CN116629917B

CN116629917B - Shop feature application method and device, equipment and medium thereof

Info

Publication number: CN116629917B
Application number: CN202310575523.8A
Authority: CN
Inventors: 谭华
Original assignee: Guangzhou Shangyan Network Technology Co ltd
Current assignee: Guangzhou Shangyan Network Technology Co ltd
Priority date: 2023-05-19
Filing date: 2023-05-19
Publication date: 2024-01-30
Anticipated expiration: 2043-05-19
Also published as: CN116629917A

Abstract

The application relates to a shop feature application method, a device, equipment and a medium thereof in the technical field of electronic commerce, wherein the method comprises the following steps: acquiring store data of an online store, wherein the store data comprises configuration information, access data, transaction data and after-sales service data; extracting basic characteristics according to store data, extracting operation characteristics according to access data, transaction data and configuration information in the store data, and constructing the basic characteristics and the operation characteristics into store image characteristics of the online store, wherein the basic characteristics are used for representing the online store operation conditions, and the operation characteristics are used for representing online store operation effects; inputting the store image features into a store classification model to obtain the store type of the online store; and adjusting a basic operation strategy corresponding to the store type according to the operation characteristics to obtain the customized operation strategy of the online store. The method and the device determine a practical operation strategy by applying multidimensional store portrait features.

Description

Shop feature application method and device, equipment and medium thereof

Technical Field

The present disclosure relates to the field of electronic commerce technologies, and in particular, to a store feature application method, and a corresponding apparatus, computer device, and computer readable storage medium thereof.

Background

In the field of electronic commerce, constructing the store characteristics of online stores is an important business, and the online stores can be characterized on the characteristic data level, so that the constructed store characteristics are applied to classify the types of the online stores, and corresponding operation strategies are adopted according to different store types.

In the prior art, the store characteristics of the online store are usually constructed manually, however, the problem of small dimension of the store characteristics occurs, the online store is difficult to comprehensively characterize, the accuracy of classifying the store types of the online store is seriously affected, and an effective operation strategy cannot be adopted.

In view of the defects of the traditional technology, the applicant has long been engaged in research in the related field, and is in order to solve the problem in the field of electronic commerce, so a new way is developed.

Disclosure of Invention

It is a primary object of the present application to solve at least one of the above problems and to provide a store feature application method and corresponding apparatus, computer device, computer readable storage medium.

In order to meet the purposes of the application, the application adopts the following technical scheme:

a store feature application method provided in accordance with one of the objects of the present application, comprising the steps of:

Acquiring store data of an online store, wherein the store data comprises configuration information, access data, transaction data and after-sales service data;

extracting basic characteristics according to the store data, extracting operation characteristics according to access data, transaction data and configuration information in the store data, and constructing the basic characteristics and the operation characteristics into store image characteristics of the online store, wherein the basic characteristics are used for representing the online store operation conditions, and the operation characteristics are used for representing online store operation effects;

inputting the store image features into a store classification model to obtain the store type of the online store;

and adjusting a basic operation strategy corresponding to the store type according to the operation characteristics to obtain the customized operation strategy of the online store.

In a further embodiment, extracting basic features according to the store data, where the basic features include statistical features, user group distribution features, and hierarchical features, extracting according to the following process includes:

counting according to the store data and with preset time granularity correspondence, obtaining statistical characteristics corresponding to each time granularity, wherein the statistical characteristics represent the operating conditions of any one aspect or any multiple aspects of the store in sales, advertising marketing and after-sales services under corresponding time spans;

Clustering is carried out according to the access data in the store data by adopting a clustering algorithm, and the number of users and the total ratio of the users corresponding to a plurality of preset user clusters are determined to be used as user cluster distribution characteristics;

layering is carried out according to access data and transaction data in the store data by adopting a preset step-type threshold interval, layering characteristics are determined, the layering characteristics represent levels corresponding to a plurality of operation key indexes of the store on a line, and the levels represent the quality of the operation key indexes.

In a further embodiment, the operational characteristics include a periodic repeating characteristic that characterizes a periodic law of sales for the online store, a fast-rise characteristic that characterizes a fast-growth fluctuation of sales for the online store, and a peer competitiveness characteristic that characterizes a peer's competitiveness for the online store.

In a further embodiment, the periodically repeated feature of the operation features is extracted according to the access data and the transaction data in the store data, and the method comprises the following steps:

acquiring the access number in the access data in the store data of the preset time granularity, and the amount of orders in the transaction data;

Determining whether each first time window in the preset duration is similar to the access number, the order number and the order amount corresponding to the first time window, and obtaining the number of the first time windows corresponding to the similar access number, the similar order number and the similar order amount;

determining whether each second time window in the preset duration is similar to the access number, the order number and the order amount corresponding to the last second time window respectively, and obtaining the number of the second time windows corresponding to the similar access number, the similar order number and the similar order amount;

and constructing the number of the first time windows and the number of the second time windows corresponding to the similar access number, the similar order number and the similar order amount as periodic repeated characteristics.

In a further embodiment, the method for extracting the quick start feature from the operation features according to the access data and the transaction data in the store data comprises the following steps:

determining the maximum value and the minimum value of the access number, the order quantity and the order amount corresponding to each first time window in the preset duration;

And correspondingly determining the number of the first time windows corresponding to the access number quick start, the order number quick start and the order amount quick start according to the maximum value and the minimum value of the access number, the order number and the order amount respectively, and constructing the quick start feature.

In a further embodiment, the method for extracting the peer-to-peer competitive feature from the operation features according to the configuration information, the access data and the transaction data in the store data includes the following steps:

determining the same-line online store group of the online stores according to the configuration information in the store data;

determining an order amount average value, a repurchase rate and a conversion rate according to the access data and the transaction data in the store data;

and determining the corresponding grades of the order amount average value, the repurchase rate and the conversion rate according to the average order amount average value, the average repurchase rate and the average conversion rate of the shop group on the same line, and constructing the same-line competitive power characteristics.

In a further embodiment, the basic features further include a combination feature and a multidimensional feature, and the method comprises the following steps of:

determining offline sales channel characteristics and advertisement marketing channel characteristics according to configuration information of store data to perform characteristic combination to obtain combination characteristics, wherein the offline sales channel characteristics represent whether online stores have offline sales channels or not, and the advertisement marketing channel characteristics represent whether online stores have advertisement marketing channels or not;

And determining the distribution of customer complaints reasons and the distribution of satisfaction according to the after-sales service data of the store data, and correspondingly performing dimension reduction processing to obtain multidimensional features.

On the other hand, the store feature application device provided by adapting to one of the purposes of the application comprises a data acquisition module, a feature extraction module, a store classification module and a strategy customization module, wherein the data acquisition module is used for acquiring store data of an online store, and the store data comprises configuration information, access data, transaction data and after-sales service data; the feature extraction module is used for extracting basic features according to the store data, extracting operation features according to access data, transaction data and configuration information in the store data, constructing the basic features and the operation features into store image features of the online store, wherein the basic features are used for representing the operation conditions of the online store, and the operation features are used for representing the operation effects of the online store; the store classification module is used for inputting the store image characteristics into a store classification model to obtain the store type of the online store; and the strategy customizing module is used for adjusting the basic operation strategy corresponding to the store type according to the operation characteristics to obtain the customized operation strategy of the online store.

In a further embodiment, the feature extraction module includes: the statistical characteristic sub-module is used for carrying out statistics according to the store data and corresponding to a plurality of preset time granularities to obtain statistical characteristics corresponding to each time granularity, wherein the statistical characteristics represent the operating conditions of any one aspect or any multiple aspects of the store in sales, advertisement marketing and after-sales service under the corresponding time span; the user group distribution characteristic sub-module is used for clustering by adopting a clustering algorithm according to the access data in the store data, and determining the number of users corresponding to a plurality of preset user groups and the total ratio of the users as user group distribution characteristics; and the layering characteristic sub-module is used for layering by adopting a preset step-type threshold interval according to the access data and the transaction data in the store data to determine layering characteristics, wherein the layering characteristics represent the levels corresponding to a plurality of operation key indexes of the store on the line, and the levels represent the quality of the operation key indexes.

In a further embodiment, the operation features in the feature extraction module include a periodic repeating feature, a fast starting feature, and a peer competitiveness feature, where the periodic repeating feature characterizes a periodic rule of online stores in terms of sales, the fast starting feature characterizes fast growth fluctuations of online stores in terms of sales, and the peer competitiveness feature characterizes competitiveness of online stores in terms of peer.

In a further embodiment, the feature extraction module includes: the granularity data acquisition sub-module is used for acquiring the access number in the access data in the store data of the preset time granularity, and the order quantity and the order amount in the transaction data; zhou Guilv determining submodule, configured to determine whether each first time window in the preset duration is similar to the access number, the order number and the order amount corresponding to the previous first time window, and obtain the number of first time windows corresponding to the similar access number, the similar order number and the similar order amount; the month rule determining submodule is used for determining whether each second time window in the preset duration is similar to the access number, the order number and the order amount corresponding to the last second time window respectively, and obtaining the number of the second time windows corresponding to the similar access number, the similar order number and the similar order amount; and the periodic repeated feature construction submodule is used for constructing the number of the first time windows and the number of the second time windows corresponding to the similar access number, the similar order number and the similar order amount as periodic repeated features.

In a further embodiment, the feature extraction module includes: the granularity data acquisition sub-module is used for acquiring the access number in the access data in the store data of the preset time granularity, and the order quantity and the order amount in the transaction data; the maximum value determining submodule is used for determining the maximum value and the minimum value of the access number, the order quantity and the order amount corresponding to each first time window in the preset duration; and the quick start characteristic construction submodule is used for correspondingly determining the number of the first time windows corresponding to the quick start of the access number, the quick start of the order number and the quick start of the order amount according to the maximum value and the minimum value of the access number, the order number and the order amount respectively, and constructing the quick start characteristic.

In a further embodiment, the feature extraction module includes: the peer determination submodule is used for determining peer online store groups of the online stores according to the configuration information in the store data; the business effect determining submodule is used for determining the average value of the order amount, the repurchase rate and the conversion rate according to the access data and the transaction data in the store data; and the peer competitive feature construction submodule is used for determining the grades corresponding to the order amount average value, the repurchase rate and the conversion rate according to the average order amount average value, the average repurchase rate and the average conversion rate of the shop group on the peer line and constructing the peer competitive feature.

In a further embodiment, the feature extraction module includes: the combined feature sub-module is used for determining offline sales channel features and advertisement marketing channel features according to the configuration information of the store data to perform feature combination, so as to obtain combined features, wherein the offline sales channel features represent whether online stores have offline sales channels or not, and the advertisement marketing channel features represent whether online stores have advertisement marketing channels or not; and the multidimensional feature sub-module is used for determining the distribution of customer complaints reasons and the distribution of satisfaction according to the after-sales service data of the store data and correspondingly carrying out dimension reduction processing to obtain multidimensional features.

In yet another aspect, a computer device is provided, adapted for one of the objects of the present application, comprising a central processor and a memory, the central processor being adapted to invoke the steps of running a computer program stored in the memory to perform the store feature application method described herein.

In yet another aspect, a computer readable storage medium adapted to another object of the present application is provided, in the form of computer readable instructions, storing a computer program implemented according to the store feature application method, which when invoked by a computer, performs the steps comprised by the method.

The technical solution of the present application has various advantages, including but not limited to the following aspects:

according to the method, basic characteristics are extracted according to store data, operating characteristics are extracted according to access data, transaction data and configuration information in the store data, the basic characteristics and the operating characteristics are constructed to be store portrait characteristics of online stores, the basic characteristics are used for representing online store operating conditions, the operating characteristics are used for representing online store operating effects, the store portrait characteristics are input into a store classification model to obtain store types of the online stores, the store types are adjusted according to the operating characteristics to correspond to preset basic operating strategies, and the customized operating strategies of the online stores are obtained. On one hand, multidimensional store image features are extracted based on store data rich in online stores, the operation conditions and operation effects of the online stores are comprehensively and accurately reflected, the accurately classified store types are ensured, on the other hand, corresponding effective basic operation strategies are determined according to the accurate store types, and further adjustment is carried out according to the operation features, so that the practicability of the operation strategies is greatly improved.

Drawings

The foregoing and/or additional aspects and advantages of the present application will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings, in which:

FIG. 1 is a flow chart of an exemplary embodiment of a store feature application method of the present application;

fig. 2 is a schematic flow chart of extracting statistical features, user group distribution features and layering features from basic features in an embodiment of the present application;

FIG. 3 is a schematic flow chart of extracting periodically repeated features from operating features according to an embodiment of the present application;

FIG. 4 is a flow chart of extracting quick start features from operating features according to an embodiment of the present application;

FIG. 5 is a schematic flow chart of extracting competitive features of the same row in operating features according to an embodiment of the present application;

FIG. 6 is a flow chart of extracting combined features and multidimensional features from basic features in an embodiment of the present application;

FIG. 7 is a schematic block diagram of a store feature application apparatus of the present application;

fig. 8 is a schematic structural diagram of a computer device used in the present application.

Detailed Description

Embodiments of the present application are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals refer to the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the drawings are exemplary only for the purpose of illustrating the present application and are not to be construed as limiting the present application.

As used herein, the singular forms "a", "an", "the" and "the" are intended to include the plural forms as well, unless expressly stated otherwise, as understood by those skilled in the art. It will be further understood that the terms "comprises" and/or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. It will be understood that when an element is referred to as being "connected" or "coupled" to another element, it can be directly connected or coupled to the other element or intervening elements may also be present. Further, "connected" or "coupled" as used herein may include wirelessly connected or wirelessly coupled. The term "and/or" as used herein includes all or any element and all combination of one or more of the associated listed items.

It will be understood by those skilled in the art that all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs unless defined otherwise. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the prior art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.

As used herein, "client," "terminal device," and "terminal device" are understood by those skilled in the art to include both devices that include only wireless signal receivers without transmitting capabilities and devices that include receiving and transmitting hardware capable of two-way communication over a two-way communication link. Such a device may include: a cellular or other communication device such as a personal computer, tablet, or the like, having a single-line display or a multi-line display or a cellular or other communication device without a multi-line display; a PCS (Personal Communications Service, personal communication system) that may combine voice, data processing, facsimile and/or data communication capabilities; a PDA (Personal Digital Assistant ) that can include a radio frequency receiver, pager, internet/intranet access, web browser, notepad, calendar and/or GPS (Global Positioning System ) receiver; a conventional laptop and/or palmtop computer or other appliance that has and/or includes a radio frequency receiver. As used herein, "client," "terminal device" may be portable, transportable, installed in a vehicle (aeronautical, maritime, and/or land-based), or adapted and/or configured to operate locally and/or in a distributed fashion, at any other location(s) on earth and/or in space. As used herein, a "client," "terminal device," or "terminal device" may also be a communication terminal, an internet terminal, or a music/video playing terminal, for example, a PDA, a MID (Mobile Internet Device ), and/or a mobile phone with music/video playing function, or may also be a device such as a smart tv, a set top box, or the like.

The hardware referred to by the names "server", "client", "service node" and the like in the present application is essentially an electronic device having the performance of a personal computer, and is a hardware device having necessary components disclosed by von neumann's principle, such as a central processing unit (including an arithmetic unit and a controller), a memory, an input device, and an output device, and a computer program is stored in the memory, and the central processing unit calls the program stored in the external memory to run in the memory, executes instructions in the program, and interacts with the input/output device, thereby completing a specific function.

It should be noted that the concept of "server" as referred to in this application is equally applicable to the case of a server farm. The servers should be logically partitioned, physically separate from each other but interface-callable, or integrated into a physical computer or group of computers, according to network deployment principles understood by those skilled in the art. Those skilled in the art will appreciate this variation and should not be construed as limiting the implementation of the network deployment approach of the present application.

One or several technical features of the present application, unless specified in the plain text, may be deployed either on a server to implement access by remotely invoking an online service interface provided by the acquisition server by a client, or directly deployed and run on the client to implement access.

The neural network model cited or possibly cited in the application can be deployed on a remote server and used for implementing remote call on a client, or can be deployed on a client with sufficient equipment capability for direct call unless specified in a clear text, and in some embodiments, when the neural network model runs on the client, the corresponding intelligence can be obtained through migration learning so as to reduce the requirement on the running resources of the hardware of the client and avoid excessively occupying the running resources of the hardware of the client.

The various data referred to in the present application, unless specified in the plain text, may be stored either remotely in a server or in a local terminal device, as long as it is suitable for being invoked by the technical solution of the present application.

Those skilled in the art will appreciate that: although the various methods of the present application are described based on the same concepts so as to be common to each other, the methods may be performed independently, unless otherwise indicated. Similarly, for each of the embodiments disclosed herein, the concepts presented are based on the same inventive concept, and thus, the concepts presented for the same description, and concepts that are merely convenient and appropriately altered although they are different, should be equally understood.

The various embodiments to be disclosed herein, unless the plain text indicates a mutually exclusive relationship with each other, the technical features related to the various embodiments may be cross-combined to flexibly construct a new embodiment, so long as such combination does not depart from the inventive spirit of the present application and can satisfy the needs in the art or solve the deficiencies in the prior art. This variant will be known to the person skilled in the art.

A store feature application method of the present application may be programmed as a computer program product that is deployed to run in a client or server, for example, in an exemplary application scenario of the present application, may be deployed in a server of an e-commerce platform, whereby the method may be performed by accessing an interface that is open after the computer program product is run, and performing man-machine interaction with a process of the computer program product through a graphical user interface.

Referring to fig. 1, in an exemplary embodiment, the store feature application method of the present application includes the following steps:

step S1100, acquiring shop data of an online shop, wherein the shop data comprises configuration information, access data, transaction data and after-sales service data;

Merchants can create and maintain their online stores on the e-commerce platform, which can provide access links to the online stores for consumers of the e-commerce platform to access and sell the online stores on-shelf goods to the consumers. Merchants may also create and maintain their own independent stations on which to deploy online stores to sell the online store-crafted merchandise to consumers accessing the independent stations.

Store databases are typically created and maintained for storing store data for online stores and corresponding data access interfaces are provided for accessing the store databases to obtain the store data, which data access interfaces and store databases are flexibly adaptable by those skilled in the art.

The configuration information may be basic information of the online store, typically edited by the merchant when creating the online store, and/or generated by the e-commerce platform, including specifically, first online time, camping commodity categories, whether there is an online off-sales channel, whether there is an advertising marketing channel.

The access data are data which are correspondingly generated when the online store is accessed, and specifically comprise an access session and an access source. The access session refers to a series of interaction processes performed by a user during one access to an online store, and generally includes any one or any multiple of operations such as browsing, clicking, searching, placing an order, and purchasing one or more pages.

The transaction data is data correspondingly generated when goods in shops on sales outlets are sold, and specifically comprises order amount and order creation time of each order.

The after-sales service data is data generated by providing a user feedback service corresponding to the commodity sold by a shop, and specifically comprises a customer complaint reason, a customer complaint amount, a customer complaint processing time and satisfaction, wherein the customer complaint reason is the content of complaints of the commodity purchased by a user, the customer complaint processing time is the waiting time from the time when the customer complains of the commodity purchased by the customer to the time when an operator of the online shop carries out corresponding reply processing, the satisfaction is an evaluation result of the satisfaction of the user on the commodity purchased by the customer, and the evaluation can be specifically carried out in a manner of scoring, evaluating the satisfaction and the like in any aspect such as the aspect of description compliance, logistics service, overall service situation and the like, and when the evaluation of the satisfaction involves multiple aspects, the corresponding average value can be taken as the evaluation result to comprehensively integrate all aspects.

Step 1200, extracting basic features according to the store data, extracting operation features according to access data, transaction data and configuration information in the store data, and constructing the basic features and the operation features into store image features of the online store, wherein the basic features are used for representing online store operation conditions, and the operation features are used for representing online store operation effects;

The basic features comprise statistical features, user group distribution features and layering features;

counting according to the order amount and the order creation time of each order in the access data in the store data and with preset time granularity, and correspondingly obtaining the corresponding statistical characteristics of each time granularity, which characterize the business conditions of the store on the sales aspect on the off-line of the corresponding time span; statistics can be carried out according to access sources in access data in store data and with preset multiple time granularities, and statistics features corresponding to each time granularity and representing business conditions of the store on the aspect of advertisement marketing on the line in corresponding time spans are correspondingly obtained; according to the customer complaint processing time in the after-sales service data in the store data, statistics is carried out according to preset various time granularities, so that corresponding mean values are determined, and statistical characteristics of the store on the aspect of after-sales service on the offline in a corresponding time span, which are corresponding to each time granularity, are correspondingly obtained. The time granularity can be any of about 10 days, about 30 days, about 90 days, about half a year and about half a year after the on-line shop operation, and can be set by a person skilled in the art as required;

And according to the access session in the access data in the store data, correspondingly determining users accessing the online store, clustering by adopting a clustering algorithm based on all the access sessions corresponding to each user, and determining the number of users corresponding to a plurality of preset user clusters and the total ratio of the users as the user cluster distribution characteristics. The plurality of user clusters may be preset as desired by those skilled in the art, with new user clusters, old user clusters, lost user clusters, and repurchase user clusters being exemplary examples, respectively. The user in the new user group can be a new user for purchasing online store commodity for the first time, the old user group can be an old user for purchasing online store commodity for many times, the user in the lost user group can be a lost user which does not purchase again for a long time after purchasing online store commodity, and the user in the re-purchase user group can be a re-purchase user which purchases online store commodity again. The clustering algorithm can be k-means, KNN and the like, and can be realized by one of the technicians according to the need;

according to the access session in the access data in the store data, the order creation time and the order amount of each order in the transaction data, any one or more of GMV (commodity transaction total), conversion rate, guest price and user quantity of access with preset time granularity are determined as operation key indexes, each operation key index is layered by adopting a step-type threshold interval corresponding to each operation key index, corresponding layers are determined as layering characteristics, and the layers represent the quality of the operation key indexes, such as high, medium and low. The user scale may be the number of users accessing the online store or the number of users purchasing the online store commodity, the preset time granularity may be any one of the last half year, the last one year and the last three years, and may be set by a person skilled in the art as required, and the preset threshold may be set by a person skilled in the art as required according to the disclosure herein.

The operation characteristics comprise a periodic repetition characteristic, a quick starting characteristic and a peer competitiveness characteristic;

the periodic repeating characteristic represents a periodic rule of a shop on the line in sales, whether the access number, the order number and the order amount in the period corresponding to adjacent periods are respectively similar or not is correspondingly determined based on the access number, the order number and the order amount in the access data and the order amount in the shop data corresponding to fine time granularity in each period in a certain period, the certain period can be set to be any one of the last half year and the last half year of the on-line shop operation, the period can be set to be any one of a week and a month, the fine time granularity can be any one of one day, two days and three days relative to the period, and the fine time granularity can be set by a person skilled in the art according to the disclosure;

the rapid-rise characteristic characterizes rapid-rise fluctuation of shops on the line in sales, which can be obtained by correspondingly determining whether the access number between adjacent periods is rapid, whether the order number is rapid and whether the order amount is rapid or not based on the access number in the access data, the order number in the transaction data and the order amount in the store data corresponding to fine time granularity in each period, wherein the number of windows corresponding to the rapid-rise number, the rapid-rise number of the order number and the rapid-rise number of the order amount are constructed as rapid-rise characteristics, the period can be set as a specific time window, such as any one of one week and one month, and the fine time granularity can be any one of one day, two days and three days relative to the period, and can be set as required by a person skilled in the art according to disclosure;

The competitive power of the online stores in the same line can be represented by comparing the average order amount average value, the average repurchase rate and the average conversion rate of the online stores which are the same line with the online stores, the corresponding difference corresponding to the order amount average value, the repurchase rate and the conversion rate of the online stores is obtained, the corresponding rating structure is determined to be the competitive power of the online stores, the order amount average value, the repurchase rate and the conversion rate are obtained by accessing data and transaction data in store data with preset time granularity of the online stores, the corresponding order amount average value, the repurchase rate and the conversion rate can be obtained by a plurality of online stores of the same line in the same line, the average order amount average value, the average repurchase rate and the average conversion rate can be obtained by averaging the number of the online stores, the preset time granularity can be any one of the near half year, the near one year and the near three years, and the technical field can be set according to the requirements.

Step S1300, inputting the store image characteristics into a store classification model to obtain the store type of the online store;

In one embodiment, the model structure of the store classification model is a Text feature extraction layer followed by a classifier, the Text feature extraction layer may be a model suitable for extracting Text features in the NLP field, the recommended selection is a BERT model, or any other model such as Text fransfomer, roBERTa, XLM-RoBERTa, MPNet, etc. may be used. The classifier can be FC (fully connected layer), MLP (multi-layer perceptron), etc., and can be implemented alternatively as needed by those skilled in the art.

Training the shop classification model to converge in advance, obtaining the capability of classifying the shop type of the online shop based on the shop image characteristics of the online shop, specifically, in one embodiment, invoking a single training sample in the training set and a supervision label thereof by acquiring a preset training set, inputting the training sample into the shop classification model, extracting deep semantic information of the training sample by the text characteristic extraction layer, outputting text characteristic vectors representing the semantics of the training sample in a vectorization mode, receiving the text characteristic vectors by a classifier, mapping the text characteristic vectors to preset multiple shop types, obtaining classification probability corresponding to each shop type, outputting the shop type with the largest classification probability as a prediction classification result of the training sample, invoking a preset cross entropy loss function, and calculating the cross entropy loss value of the prediction result based on the supervision label according to the training sample by a person skilled in the art according to priori knowledge or experimental experience; when the cross entropy loss value reaches a preset threshold value, indicating that the shop classification model is trained to a convergence state, so that model training can be terminated; when the cross entropy loss value does not reach the preset threshold, the model is indicated to be not converged, gradient update is carried out on the model according to the cross entropy loss value, the model is further approximated to convergence by correcting weight parameters of each link of the model through back propagation, then iteration training is carried out on the model by continuously calling other training samples in the training set and supervision labels thereof until the model is trained to be in a convergence state, and the preset threshold can be set by a person skilled in the art according to requirements.

The training set is constructed in advance, wherein a training sample is a shop portrait feature of an online shop, and the type of the shop to which the online shop belongs is manually determined according to the shop portrait feature to be a supervision label of the training sample.

And step 1400, adjusting a basic operation strategy corresponding to the store type according to the operation characteristics to obtain the customized operation strategy of the online store.

The basic operation strategies corresponding to the multiple store types can be preset by the skilled person according to flexible variation of business needs, and the corresponding basic operation strategies are, for example, the store types with good operation conditions and poor operation effects, and the corresponding basic operation strategies are to expand advertisement marketing channels or increase exposure on the existing advertisement marketing channels; the operation conditions are poor and the operation effects are poor, and the corresponding basic operation strategies are to comprehensively and greatly improve the visual page effect of shops, commodity pictures and texts, advertisement marketing and the corresponding quality of after-sales service, expand a large number of advertisement marketing channels or increase a large number of exposure on the existing advertisement marketing channels; the corresponding basic management strategies are to comprehensively and slightly improve the visual page effect of shops, commodity pictures and texts, advertisement marketing and corresponding quality of after-sales service, and expand a small amount of advertisement marketing channels or select a popular advertisement marketing channel to incline more exposure in the existing advertisement marketing channels; the corresponding basic management strategy is to comprehensively and slightly improve the corresponding quality of visual pages, commodity pictures and texts, advertisement marketing and after-sales service of the store; the store type with good operation condition and good operation effect has the corresponding basic operation strategy of managing and controlling the corresponding quality of visual page effect, commodity pictures and texts, advertisement marketing and after-sale service of the store and keeping the quality stable.

Further, the person skilled in the art can flexibly adjust the basic operation policy corresponding to the store type according to the operation characteristics according to the service requirement, and as an exemplary example, the operation policy is added on the basis of the basic operation policy according to the operation characteristics, the period rule of the online store shown by the operation characteristics is not outstanding or is not provided, the quick quantity is less or the operation effect is not enough, and the operation policy is added in the period that the corresponding industry general sales quantity is more is determined according to the category of the commodity of the online store, and the promotion frequency is increased compared with that of the online store of the same line during the period; for online stores with little or no quick play, low peer competitiveness or generally insufficient business results, the operation strategy may be to determine a period with a relatively large amount of common sales in the industry according to the category of the commodity of the online store, and during this period, the sales promotion discount force is increased compared with the online store of the same peer.

As can be appreciated from the exemplary embodiments of the present application, the technical solution of the present application has various advantages, including but not limited to the following aspects:

according to the method, basic characteristics are extracted according to store data, operating characteristics are extracted according to access data, transaction data and configuration information in the store data, the basic characteristics and the operating characteristics are constructed to be store portrait characteristics of online stores, the basic characteristics are used for representing online store operating conditions, the operating characteristics are used for representing online store operating effects, the store portrait characteristics are input into a store classification model to obtain store types of the online stores, the store types are adjusted according to the operating characteristics to correspond to preset basic operating strategies, and the customized operating strategies of the online stores are obtained. On one hand, multidimensional store image features are extracted based on store data rich in online stores, the operation conditions and operation effects of the online stores are comprehensively and accurately reflected, the accurately classified store types are ensured, on the other hand, corresponding effective basic operation strategies can be determined in a targeted mode, and further adjustment is carried out according to the operation features, so that the practicability of the operation strategies is improved.

Referring to fig. 2, in a further embodiment, step S1200 extracts basic features according to the store data, where the basic features include statistical features, user group distribution features, and hierarchical features, and the extracting includes:

step S1210, counting according to the store data and with preset time granularity, obtaining statistical characteristics corresponding to each time granularity, wherein the statistical characteristics represent the operation conditions of any one aspect or any multiple aspects of the store in sales, advertisement marketing and after-sales services under the corresponding time span;

the various time granularities may be any of about 10 days, about 30 days, about 90 days, about half a year, and about the last half of an on-line shop business, and may be set as desired by one skilled in the art.

The statistics of the operation condition of the online store in sales aspect can be performed by counting the number of orders and the amount of the orders obtained by adding all the orders within the corresponding time granularity according to the order creation time of each order in the access data; and counting the statistical characteristics representing the business conditions of the online shops in the aspect of advertisement marketing, counting the total number of each flow source and all flow sources in the corresponding time granularity according to the access sources in the access data, namely the flow sources, further calculating the total ratio of each flow source corresponding to the total number of all flow sources respectively, and obtaining the flow concentration distribution. And counting the statistical characteristics representing the business conditions of the online shops in the aspect of after-sales service, and counting the average value obtained by averaging all customer complaint processing time lengths in the corresponding time granularity according to the after-sales service data.

Step S1220, clustering by using a clustering algorithm according to the access data in the store data, and determining the number of users and the total ratio of the users corresponding to the preset plurality of user clusters as the user cluster distribution characteristics;

and determining users accessing the online stores according to the access session in the access data, clustering all users by adopting a clustering algorithm based on all access sessions corresponding to each user, classifying user clusters to which each user belongs, determining the number of users corresponding to each user cluster, further calculating the sum of the number of users corresponding to each user cluster and the number of users corresponding to all user clusters, and correspondingly obtaining the corresponding user occupation ratio of each user cluster to obtain the user cluster distribution characteristics. A person skilled in the art may preset a plurality of the user clusters as required, and exemplary examples are a new user cluster, an old user cluster, a lost user cluster, and a repurchase user cluster, respectively. The user in the new user group can be a new user for purchasing online store commodity for the first time, the old user group can be an old user for purchasing online store commodity for many times, the user in the lost user group can be a lost user which does not purchase again for a long time after purchasing online store commodity, and the user in the re-purchase user group can be a re-purchase user which purchases online store commodity again. The clustering algorithm can be k-means, KNN and the like, and can be realized by one skilled in the art according to the need.

Step S1230, layering is performed by adopting a preset step-type threshold interval according to the access data and the transaction data in the store data, so as to determine layering characteristics, wherein the layering characteristics represent the levels corresponding to a plurality of operation key indexes of the store on the line, and the levels represent the quality of the operation key indexes.

In this embodiment, by extracting statistical features, user group distribution features, and layering features corresponding to various time granularities according to store data of online stores, on one hand, the operating conditions corresponding to the online stores in terms of users, sales, advertising, after-sales service, and operation key indexes are fully and deeply mined, and on the other hand, a foundation can be laid for the accuracy of store types obtained by classifying by using a store classification model in the following manner.

Referring to fig. 3, in a further embodiment, step S1200, extracting a periodically repeated feature of the operation features according to the access data and the transaction data in the store data, includes the following steps:

step S2200, obtaining the access number in the access data in the store data of the preset time granularity, and the amount of orders in the transaction data;

the predetermined time granularity may be set to one day or two days, and may be set as needed by those skilled in the art.

In one embodiment, the preset time granularity is set to be one day, the number of the access sessions in the access data in the store data in one day is counted in advance to obtain the corresponding access number, and the number of the orders and the amount of the orders obtained by adding all the orders in one day are counted according to the order creation time of each order in the transaction data in the store data.

Step S2210, determining whether each first time window in the preset duration is similar to the access number, the order number and the order amount corresponding to the last first time window respectively, and obtaining the number of the first time windows corresponding to the similar access number, the similar order number and the similar order amount;

the preset time period can be set to be any one of the near half year and the last half year of the on-line shop operation, and can be set by a person skilled in the art as required, and the first time window is set to be one week.

In an embodiment, the preset time period is set to be about half a year, a similarity algorithm is adopted to calculate the similarity of the visit number, the order number and the order amount of each day in a week corresponding to the last week in about half a year, the number of first time windows and the number of first time windows, wherein the similarity of the visit number and the order amount exceeds a preset threshold, the number of the first time windows and the number of the first time windows, the number of the first time windows is the number of the order windows, the number of the order windows is the number of the order windows, and the number of the order windows is the order amount is the number of the order. The similarity algorithm can be any one of cosine similarity algorithm, vector dot product algorithm, manhattan distance, euclidean distance algorithm, pierson correlation coefficient and the like, and can be realized by one skilled in the art according to the need.

Step S2220, determining whether each second time window in the preset duration is similar to the access number, the order number and the order amount corresponding to the last second time window, and obtaining the number of second time windows corresponding to the similar access number, the similar order number and the similar order amount;

the preset time period can be set to be any one of the near half year and the last half year of the on-line shop operation, and can be set by a person skilled in the art as required, and the first time window is set to be one month.

In an embodiment, the preset time period is set to be about half a year, a similarity algorithm is adopted to calculate the respective similarity among the number of accesses, the number of orders and the amount of orders of each month in each month corresponding to the last month in the about half a year, the number of second time windows and the number of second time windows, wherein the similarity represents that the number of accesses is similar to the number of the second time windows, the number of the second time windows represents that the similarity represents that the number of orders is similar to the number of the second time windows exceeds the preset threshold, and the preset threshold can be set by a person skilled in the art according to requirements. The similarity algorithm can be any one of cosine similarity algorithm, vector dot product algorithm, manhattan distance, euclidean distance algorithm, pierson correlation coefficient and the like, and can be realized by one skilled in the art according to the need.

Step S2230, the number of the first time windows and the number of the second time windows corresponding to the similar access number, the similar order number and the similar order amount are configured as periodic repeating features.

In one embodiment, the number of first time windows corresponding to the similar number of orders and the similar amount of orders is multiplied to obtain a corresponding first period combination feature, the number of second time windows corresponding to the similar number of orders and the similar amount of orders is multiplied to obtain a corresponding second period combination feature, and the number of first time windows and the number of second time windows corresponding to the similar number of orders and the similar amount of orders are configured as periodic repetition features

In this embodiment, by extracting the periodic repetitive feature from the access data and the transaction data in the store data divided in the period, the periodic law of the online store in sales can be accurately and deeply captured.

Referring to fig. 4, in a further embodiment, step S1200, extracting the quick start feature from the operation features according to the access data and the transaction data in the store data, includes the following steps:

Step S3200, obtaining the access number in the access data in the store data with preset time granularity, and the amount of orders in the transaction data;

Step S3210, determining the maximum value and the minimum value of the access number, the order quantity and the order amount corresponding to each first time window in the preset duration;

In one embodiment, the preset time period is set to be about half a year, and the maximum value and the minimum value of the visit number, the order number and the order amount each day in a week corresponding to each week in about half a year are determined.

Step S3220, according to the maximum value and the minimum value of the access number, the order number and the order amount respectively, the number of the first time windows corresponding to the access number quick start, the order number quick start and the order amount quick start is correspondingly determined, and the quick start characteristic is constructed.

According to the maximum value and the minimum value of the access number, the order number and the order amount respectively, the increment fluctuation value corresponding to the access number, the order number and the order amount is calculated, and an exemplary formula is as follows:

here, score1 is a growth fluctuation value, max is a maximum value, and min is a minimum value.

When the increment fluctuation value corresponding to the access number, the order number and the order amount is larger than a preset threshold value, the corresponding first time window is indicated to have quick increment fluctuation, namely the quick start quantity is included, and the preset threshold value can be set by a person skilled in the art as required, and is recommended to be set to be 0.6.

In this embodiment, by extracting the quick start feature from the access data and the transaction data in the store data divided in the period, the quick growth fluctuation of the online store in sales can be accurately and deeply captured.

Referring to fig. 5, in a further embodiment, step S1200, extracting peer-to-peer competitive features from the operation features according to configuration information, access data and transaction data in store data, includes the following steps:

Step S4200, determining the same-line store group of the online store according to the configuration information in the store data;

and determining a plurality of online stores which are the same as the online stores as the online store group of the same line according to the category of the main commodity in the configuration information in the store data.

Step S4210, determining an order amount mean value, a repurchase rate and a conversion rate according to the access data and the transaction data in the store data;

according to the access session in the access data in the store data with the preset time granularity, comparing the times of the user purchasing the online store commodity with the sum of the times of the user searching and browsing the online store commodity, calculating the conversion rate, and comparing the number of users purchasing the same commodity in the online store at least twice with the number of users purchasing all the online store commodity, calculating the repurchase rate; according to the order creation time of each order in the transaction data in the store data with the preset time granularity, calculating the order amount of all the average orders, and calculating the average order amount. The preset time granularity can be set to be any one of the near half year and the last half year of the on-line shop operation, and can be set as required by a person skilled in the art.

Step S4220, determining the grades corresponding to the average order amount mean value, the average repurchase rate and the average conversion rate of the shop group on the same line, and constructing the grade corresponding to the order amount mean value, the repurchase rate and the conversion rate as competitive characteristics of the same line.

And determining an average order amount value, an average repurchase rate and an average conversion rate corresponding to each online store in the online store group of the same line according to the step S4210, and further correspondingly calculating the average order amount value, the average repurchase rate and the average conversion rate compared with the online store number.

Determining the rating of the order amount mean according to the order amount mean of the online stores and the average order amount mean of the online store groups of the same line, wherein the determining comprises the following steps: when the average value of the order amount of the online store is larger than the average value of the order amount of the online store, and the difference between the average value of the order amount and the average value of the order amount is calculated and divided by the average order amount, determining that the grade of the average value of the order amount is high in peer competitiveness; when the average value of the order amount of the online store is smaller than the average value of the order amounts of the online stores of the same row, and the difference between the average value of the order amounts and the average value of the order amounts is calculated and divided by the average order amount, and the obtained calculation result is larger than a preset threshold value, determining that the rating of the average value of the order amounts is low in competition of the same row; and in other cases, determining the average value of the order amount as the rank of the same-row competitive power. The preset threshold may be set by those skilled in the art as desired, preferably to 0.2.

And determining the grade corresponding to the repurchase rate and the conversion rate according to the repurchase rate and the conversion rate of the online shops and the average repurchase rate and the average conversion rate of the online shops of the same line according to the same principle.

And constructing the grades corresponding to the average value of the order amount, the repurchase rate and the conversion rate of the online shops as competitive characteristics of the same line.

In this embodiment, the on-line shop group of the on-line shops is determined according to the configuration information in the shop data, and the order amount average value, the repurchase rate and the conversion rate of the on-line shops are determined according to the access data and the transaction data in the shop data, so that the order amount average value, the repurchase rate and the conversion rate are respectively compared with the corresponding on-line average levels of the on-line shops, and the corresponding rating structure is determined to be the on-line competitive feature, so that the competitive power of the on-line shops in the on-line aspect can be accurately evaluated.

Referring to fig. 6, in a further embodiment, step S1200, the basic features further include a combination feature and a multidimensional feature, and the extraction includes:

step S1201, determining offline sales channel characteristics and advertisement marketing channel characteristics according to configuration information of store data, and obtaining combined characteristics, wherein the offline sales channel characteristics represent whether online stores have offline sales channels or not, and the advertisement marketing channel characteristics represent whether online stores have advertisement marketing channels or not;

And correspondingly determining the offline sales channel characteristics and the advertisement marketing channel characteristics according to whether the offline sales channel and the advertisement marketing channel exist in the configuration information of the store data, wherein the offline sales channel characteristics can be expressed as 1 if the offline sales channel is for example, or the offline sales channel characteristics can be expressed as 0, and the advertisement marketing channel characteristics can be expressed in the same way. Further, the off-line sales channel feature and the advertising marketing channel feature are added to obtain a combined feature.

And step 1202, determining the distribution of customer complaints reasons and the corresponding satisfaction degree distribution according to the after-sales service data of the store data, and performing dimension reduction processing to obtain multidimensional features.

Clustering by adopting a clustering algorithm according to the complaint reasons in the after-sales service data in the store data with preset time granularity, classifying the reason clusters to which each complaint reason belongs, determining the number of the complaint reasons corresponding to each reason cluster, further calculating the sum of the number of the complaint reasons corresponding to each reason cluster and the number of the complaint reasons corresponding to all reason clusters, and correspondingly obtaining the total ratio corresponding to each reason cluster to obtain the distribution of the complaint reasons. The clustering algorithm can be k-means, KNN and the like, and can be realized by one skilled in the art according to the need. The specific meaning of the reason cluster is obtained by analyzing the complaint reasons in the original cluster obtained after clustering, such as non-compliance of commodity description, poor commodity quality, poor customer service attitude and the like, and the specific number of the reason cluster can be determined by algorithms such as an elbow method, a contour coefficient method, a Gap statistic method, a DB index method, a hierarchical clustering method and the like, so that the reason cluster can be realized by one of the technicians according to the requirements.

And further calculating the number of users corresponding to each satisfaction according to the satisfaction in the after-sales service data in the store data with the preset time granularity, and comparing the number of users corresponding to each satisfaction with the sum of the numbers of users corresponding to the satisfaction respectively, and correspondingly obtaining the occupation ratio corresponding to each satisfaction to obtain the satisfaction distribution.

The preset time granularity can be set to be any one of the near half year and the last half year of the on-line shop operation, and can be set as required by a person skilled in the art.

Further, the dimensionality reduction algorithm is adopted to carry out dimensionality reduction processing on the customer complaint reason distribution and the satisfaction distribution, so as to obtain multidimensional features, wherein the dimensionality reduction algorithm can be any one of factor analysis, principal component analysis and the like, and can be realized by a person skilled in the art as required.

In this embodiment, the richness of the basic features can be ensured by constructing the corresponding combination features according to the configuration information in the store data and constructing the multidimensional features according to the after-sales service data.

Referring to fig. 7, a store feature application apparatus provided in accordance with one of the purposes of the present application is a functional implementation of a store feature application method of the present application, and on the other hand, the apparatus provided in accordance with one of the purposes of the present application includes a data acquisition module 1100, a feature extraction module 1200, a store classification module 1300, and a policy customization module 1400, where the data acquisition module 1100 is configured to acquire store data of an online store, and the store data includes configuration information, access data, transaction data, and after-sales service data; the feature extraction module 1200 is configured to extract basic features according to the store data, extract operating features according to access data, transaction data and configuration information in the store data, and construct the basic features and the operating features as store portrait features of the online store, where the basic features are used for representing online store operating conditions, and the operating features are used for representing online store operating effects; a store classification module 1300 for inputting the store image feature into a store classification model to obtain a store type to which the online store belongs; the policy customizing module 1400 is configured to adjust a basic operation policy corresponding to the store type according to the operation feature, so as to obtain a customized operation policy of the online store.

In a further embodiment, the feature extraction module 1200 includes: the statistical characteristic sub-module is used for carrying out statistics according to the store data and corresponding to a plurality of preset time granularities to obtain statistical characteristics corresponding to each time granularity, wherein the statistical characteristics represent the operating conditions of any one aspect or any multiple aspects of the store in sales, advertisement marketing and after-sales service under the corresponding time span; the user group distribution characteristic sub-module is used for clustering by adopting a clustering algorithm according to the access data in the store data, and determining the number of users corresponding to a plurality of preset user groups and the total ratio of the users as user group distribution characteristics; and the layering characteristic sub-module is used for layering by adopting a preset step-type threshold interval according to the access data and the transaction data in the store data to determine layering characteristics, wherein the layering characteristics represent the levels corresponding to a plurality of operation key indexes of the store on the line, and the levels represent the quality of the operation key indexes.

In a further embodiment, the operation features in the feature extraction module 1200 include a periodic repeating feature, a fast-rise feature, and a peer competitiveness feature, where the periodic repeating feature characterizes a periodic law of sales of online stores, the fast-rise feature characterizes fast-growth fluctuations of sales of online stores, and the peer competitiveness feature characterizes competitiveness of online stores in peer.

In a further embodiment, the feature extraction module 1200 includes: the granularity data acquisition sub-module is used for acquiring the access number in the access data in the store data of the preset time granularity, and the order quantity and the order amount in the transaction data; zhou Guilv determining submodule, configured to determine whether each first time window in the preset duration is similar to the access number, the order number and the order amount corresponding to the previous first time window, and obtain the number of first time windows corresponding to the similar access number, the similar order number and the similar order amount; the month rule determining submodule is used for determining whether each second time window in the preset duration is similar to the access number, the order number and the order amount corresponding to the last second time window respectively, and obtaining the number of the second time windows corresponding to the similar access number, the similar order number and the similar order amount; and the periodic repeated feature construction submodule is used for constructing the number of the first time windows and the number of the second time windows corresponding to the similar access number, the similar order number and the similar order amount as periodic repeated features.

In a further embodiment, the feature extraction module 1200 includes: the granularity data acquisition sub-module is used for acquiring the access number in the access data in the store data of the preset time granularity, and the order quantity and the order amount in the transaction data; the maximum value determining submodule is used for determining the maximum value and the minimum value of the access number, the order quantity and the order amount corresponding to each first time window in the preset duration; and the quick start characteristic construction submodule is used for correspondingly determining the number of the first time windows corresponding to the quick start of the access number, the quick start of the order number and the quick start of the order amount according to the maximum value and the minimum value of the access number, the order number and the order amount respectively, and constructing the quick start characteristic.

In a further embodiment, the feature extraction module 1200 includes: the peer determination submodule is used for determining peer online store groups of the online stores according to the configuration information in the store data; the business effect determining submodule is used for determining the average value of the order amount, the repurchase rate and the conversion rate according to the access data and the transaction data in the store data; and the peer competitive feature construction submodule is used for determining the grades corresponding to the order amount average value, the repurchase rate and the conversion rate according to the average order amount average value, the average repurchase rate and the average conversion rate of the shop group on the peer line and constructing the peer competitive feature.

In a further embodiment, the feature extraction module 1200 includes: the combined feature sub-module is used for determining offline sales channel features and advertisement marketing channel features according to the configuration information of the store data to perform feature combination, so as to obtain combined features, wherein the offline sales channel features represent whether online stores have offline sales channels or not, and the advertisement marketing channel features represent whether online stores have advertisement marketing channels or not; and the multidimensional feature sub-module is used for determining the distribution of customer complaints reasons and the distribution of satisfaction according to the after-sales service data of the store data and correspondingly carrying out dimension reduction processing to obtain multidimensional features.

In order to solve the technical problems, the embodiment of the application also provides computer equipment. As shown in fig. 8, the internal structure of the computer device is schematically shown. The computer device includes a processor, a computer readable storage medium, a memory, and a network interface connected by a system bus. The computer readable storage medium of the computer device stores an operating system, a database and computer readable instructions, the database can store a control information sequence, and the computer readable instructions, when executed by a processor, can enable the processor to realize a store feature application method. The processor of the computer device is used to provide computing and control capabilities, supporting the operation of the entire computer device. The memory of the computer device may have stored therein computer readable instructions that, when executed by the processor, cause the processor to perform the store feature application method of the present application. The network interface of the computer device is for communicating with a terminal connection. It will be appreciated by those skilled in the art that the structure shown in fig. 8 is merely a block diagram of some of the structures associated with the present application and is not limiting of the computer device to which the present application may be applied, and that a particular computer device may include more or fewer components than shown, or may combine certain components, or have a different arrangement of components.

The processor in this embodiment is configured to execute specific functions of each module and its sub-module in fig. 7, and the memory stores program codes and various data required for executing the above modules or sub-modules. The network interface is used for data transmission between the user terminal or the server. The memory in the present embodiment stores program codes and data necessary for executing all modules/sub-modules in the store feature application device of the present application, and the server can call the program codes and data of the server to execute the functions of all sub-modules.

The present application also provides a storage medium storing computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform the steps of the store feature application method of any of the embodiments of the present application.

Those skilled in the art will appreciate that implementing all or part of the above-described methods of embodiments of the present application may be accomplished by way of a computer program stored on a computer readable storage medium, which when executed, may comprise the steps of embodiments of the methods described above. The storage medium may be a computer readable storage medium such as a magnetic disk, an optical disk, a Read-Only Memory (ROM), or a random access Memory (Random Access Memory, RAM).

In summary, the method constructs multidimensional store portrait features, accurately classifies corresponding store types, and determines practical operation strategies.

Those of skill in the art will appreciate that the various operations, methods, steps in the flow, actions, schemes, and alternatives discussed in the present application may be alternated, altered, combined, or eliminated. Further, other steps, means, or steps in a process having various operations, methods, or procedures discussed in this application may be alternated, altered, rearranged, split, combined, or eliminated. Further, steps, measures, schemes in the prior art with various operations, methods, flows disclosed in the present application may also be alternated, altered, rearranged, decomposed, combined, or deleted.

The foregoing is only a partial embodiment of the present application, and it should be noted that, for a person skilled in the art, several improvements and modifications can be made without departing from the principle of the present application, and these improvements and modifications should also be considered as the protection scope of the present application.

Claims

1. A store feature application method, comprising the steps of:

inputting the store image features into a store classification model to obtain the store type of the online store; the model structure of the store classification model is a text feature extraction layer connected with a classifier, and training is carried out in advance until convergence, so that the capability of classifying the store type of the online store based on the store image features of the online store is obtained;

according to the operation characteristics, adjusting a basic operation strategy corresponding to the store type to obtain a customized operation strategy of the online store;

the basic features comprise statistical features, user group distribution features and layering features, and the basic features are extracted according to the following process, wherein the basic features comprise:

layering is carried out according to access data and transaction data in the store data by adopting a preset step-type threshold interval, layering characteristics are determined, the layering characteristics represent levels corresponding to a plurality of operation key indexes of the store on a line, and the levels represent the quality of the operation key indexes;

the operation characteristics comprise a periodical repeated characteristic, a quick starting characteristic and a peer competitiveness characteristic, wherein the periodical repeated characteristic represents a periodical rule of online stores in sales, the quick starting characteristic represents quick growth fluctuation of online stores in sales, and the peer competitiveness characteristic represents competitiveness of online stores in peer.

2. The store feature application method according to claim 1, wherein the periodically repeated feature of the operation features is extracted from the access data and the transaction data in the store data, comprising the steps of:

3. The store feature application method according to claim 1, wherein the quick start feature in the operation feature is extracted from access data and transaction data in store data, comprising the steps of:

4. The store feature application method according to claim 1, wherein the step of extracting the peer-to-peer competitive feature of the operation features from the configuration information, the access data, and the transaction data in the store data comprises the steps of:

5. The store feature application method according to any one of claims 1 to 4, wherein the basic features further comprise combined features, multidimensional features, extracted according to the following process, including:

6. A store characterization application device, comprising:

the system comprises a data acquisition module, a data processing module and a data processing module, wherein the data acquisition module is used for acquiring store data of an online store, and the store data comprises configuration information, access data, transaction data and after-sales service data;

the feature extraction module is used for extracting basic features according to the store data, extracting operation features according to access data, transaction data and configuration information in the store data, constructing the basic features and the operation features into store image features of the online store, wherein the basic features are used for representing the operation conditions of the online store, and the operation features are used for representing the operation effects of the online store;

the store classification module is used for inputting the store image characteristics into a store classification model to obtain the store type of the online store; the model structure of the store classification model is a text feature extraction layer connected with a classifier, and training is carried out in advance until convergence, so that the capability of classifying the store type of the online store based on the store image features of the online store is obtained;

The strategy customizing module is used for adjusting a basic operation strategy corresponding to the store type according to the operation characteristics to obtain a customized operation strategy of the online store;

the feature extraction module comprises: the statistical characteristic sub-module is used for carrying out statistics according to the store data and corresponding to a plurality of preset time granularities to obtain statistical characteristics corresponding to each time granularity, wherein the statistical characteristics represent the operating conditions of any one aspect or any multiple aspects of the store in sales, advertisement marketing and after-sales service under the corresponding time span; the user group distribution characteristic sub-module is used for clustering by adopting a clustering algorithm according to the access data in the store data, and determining the number of users corresponding to a plurality of preset user groups and the total ratio of the users as user group distribution characteristics; the layering characteristic submodule is used for layering according to access data and transaction data in the store data by adopting a preset step-type threshold interval, so that layering characteristics are determined, the layering characteristics represent levels corresponding to a plurality of operation key indexes of the store on a line, and the levels represent good and good conditions of the operation key indexes;

the operation features in the feature extraction module comprise periodic repeated features, quick starting features and peer competitiveness features, wherein the periodic repeated features represent the periodic law of online stores in sales, the quick starting features represent the quick growth fluctuation of the online stores in sales, and the peer competitiveness features represent the competitiveness of the online stores in peer.

7. A computer device comprising a central processor and a memory, characterized in that the central processor is arranged to invoke a computer program stored in the memory for performing the steps of the method according to any of claims 1 to 5.

8. A computer-readable storage medium, characterized in that it stores in the form of computer-readable instructions a computer program implemented according to the method of any one of claims 1 to 5, which, when invoked by a computer, performs the steps comprised by the corresponding method.