CN117522500A - Method, system, computer device and computer storage medium for authenticating images of electronic store - Google Patents

Abstract

The application discloses an image authentication method of an electronic store, which comprises the steps of extracting data related to the store to form an original data set, evaluating the business condition of the store based on the original data set, refining a characteristic label of the store based on the evaluation result, and generating at least one of a store business card, a business file and/or an authentication report of the store. The method in some embodiments of the invention utilizes the salient features before feature engineering extraction to make the model more explanatory for the evaluation and authentication results of the store operation condition. Corresponding systems, computer devices, and computer storage media are also disclosed.

Description

Method, system, computer device and computer storage medium for authenticating images of electronic store

Technical Field

The invention belongs to big data products, and relates to a system and a method for authenticating images of an electronic shop by using big data, cloud computing, machine learning, image technology, customer image technology and other technologies.

Background

Accurately evaluating business entities via data to predict transaction security and then determine whether to conduct a transaction has always been a problem to be solved in the business field. This problem is also pronounced on e-commerce platforms where the buyer and seller are not transacted properly and are remote. On the e-commerce platform there are a number of stores, some of which are associated with business entities in real life, such as registry, while others are operated by individuals or teams, whereas the operators of the stores are called sellers, whereby it is seen that sellers may be registry or perhaps just individuals or teams. Buyers acquire goods or services by transacting with stores on these e-commerce platforms. As e-commerce platform participants mature, buyers are more concerned about the security of transactions than just whether related products or services exist, and also about the price of the transaction. Therefore, prior to conducting the transaction, the buyer wishes to have a systematic knowledge of the seller's overall situation to determine whether the transaction is secure, and thus whether to conduct the e-commerce transaction.

In the prior art, a method for calculating the integrity level of an enterprise from four dimensions of enterprise profile, business strength, service capability and enterprise movement by using enterprise online transaction data and offline public data exists, but the method has the following problems: firstly, the method only calculates and displays the integrity grade of the enterprise, the result is single, and no customer portrait is carried out on the enterprise, so that the interpretation of the result is poor, and the difficulty and threshold of using the result by a user are high; secondly, the method does not evaluate and check the real situation of the transaction after collecting the transaction data on the enterprise line, and the problem that the calculated data base is distorted due to common phenomena such as bill swiping and the like, and finally, the calculation result and the actual situation possibly have larger deviation is avoided; thirdly, the method is that the operation of the shops is completely equivalent to the operation of enterprises, namely, the shops operated by the same enterprises have the same operation strength and service capability, which is different from the facts: firstly, the electronic commerce has the case of shell borrowing operation, and the actual control persons behind the shops operated by the same enterprise are not the same; secondly, even the shops of the same business enterprise and the same actual person have essential differences due to the difference of years and industries of the shops.

Disclosure of Invention

The invention aims to provide a method and a system for taking a store of an electronic commerce platform as a research object, which can image and authenticate the store so as to objectively and intuitively reflect the overall operation condition of the store and provide a system reference for a buyer of the electronic commerce platform.

To achieve this object, some embodiments of the present application provide an electronic shop portrait authentication method including the steps of: step one: extracting data related to the store to obtain an original data set; and step two: and based on the original data set, portraying the store from a plurality of feature dimensions by utilizing feature engineering.

Some embodiments of the present application provide an electronic shop portrait authentication method comprising the steps of: step one: extracting data related to the store to obtain an original data set; step two: checking the authenticity of the data by using an identification rule model and/or a false transaction identification model on the original data set, and storing the data screened by the checking, wherein the screened data represents the real operation condition of the store; step three: portraying the store from a plurality of feature dimensions using a feature project based on the screened data; and step four: and quantitatively evaluating the overall business condition of the store by using a machine learning algorithm, and presenting the overall business condition in the form of scores and star grades.

Optionally, in the above method, the data is extracted after obtaining authorized consent of a seller of the store to meet the requirements in terms of privacy rules.

In the above method, the store-related data includes a raw data set such as the store basic information, the business data, and the fund data.

In the above method, the original data set may be data having the same structure or data having different structures.

In the above method, the raw data set is obtained from an e-commerce platform and/or a third party platform, such as a third party public information platform.

In the above method, the structure of the original data set may be: attribute indexes: such as operating duration, camping industry, etc.; evaluation index: such as good score, commodity score, etc.; commodity index: such as the number of goods, the commodity dynamic sales rate; order index: such as the amount of orders, etc., for the last 12 months; funding index: such as a refund amount, etc., for approximately 3 months; buyer index: such as the number of buyers, the buyers' repurchase rate, etc.; behavioral indexes such as number of times of occurrence, amount of occurrence, etc. in the month of the last 24 months; and compliance indicators such as the number of times the commodity was penalized for the last 6 months, the number of red cards, etc.

In the above method, the characteristic dimension includes one or more of six characteristic dimensions of qualification characteristics, commodity operation, customer quality, business capability, quality of service, and compliance history.

In some embodiments of the above method, the order data is checked, wherein the recognition rule model is established based on experience of recognition of false orders for many years, specifically, the recognition rule model uses logistics information, order information and the like to perform rule filtering on the false orders, and related methods are more in the prior art, and the recognition method of the false orders is not the key point of the invention; the false transaction recognition model is a machine learning model, and can be obtained by supervised training by using a neural network algorithm based on a large number of historical accumulated false transaction samples. The false transaction recognition model and the training method thereof are various in the prior art, and the model and the training method thereof are not the key points of the invention.

In the method, the performing feature engineering includes the steps of: performing data preprocessing on the screened data; performing feature selection on the preprocessed data to establish a feature matrix; and performing dimension reduction processing on the feature matrix.

Wherein the preprocessing includes one or more of normalizing the data, missing value calculation, and data transformation.

Wherein, the feature selection of the preprocessed data to establish the feature matrix includes scoring at least one feature determined in advance according to the correlation by using a filter feature selection method, and setting a threshold or the number of thresholds to be selected to obtain the feature matrix. Wherein the feature is associated with the data.

Wherein the dimension reduction processing of the feature matrix includes dimension reduction of the feature matrix using Principal Component Analysis (PCA).

Optionally, the image result is displayed in the form of word cloud.

Furthermore, the whole business condition of the shops can be quantitatively evaluated by using a machine learning algorithm and displayed in the form of scores and/or star marks. Specifically, whether the store breaks or closes is used as a dependent variable, other data of the store is used as an independent variable, the store breaking/breaking probability is predicted by using a logistic regression (logist ic recess ion) model, and the store breaking/breaking probability is converted into a score by using an s igmoid conversion, for example, the score between 300 and 999 is higher, and the probability representing the store breaking/breaking is lower.

Output results related to the store, such as a business file of the store and/or a store business card and/or a store evaluation report, can be automatically generated based on the customer representation and the result of the quantitative evaluation.

Alternatively, these output results may be shared online by the store itself.

Optionally, the business archive is configured to be automatically generated and recorded according to a preset major event and a corresponding threshold value. Optionally, the store business card is configured to be automatically generated in a predetermined format according to the store portrait and the result of the quantitative evaluation. Optionally, the evaluation report is automatically generated in the preset report template by filling the data of the specific store.

Alternatively, further, the authentication result of the store may be updated periodically or aperiodically.

Optionally, further, experience is accumulated in practice and the model is iteratively upgraded.

Further embodiments of the present application provide an electronic store shop image authentication system comprising a central processing unit configured to perform the electronic store shop image authentication method of any one of the above.

Still further embodiments of the present application provide a computing device comprising: the device comprises a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface complete communication with each other through the communication bus; the memory is configured to store at least one executable instruction, where the executable instruction causes the processor to perform operations corresponding to the method for authenticating an e-shop portrait according to any one of the above.

Still further embodiments of the present application provide a computer storage medium having stored therein at least one executable instruction for causing a processor to perform operations corresponding to the e-shop portrait authentication method as described in any one of the above.

The technical solutions in the different embodiments of the present application have one or more of the following advantages over the prior art:

the system in some embodiments of the invention utilizes the significant features extracted by the feature engineering to make the model more explanatory to the evaluation and authentication result of the store operation condition;

before modeling calculation, the system in some embodiments of the invention performs authenticity verification on the acquired store data, and only the data passing the authenticity verification participates in calculation, so that the influence of dirty data such as a bill order and the like on a calculation result can be prevented, the real operation condition of the store can be reflected, and the requirement that buyers and institution clients wish to know the real operation condition of the store as much as possible is met;

the system in some embodiments of the invention takes the store as a research object, distinguishes the store from a store business enterprise, and is more in line with the actual situation of the electronic market scene;

the system in some embodiments of the present invention provides the model with self-learning capabilities through a mechanism for manually marking the backfeeding model, such as may be found with a novel brush bill.

The system in some embodiments of the invention intuitively presents store features in the form of word clouds.

It is to be understood that the foregoing description is only an overview of the present invention, and is intended to provide a better understanding of the technical means of the present invention, as it is embodied in accordance with the present invention, and is intended to provide a better understanding of the above and other objects, features and advantages of the present invention, as it is embodied in the following specific examples.

Drawings

FIG. 1 is a flow chart of an electronic store shop image authentication method according to an embodiment of the present application;

FIG. 2 is a flow chart of data extraction steps in an electronic shop image authentication method according to an embodiment of the present application;

FIG. 3 is a flow chart of feature engineering of an electronic store shop image authentication method according to an embodiment of the present application;

FIG. 4 is a schematic diagram of a store feature word cloud drawn by an electronic store image authentication method according to an embodiment of the present application;

FIG. 5 is a schematic diagram of an authentication report obtained by an electronic store shop image authentication method according to an embodiment of the present application.

Detailed Description

Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

It should be appreciated that the store representation authentication method of the present application may be implemented in the form of a computer program on a computing system, such as a desktop, notebook, tablet, workstation, centralized, distributed server, cloud computing service, or the like. In particular, such a computer program may be deployed in a control system of an e-commerce platform as part of the control system.

The following will take a store named "Vikey13" as an example to show the steps of the method for authenticating images of a store according to the present application. "Vikey13" is a store on a home electronics business platform that is operated by its vendor. The buyer makes purchases through the store to generate orders, and pays the orders through a bank system or a payment system of the electronic commerce platform. The commodity in the order is delivered to the buyer through express service, fixed-point self-taking and other modes.

After determining that the system of the present application will be run for the store named "Vikey 13". As shown in fig. 1, first, the relevant data of the store is obtained step S101. This step may be further subdivided into several sub-steps, which as shown in fig. 2, may include sending a request to the manager of the store, also called the seller, to extract the data, step S1011. The request may be sent, for example, in the form of an internal message through the e-commerce platform's management system, or may be sent by external information, email, or other means. The request for extracting data may be a request for extracting all data associated with the store, or may be a request for extracting part of data associated with the store. The extracted data can come from an e-commerce platform where a store is located, and can come from other platforms, such as a receipt mechanism, a bank and the like. It should be understood that the authorization of the user is not a requirement of the system in the present application, and is generally understood to be a requirement for regulatory compliance, i.e., the system in some embodiments according to the present application may not be authorized by the seller of the store so long as all or part of the data associated with the store is available.

After the approval of the seller of the store is obtained, step S1012, the store image authentication system acquires the data accumulated in the electronic commerce platform and the acquirer by the store, including the store attribute, the order data, the fund data, and the like, step S1013. For example, the store gets 446 dimensions of data;

in some embodiments of the present application, all or part of the extracted data may be checked to improve accuracy of the basic data that is the portrait, step S104, as shown in fig. 1. The verification may be performed by means of external data using an identification rule model, for example, by means of an identification rule model, which may be used to filter false orders in order information by means of logistic information, and related methods are more numerous in the prior art, for example, the identification rule model may be used to identify a related order as a false order if certain order information cannot match successfully delivered logistic information. The method of false order identification itself is not the focus of the present invention.

The identification rule model may be performed, for example, for an order amount in order data, and the external data may be a borrowing record for the store-related account from the bank. Thereby checking whether the order amount is accurate.

A dummy transaction recognition model, which is a machine learning model that can be obtained by supervised training using neural network algorithms based on a large number of dummy transaction samples accumulated historically, can be further provided as a check of data authenticity.

For example, in the present embodiment, the partially extracted data may be checked. Verification may be performed for store order data to verify authenticity, for example 2092 orders for store for the last 2 years are identified by an identification rule model and a false transaction identification model, involving an order amount of 131796 dollars, and finally a total of 1841 dollars pass verification of authenticity, an order amount of 119934 dollars, i.e. an order amount of 88% authenticity and an order amount of 91% authenticity. After this step, the data that is not authentic in the extracted data, such as order data that is not authentic in order quantity and/or order amount, may be filtered out and filtered out.

The applicant finds that, in implementing the system of the present application, the authenticity check can improve the accuracy and referenceability of image authentication, but requires additional overhead and delay of speaking, and external data is required to be acquired from outside the platform, which may cause a decrease in the operation efficiency of the system. In fact, the data authenticity of stores of e-commerce platforms, such as applicant platforms, is high, and in some embodiments, images and certificates of stores may be directly made with the extracted data without authenticity verification.

After determining the data for performing the portrayal, for example, the extracted raw data or the filtered data, the store may be portrayed from a plurality of feature dimensions by using feature engineering based on these data, step S102, as shown in fig. 1. Specifically, the feature dimensions may include a plurality of feature dimensions, and the plurality of feature dimensions may be, for example, one or more of six feature dimensions of qualification characteristics, commodity operations, customer quality, business capabilities, quality of service, and compliance history.

A feature tag system may be built corresponding to the feature dimensions described above. The feature tag system may include multiple types of tags, such as one, two, or more. Each feature dimension may correspond to one or two or more types of feature labels.

For example, attribute tags corresponding to the qualification feature dimensions described above may be included; commodity labels corresponding to the commodity operation dimensions; customer labels corresponding to the customer quality dimensions; an operation label corresponding to the operation capability dimension; a service tag corresponding to the quality of service dimension; a compliance tag corresponding to the compliance history dimension; a seller tag corresponding to the qualification characteristic dimension and a risk tag corresponding to the compliance history dimension.

Different types of tags may include one or more options, and attribute tags may be such as "ten years old store", "new store"; the commodity label can be such as "high movable sales rate", "general movable sales rate", "low movable sales rate", etc.; customer labels such as "customer loyalty high", "customer loyalty general", "customer loyalty low", etc.; business labels such as "high growth", "low growth", "no high growth", etc.; service labels such as "high score", "general score", "low score", etc.; compliance tags such as "yellow cards", "red cards", "model shops", and the like; seller tags such as "e-commerce aged man", "e-commerce new and expensive", etc., and risk tags such as "red early warning", "yellow early warning", "blue early warning", etc.

In general, store features represented by each type of feature tag do not overlap each other to provide a stronger differentiation.

The specific flow of the feature engineering may include, as shown in fig. 3:

first, data preprocessing is performed on the extracted data/the filtered data, step S1021. The preprocessing includes normalizing the data, missing value calculation, and data transformation. Wherein the normalizing includes normalizing the various heterogeneous data into data in a uniform format for subsequent processing. The missing value calculation comprises possible value interpolation missing values to fill missing data under a standardized data structure, wherein the possible value interpolation missing values comprise mean value interpolation, and the like mean value interpolation, the maximum likelihood estimation and multiple interpolation are utilized. The data transformation includes discretizing the continuous attributes using a K-Means clustering algorithm and constructing new attributes using existing attribute sets.

Further, feature selection is performed on the preprocessed data to build a feature matrix, step S1022. The step of establishing the feature matrix comprises the step of scoring at least one feature determined in advance according to the relevance by using a filter type (Fi lter) feature selection method, and setting a threshold value or the number of threshold values to be selected to obtain the feature matrix. Wherein the feature is associated with the data. The filtering type feature selection method is based on the general performance of the features, such as target correlation, autocorrelation, divergence and the like. The method has the advantages that the feature selection calculation cost is small; the disadvantage is that if the feature subset is not selected for the learner to be used subsequently, the learner's ability to fit may be weak. When the filtering type characteristic selection method is used for examining the variable, whether the variable is filtered out is judged from the relationship between the condition of the single variable and the multiple variable. For univariates, the percent missing (Miss ing Percentage) can be from (1): missing the feature that the sample ratio is too high and difficult to fill, it is recommended to reject the variable. (2) Variance (Variance): if the variance of a continuous variable is close to 0, indicating that its characteristic value tends to a single value, the model is not helpful, and it is recommended to reject the variable. (3) Frequency (Frequency): if the enumeration value sample size of a certain type of variable is distributed in proportion to the ratio, the enumeration value is concentrated on a single enumeration value, and the variable is recommended to be removed. Three variables consider whether to cull. The relation between the multiple variables is mainly two, namely 1) the higher the correlation between the independent variables is, the multiple collinearity problem is caused, the stability of the model is further reduced, the sample tiny disturbance brings large parameter change, one of the characteristics with collinearity is suggested, and the rest is removed. 2) The higher the correlation between the independent and dependent variables, the more important the explanatory features are to model predictive goals, suggesting retention.

Further, the feature matrix is subjected to dimension reduction processing, step S1023: the feature matrix is reduced in dimension using Principal Component Analysis (PCA). The principal component analysis method comprises the steps of recombining a plurality of original P indexes with certain correlation into a new group of Q comprehensive indexes which are irrelevant to each other to replace the original P indexes. For example, the original P indices are linearly combined as a new overall index. For example, expressed in terms of the variance of a first composite index F1, where F1 is the first linear combination selected; that is, the larger Var (F1) is, the more information F1 contains. Therefore, F1 selected from all linear combinations should be the largest variance, so also called F1 is the first principal component. If the first principal component F1 is insufficient to represent the information of the original P indices, then consider selecting a second linear combination F2, i.e., the second principal component. In some embodiments, in order to effectively reflect the original information, the information existing in the first linear combination F1 is set to be not included in F2, i.e. Cov (F1, F2) =0 is required, and so on, the third, fourth, … …, and Q-th principal components can be constructed. The dimension reduction process can solve the problems of large calculation amount and long training time caused by overlarge feature matrix. After the dimension reduction treatment, the running time and the consumed system resources of the method are greatly reduced.

After the feature engineering is used to obtain the feature matrix after the dimension reduction, the feature label 205 may be extracted based on the feature matrix after the dimension reduction, step S1024. The feature tag is extracted from the feature tag system. The feature tag system may include multiple types of tags, for example, may include attribute tags, such as "ten year old store"; commodity labels such as "high dynamic sales rate"; customer labels, such as "customer loyalty high"; business labels, such as "high growth"; service tags, such as "high score"; compliance labels, such as "yellow cards"; a seller tag, such as "e-business senior" and a risk tag, such as "red warning" are comprised of one, two or more items.

For example, may be based on one or more of six characteristic dimensions of qualification, commodity operation, customer quality, business capability, quality of service, and compliance history.

Wherein each type of tag may correspond to one or more feature dimensions.

For the store named as "Vikey13", the feature labels such as "ten years old store", "excellent service", "cross-category management", "high customer loyalty", "good growth", "low bad experience" can be extracted by using the feature engineering steps, as shown by feature label 205 in FIG. 4. .

After the feature tag 205 is determined, the display method may be any of the prior art to display the feature tag to the user, step S1025. For example, all tags may be placed with their saliency level to generate a tag word cloud, i.e., feature tags with higher saliency level are placed in a larger font at the middle and/or upper layer of the word cloud image, while feature tags with lower saliency level are placed in a larger font at the edge and/or lower layer of the word cloud image, as shown in portrait word cloud 206 in fig. 4.

The representation of the store may be provided to the user in the form of an administration archive 100 and/or a store business card 200 and/or a store assessment report 300 in association with basic information of the store, step S103, as shown in fig. 1.

The business profile 100 and/or the business card 200 and/or the store assessment report 300 may also include other content. For example, in some embodiments of the present application, the image authentication method may further include inputting directly extracted data or filtered data of the store into a quantitative evaluation model to obtain a score and a rating of the store, step S105, as shown in fig. 1. The score of the store may be calculated from the data extracted for the store and the store may be ranked according to the score. The particular quantitative assessment model may be arbitrarily set and is not critical to the present application. According to these embodiments, as shown in FIG. 5, the store named "Vikey13" may have a score of 632 points, for example, and a score of two stars. Correspondingly, in this embodiment of the present application, the portrait authentication method may further include automatically generating the business record 100, the business card 200, and the authentication report 300 of the store according to the determined feature tag and/or the score and/or the rating result, step S103.

The business card 200 may be composed of, for example, a score 201, a rating 202, an industry rank 203, a six-dimensional radar chart 204, a feature tag 205, and a portrait word cloud 206, as shown in fig. 5. In some embodiments, the portrayal authentication system may be further configured to allow a buyer to review the business card 200 and/or allow a seller to share his business card 200 with a buyer or other partner entity in a single key.

The business file 100 records major events of the store since registration is established, such as successful registration, first order transaction, 100 ten thousand yuan breakthrough of transaction amount, etc., as shown in fig. 5. In some embodiments, the portrayal authentication system may be further configured to allow the buyer to review the business profile and/or allow the seller to share his business profile with the buyer or other partner entity in one-touch.

In some embodiments, information of the business, as well as, for example, scores, ranks, etc., may also be provided to the seller or other entity in the form of basic information 100 and/or presented with the business profile 100, business card 200, and store authentication report 300.

In some embodiments, as an optional step, the authentication result of the store may be updated periodically or aperiodically, S106, as shown in fig. 1.

The method and the system for authenticating the pictures of the electronic store shop are particularly suitable for cross-border electronic store shops which are difficult to examine and verify on site by users.

The algorithms or displays presented herein are not inherently related to any particular computer, virtual system, or other apparatus. Various general-purpose systems may also be used with the teachings herein. The required structure for a construction of such a system is apparent from the description above. In addition, embodiments of the present invention are not directed to any particular programming language. It will be appreciated that the teachings of the present invention described herein may be implemented in a variety of programming languages, and the above description of specific languages is provided for disclosure of enablement and best mode of the present invention.

In the description provided herein, numerous specific details are set forth. However, it is understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description.

Similarly, it should be appreciated that in the above description of exemplary embodiments of the invention, various features of the embodiments of the invention are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure and aiding in the understanding of one or more of the various inventive aspects. However, the disclosed method should not be construed as reflecting the intention that: i.e., the claimed invention requires more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of this invention.

Those skilled in the art will appreciate that the modules in the apparatus of the embodiments may be adaptively changed and disposed in one or more apparatuses different from the embodiments. The modules or units or components of the embodiments may be combined into one module or unit or component and, furthermore, they may be divided into a plurality of sub-modules or sub-units or sub-components. Any combination of all features disclosed in this specification (including any accompanying claims, abstract and drawings), and all of the processes or units of any method or apparatus so disclosed, may be used in combination, except insofar as at least some of such features and/or processes or units are mutually exclusive. Each feature disclosed in this specification (including any accompanying claims, abstract and drawings), may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise.

Furthermore, those skilled in the art will appreciate that while some embodiments herein include some features but not others included in other embodiments, combinations of features of different embodiments are meant to be within the scope of the invention and form different embodiments. For example, in the claims, any of the claimed embodiments may be used in any combination.

Various component embodiments of the invention may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art will appreciate that some or all of the functionality of some or all of the components according to embodiments of the present invention may be implemented in practice using a microprocessor or Digital Signal Processor (DSP). The present invention can also be implemented as an apparatus or device program (e.g., a computer program and a computer program product) for performing a portion or all of the methods described herein. Such a program embodying the present invention may be stored on a computer readable medium, or may have the form of one or more signals. Such signals may be downloaded from an internet website, provided on a carrier signal, or provided in any other form.

It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and that those skilled in the art will be able to design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention may be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by one and the same item of hardware. The use of the words first, second, third, etc. do not denote any order. These words may be interpreted as names. The steps in the above embodiments should not be construed as limiting the order of execution unless specifically stated.

Claims (10)

1. An electronic shop authentication method is characterized in that: the method includes the steps of extracting data related to the store to form an original dataset, evaluating the business condition of the store based on the original dataset, refining a feature tag of the store based on the evaluation result, and generating at least one of a store business card, a business profile and/or an authentication report of the store.

2. An electronic shop portrait authentication method is characterized in that: the method includes extracting data associated with the store to form an original dataset; checking the authenticity of the data by using an identification rule model and/or a false transaction identification model on the original data set, and storing the data screened by the checking, wherein the screened data represents the real operation condition of the store; and portraying the store from a plurality of feature dimensions using feature engineering based on the screened data.

3. The electronic shop image authentication method according to claim 1 or 2, characterized in that: the data is extracted after authorized consent of the seller of the store to meet the requirements in terms of privacy rules.

4. The electronic shop image authentication method according to claim 1 or 2, characterized in that: the store-related data includes the store basic information, business data, fund data and other original data sets.

5. The method for authenticating an electronic shop portrait according to claim 4, characterized in that: the original data sets are data having the same structure or data having different structures.

6. The method for authenticating an electronic shop portrait according to claim 5, characterized in that: the raw data set is obtained from an e-commerce platform and/or a third party platform.

7. The electronic shop portrait authentication method according to any one of the preceding claims, characterised in that: the operation file is configured to be automatically generated and recorded according to a preset major event and a corresponding threshold value;

the store business card is configured to automatically generate according to a predetermined format according to store images and the result of quantitative evaluation;

the evaluation report is automatically generated in the data of the filling specific shops in the preset report template.

8. The shop portrait authentication system is characterized in that: comprising a central processing unit configured to perform the shop portrait authentication method according to any one of claims 1 to 7.

9. A computing device, comprising: the device comprises a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface complete communication with each other through the communication bus; the memory is configured to store at least one executable instruction that causes the processor to perform operations corresponding to the method for authenticating an e-shop portrait according to any one of claims 1 to 7.

10. A computer storage medium having stored therein at least one executable instruction for causing a processor to perform operations corresponding to the e-shop portrait authentication method according to any one of claims 1 to 7.