessment model.
Further, the system further comprises:
the website safety quality assessment coefficient output module is used for taking website data information as an input layer, inputting the website data information into the quality assessment logic layer and outputting a website safety quality assessment coefficient;
the model output result output module is used for outputting the website security quality assessment coefficient as a model output result based on the output layer.
Through the foregoing detailed description of the method for evaluating the security quality of an intelligent cross-border e-commerce website, those skilled in the art can clearly know the method and the system for evaluating the security quality of an intelligent cross-border e-commerce website in this embodiment, and for the device disclosed in the embodiment, the description is relatively simple because the device corresponds to the method disclosed in the embodiment, and relevant places refer to the description of the method section.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (7)
1. An intelligent cross-border e-commerce website security quality assessment method is characterized by comprising the following steps:
acquiring a cross-border E-commerce website set through big data acquisition;
the data crawling is carried out on the cross-border e-commerce website set, and a cross-border e-commerce website database set is obtained;
acquiring commodity release information sets of all electronic commerce websites in the cross-border electronic commerce website sets;
constructing an e-commerce website security quality assessment model based on the cross-border e-commerce website database set;
constructing a website commodity security quality assessment model based on the commodity release information set;
combining the E-commerce website security quality assessment model and the website commodity security quality assessment model to generate a cross-border E-commerce website security quality assessment model;
inputting website data information of a cross-border e-commerce website to be evaluated into the cross-border e-commerce website security quality evaluation model for evaluation, and obtaining website security quality evaluation coefficients;
setting a website quality qualification threshold, and marking the cross-border e-commerce website to be evaluated as a trusted e-commerce website when the website safety quality evaluation coefficient reaches the website quality qualification threshold;
the construction of the e-commerce website security quality assessment model based on the cross-border e-commerce website database set comprises the following steps:
Classifying websites of the cross-border e-commerce website database collection to obtain information service cross-border e-commerce website database information and online transaction cross-border e-commerce website database information;
extracting abnormal characteristics of the information service cross-border e-commerce website database information and the online transaction cross-border e-commerce website database information to obtain an information service cross-border e-commerce website abnormal characteristic set and an online transaction cross-border e-commerce website abnormal characteristic set;
respectively training a neural network model based on the information service cross-border e-commerce website abnormal feature set and the online transaction cross-border e-commerce website abnormal feature set to generate an information service website safety quality assessment model and an online transaction website safety quality assessment model;
and carrying out joint training on model parameters of the information service website safety quality assessment model and the online transaction website safety quality assessment model to obtain the e-commerce website safety quality assessment model.
2. The method of claim 1, wherein the obtaining the information service cross-border e-commerce web site anomaly feature set and the online transaction cross-border e-commerce web site anomaly feature set comprises:
Constructing an E-commerce website security evaluation dimension set;
evaluating the information service cross-border e-commerce website database information and the online transaction cross-border e-commerce website database information according to the e-commerce website security evaluation dimension set to obtain information service website security evaluation information and online transaction website evaluation information;
obtaining a website abnormal characteristic support vector machine;
inputting the information service website security evaluation information and the online transaction website evaluation information into the website anomaly feature support vector machine, and respectively outputting the information service cross-border e-commerce website anomaly feature set and the online transaction cross-border e-commerce website anomaly feature set.
3. The method of claim 2, wherein obtaining a website anomaly characteristic support vector machine comprises:
acquiring historical website security evaluation information, wherein the historical website security evaluation information comprises historical information service website security evaluation information and historical online transaction website evaluation information;
dividing the historical website security evaluation information according to a preset proportion to obtain a website evaluation information training sample and a website evaluation information test sample;
according to the website evaluation information training sample and the website evaluation information test sample, obtaining a training sample evaluation characteristic label and a test sample evaluation characteristic label;
And taking the website evaluation information training sample and the training sample evaluation feature label as training data to construct the website abnormal feature support vector machine.
4. A method according to claim 3, wherein the method comprises:
inputting the website evaluation information test sample into the abnormal feature support vector machine to obtain feature results of each sample in the website evaluation information test sample;
comparing the characteristic results of each sample with the characteristic labels of the test samples to obtain the accuracy of the characteristic labels of the samples;
and taking the accuracy of the sample feature label as the analysis accuracy of the website abnormal feature support vector machine.
5. The method of claim 1, wherein the method comprises:
popularity evaluation is carried out on the website searching times, and website level information of the E-commerce is obtained;
analyzing the satisfaction degree of the commodity purchase of the website to obtain the quality index of the purchase of the website of the electronic commerce;
generating a website quality influence factor according to the e-commerce website level information and the e-commerce website purchase quality index;
and carrying out optimization learning on the website commodity safety quality model based on the website quality influence factor to obtain a website commodity safety quality optimization assessment model.
6. The method of claim 1, wherein obtaining website security quality assessment coefficients comprises:
the cross-border e-commerce website safety quality assessment model comprises an input layer, a quality assessment logic layer and an output layer;
inputting the website data information into the quality evaluation logic layer as an input layer, and outputting a website security quality evaluation coefficient;
and outputting the website security quality assessment coefficient as a model output result based on the output layer.
7. An intelligent cross-border e-commerce website security quality assessment system, the system comprising:
the cross-border e-commerce website acquisition module is used for acquiring a cross-border e-commerce website set through big data acquisition;
the database acquisition module is used for crawling data of the cross-border e-commerce website set to obtain the cross-border e-commerce website database set;
the commodity release information acquisition module is used for acquiring commodity release information sets of all electronic commerce websites in the cross-border electronic commerce website sets;
the website evaluation model building module is used for building an e-commerce website safety quality evaluation model based on the cross-border e-commerce website database set;
The commodity evaluation model building module is used for building a website commodity security quality evaluation model based on the commodity release information set;
the cross-border e-commerce website evaluation model building module is used for combining the e-commerce website security quality evaluation model and the website commodity security quality evaluation model to generate a cross-border e-commerce website security quality evaluation model;
the website evaluation coefficient acquisition module is used for inputting website data information of the cross-border e-commerce website to be evaluated into the cross-border e-commerce website security quality evaluation model for evaluation to obtain a website security quality evaluation coefficient;
the website evaluation coefficient judgment module is used for setting a website quality qualification threshold, and marking the cross-border e-commerce website to be evaluated as a trusted e-commerce website when the website safety quality evaluation coefficient reaches the website quality qualification threshold;
the website classification module is used for classifying websites of the cross-border e-commerce website database collection to obtain information service cross-border e-commerce website database information and online transaction cross-border e-commerce website database information;
The abnormal feature extraction module is used for extracting abnormal features of information service cross-border e-commerce website database information and online transaction cross-border e-commerce website database information to obtain an information service cross-border e-commerce website abnormal feature set and an online transaction cross-border e-commerce website abnormal feature set;
the neural network model training module is used for respectively training the neural network model based on the information service cross-border e-commerce website abnormal feature set and the online transaction cross-border e-commerce website abnormal feature set to generate an information service website safety quality assessment model and an online transaction website safety quality assessment model;
the joint training module is used for joint training of model parameters of the information service website safety quality assessment model and the online transaction website safety quality assessment model to obtain the e-commerce website safety quality assessment model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211162298.7A CN115496555B (en) | 2022-09-23 | 2022-09-23 | Intelligent cross-border e-commerce website security quality assessment method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211162298.7A CN115496555B (en) | 2022-09-23 | 2022-09-23 | Intelligent cross-border e-commerce website security quality assessment method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115496555A CN115496555A (en) | 2022-12-20 |
| CN115496555B true CN115496555B (en) | 2023-10-13 |
Family
ID=84470386
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211162298.7A Active CN115496555B (en) | 2022-09-23 | 2022-09-23 | Intelligent cross-border e-commerce website security quality assessment method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115496555B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116662631B (en) * | 2023-04-23 | 2024-05-28 | 无锡览山信息科技有限公司 | Cloud operation and maintenance management platform multi-cloud nano tube system and method based on Internet |
| CN117745309B (en) * | 2024-02-20 | 2024-05-10 | 江苏中创供应链服务有限公司 | Processing method and system for cross-border e-commerce industry data trusted service |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104767757A (en) * | 2015-04-17 | 2015-07-08 | 国家电网公司 | Multiple-dimension security monitoring method and system based on WEB services |
| CN108733748A (en) * | 2018-04-04 | 2018-11-02 | 浙江大学城市学院 | A kind of cross-border product quality risk fuzzy prediction method based on comment on commodity public sentiment |
| CN111967503A (en) * | 2020-07-24 | 2020-11-20 | 西安电子科技大学 | Method for constructing multi-type abnormal webpage classification model and abnormal webpage detection method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8700487B2 (en) * | 2012-02-03 | 2014-04-15 | Buysafe, Inc. | User to website guaranteed shopping |
-
2022
- 2022-09-23 CN CN202211162298.7A patent/CN115496555B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104767757A (en) * | 2015-04-17 | 2015-07-08 | 国家电网公司 | Multiple-dimension security monitoring method and system based on WEB services |
| CN108733748A (en) * | 2018-04-04 | 2018-11-02 | 浙江大学城市学院 | A kind of cross-border product quality risk fuzzy prediction method based on comment on commodity public sentiment |
| CN111967503A (en) * | 2020-07-24 | 2020-11-20 | 西安电子科技大学 | Method for constructing multi-type abnormal webpage classification model and abnormal webpage detection method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115496555A (en) | 2022-12-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115496555B (en) | Intelligent cross-border e-commerce website security quality assessment method and system | |
| CN110415087B (en) | Electronic commerce transaction system based on internet | |
| CN110009372B (en) | User risk identification method and device | |
| CN108648038A (en) | A kind of credit propagation and maliciously evaluation recognition methods excavated based on subgraph | |
| CN113034238A (en) | Commodity brand feature extraction and intelligent recommendation management method based on electronic commerce platform transaction | |
| CN112036985A (en) | Marketing system of E-commerce operation | |
| CN110414902B (en) | E-commerce platform system based on internet | |
| CN115187065A (en) | Garment commodity quality analysis system based on electronic commerce platform | |
| CN112069315A (en) | Method, device, server and storage medium for extracting text multidimensional information | |
| CN108985431B (en) | Anti-counterfeit label and method for detecting and identifying genuine goods of registered trademark | |
| CN113988961A (en) | Target object prediction method and device | |
| CN112465600A (en) | Order data detection method and system applied to online business | |
| KR102482969B1 (en) | Artificial intelligence-based system and method for online counterfeit product crackdown | |
| CN114266594A (en) | Big data analysis method based on southeast Asia cross-border e-commerce platform | |
| CN116502813A (en) | Abnormal order detection method based on ensemble learning | |
| CN107092600B (en) | Information identification method and device | |
| CN108846683A (en) | Online shopping city data interactive method and device | |
| KR20180092053A (en) | Intelligent shopping managing system | |
| CN114155057A (en) | Commodity recommendation system for electronic commerce platform | |
| KR102533239B1 (en) | Image library system and counterfeit detection method for artificial intelligence-based counterfeit product detection | |
| KR102470248B1 (en) | Natural language processing method for identification of counterfeit products and counterfeit product detection system | |
| CN111461797B (en) | Goods pricing method based on user behaviors | |
| CN110648173B (en) | Unsupervised abnormal commodity data detection method based on good evaluation and poor evaluation rates of commodities | |
| KR102283191B1 (en) | Method and server for inspecting food advertising images | |
| KR102483987B1 (en) | Automated analysis method and system for target item and target market in open market |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |