CN113516534A - Service management method and system based on big data - Google Patents

Abstract

The embodiment of the invention provides a big data-based business management method and a big data-based business management system, wherein a plurality of similar commodities similar to a target commodity are obtained through business data of the received target commodity under a plurality of sales attributes, a relevant commodity list matched with the similar commodities respectively is obtained, the relevant commodity in the relevant commodity list is at least one other commodity relevant to the similar commodity during information display, the relevant commodity with the click rate meeting preset conditions of each relevant commodity in each relevant commodity list is used as an alternative relevant commodity, and finally a quasi-relevant commodity of the target commodity during information display is determined from the alternative relevant commodities. The related commodities of the similar commodities are directly called through matching the similar commodities, reasonable matching is carried out through combination judgment of the click rate and the similarity, the related commodities serve as the related commodities of the target commodity, matching is carried out in the database through the related data of the target commodity directly in the prior art, and efficiency and accuracy of searching the related commodities are improved.

Description

Service management method and system based on big data

Technical Field

The present application relates to the field of data processing, and in particular, to a service management method and system based on big data.

Background

With the rapid development of internet economy, the competitive environment of e-commerce is more and more intense, and when a user clicks and watches a certain commodity, each large e-commerce platform often pushes other commodities related to the commodity to the user at a proper position and time so as to attract the attention of the user and facilitate the bargaining of an order when the user shows the commodity to the client, so that how to accurately push the related commodities to the user according to the consumption habits and the consumption psychology of the user is a direction of key attention of each e-commerce platform at present.

Disclosure of Invention

The invention aims to provide a business management method and system based on big data, so that interested commodities can be accurately pushed to a user when the user purchases E-commerce commodities.

In order to achieve the above object, the embodiment of the present invention is implemented as follows:

in a first aspect, an embodiment of the present invention provides a method, including: receiving business data of a target commodity under a plurality of sales attributes, and obtaining a plurality of similar commodities similar to the target commodity; acquiring an associated commodity list matched with each of the similar commodities, wherein the associated commodity in the associated commodity list is at least one other commodity associated with the similar commodity during information display; taking the associated commodities of which the click rate of each associated commodity in each associated commodity list meets the preset condition as alternative associated commodities; and determining a to-be-associated commodity of the target commodity when the information is displayed from the alternative associated commodities.

Further, acquiring an associated commodity list matched with each of the plurality of similar commodities, including:

acquiring a plurality of prior associated commodities of each similar commodity during display;

respectively obtaining group click rates of a plurality of prior associated commodities in different user groups;

obtaining click rates corresponding to a plurality of prior associated commodities through the obtained click rates of the plurality of groups and the weights corresponding to the user groups respectively;

and selecting the associated commodities with the click rates higher than the preset click rate according to the obtained click rates, wherein the selected associated commodities with the click rates higher than the preset click rate are associated commodities corresponding to similar commodities.

Further, taking the associated commodity of which the click rate of each associated commodity in each associated commodity list meets the preset condition as a candidate associated commodity, including:

respectively obtaining the weighted click rate corresponding to each associated commodity in the associated commodity list according to the similarity between the similar commodity and the target commodity and the click rate of each associated commodity in the associated commodity list of the similar commodity;

and selecting at least one associated commodity with the weighted click rate larger than the preset click rate through the weighted click rate of each associated commodity in the associated commodity list, wherein the selected at least one associated commodity with the weighted click rate larger than the preset click rate is an alternative associated commodity.

Further, before selecting at least one associated commodity with a weighted click rate greater than a preset click rate through the weighted click rate of each associated commodity in the associated commodity list, the method further comprises:

determining the same associated commodities in the associated commodity list;

and obtaining the weighted click rate selected from the same associated commodities through the weighted click rate of the same associated commodity corresponding to each similar commodity, wherein the same associated commodity is the same associated commodity in the associated commodity list corresponding to different similar commodities.

Further, taking the associated commodity of which the click rate of each associated commodity in each associated commodity list meets the preset condition as a candidate associated commodity, including:

obtaining a plurality of weighted group click rates of the associated commodities according to the group click rates of the associated commodities corresponding to a plurality of user groups and the similarity corresponding to the associated commodities; the similarity corresponding to the associated commodity is the similarity between the similar commodity corresponding to the associated commodity and the target commodity;

obtaining the weighted click rate of the associated commodity according to the obtained click rates of the plurality of weighted groups;

and selecting at least one associated commodity with the weighted click rate larger than the preset click rate through the weighted click rate of each associated commodity, wherein the selected at least one associated commodity with the weighted click rate larger than the preset click rate is a candidate associated commodity.

Further, obtaining a plurality of similar commodities similar to the target commodity based on the business data of the target commodity under the plurality of sales attributes comprises:

inputting the business data of the target commodity corresponding to a plurality of sales attributes into a commodity information vector model to obtain a commodity information vector of the target commodity, obtaining at least one similar commodity information vector close to the commodity information vector, and determining the similar commodity corresponding to the similar commodity information vector. 7. The method of claim 6,

the commodity information vector model is obtained by training according to the following steps:

acquiring a training sample, wherein the training sample is a plurality of training vectors obtained by performing feature extraction on business data of a commodity under a plurality of sales attributes, and each training vector corresponds to one sales attribute;

performing feature fusion processing on the training vectors to obtain commodity information vectors corresponding to commodities;

obtaining type prediction corresponding to the commodity through the commodity information vector;

determining a loss value between the type prediction and the actual type of the commodity through a first loss function;

and adjusting parameters of the commodity information vector model through the loss value until a preset condition is met.

Further, the training of the commodity information vector model further comprises:

acquiring similar training samples, wherein the similar training samples comprise service data of two commodities;

acquiring commodity information vectors of the two service data and performing feature fusion to obtain a fusion commodity information vector of a similar training sample;

obtaining the proximity degree predicted values of the two commodities by fusing commodity information vectors;

determining a loss value between the predicted proximity degree value and the pre-labeled real proximity degree through a second loss function;

and adjusting parameters of the commodity information vector model through the loss value until a preset condition is met.

Further, determining a quasi-associated commodity of the target commodity when the information is displayed from the alternative associated commodities includes:

inputting business data of the alternative associated commodities into a vector generation model to obtain commodity vectors of the alternative associated commodities;

obtaining the association degree of the alternative associated commodities and the account numbers through the commodity vector and the account number vector of the user receiving the alternative associated commodities;

inputting the click rate data of the alternative associated commodities into a conjecture model to obtain click rate conjecture results of the alternative associated commodities;

obtaining the pushing degree of the alternative associated commodities according to the association degree and click rate guess result;

and selecting the to-be-associated commodity from at least one candidate associated commodity according to the obtained pushing degree of each candidate associated commodity.

In a second aspect, an embodiment of the present invention further provides a big data based business management system, where the system includes a client and a server that are in communication with each other, the client is configured to send business data of a target product under multiple sales attributes to the server, the server includes a processor and a memory that are in communication with each other, and the processor is configured to retrieve a computer program from the memory and implement the method provided in the first aspect by running the computer program.

According to the big data-based business management method and system provided by the embodiment of the invention, a plurality of similar commodities similar to the target commodity are obtained through the received business data of the target commodity under a plurality of sales attributes, a relevant commodity list matched with the similar commodities is obtained, the relevant commodity in the relevant commodity list is at least one other commodity associated with the similar commodity during information display, the relevant commodity with the click rate meeting the preset condition of each relevant commodity in each relevant commodity list is used as an alternative relevant commodity, and finally a quasi-relevant commodity of the target commodity during information display is determined from the alternative relevant commodities. The related commodities of the similar commodities are directly called through matching the similar commodities, reasonable matching is carried out through combination judgment of the click rate and the similarity, the related commodities serve as the related commodities of the target commodity, matching is carried out in the database through the related data of the target commodity directly in the prior art, and efficiency and accuracy of searching the related commodities are improved.

In the description that follows, additional features will be set forth, in part, in the description. These features will be in part apparent to those skilled in the art upon examination of the following and the accompanying drawings, or may be learned by production or use. The features of the present application may be realized and attained by practice or use of various aspects of the methodologies, instrumentalities and combinations particularly pointed out in the detailed examples that follow.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained from the drawings without inventive effort.

The methods, systems, and/or processes of the figures are further described in accordance with the exemplary embodiments. These exemplary embodiments will be described in detail with reference to the drawings. These exemplary embodiments are non-limiting exemplary embodiments in which reference numerals represent similar mechanisms throughout the various views of the drawings.

FIG. 1 is a block diagram of a big data based business management system, shown in accordance with some embodiments of the present application.

FIG. 2 is a schematic diagram illustrating the hardware and software components in a server according to some embodiments of the present application.

FIG. 3 is a flow diagram illustrating a big data based business management method according to some embodiments of the present application.

Fig. 4 is a schematic architecture diagram of a big data based traffic management device according to an embodiment of the present application.

Detailed Description

In order to better understand the technical solutions, the technical solutions of the present application are described in detail below with reference to the drawings and specific embodiments, and it should be understood that the specific features in the embodiments and examples of the present application are detailed descriptions of the technical solutions of the present application, and are not limitations of the technical solutions of the present application, and the technical features in the embodiments and examples of the present application may be combined with each other without conflict.

In the following detailed description, numerous specific details are set forth by way of examples in order to provide a thorough understanding of the relevant guidance. It will be apparent, however, to one skilled in the art that the present application may be practiced without these specific details. In other instances, well-known methods, procedures, systems, compositions, and/or circuits have been described at a relatively high-level, without detail, in order to avoid unnecessarily obscuring aspects of the present application.

These and other features, functions, methods of execution, and combination of functions and elements of related elements in the structure and economies of manufacture disclosed in the present application may become more apparent upon consideration of the following description with reference to the accompanying drawings, all of which form a part of this application. It is to be expressly understood, however, that the drawings are for the purpose of illustration and description only and are not intended as a definition of the limits of the application. It should be understood that the drawings are not to scale. It is to be expressly understood, however, that the drawings are for the purpose of illustration and description only and are not intended as a definition of the limits of the application. It should be understood that the drawings are not to scale.

Flowcharts are used herein to illustrate the implementations performed by systems according to embodiments of the present application. It should be expressly understood that the processes performed by the flowcharts may be performed out of order. Rather, these implementations may be performed in the reverse order or simultaneously. In addition, at least one other implementation may be added to the flowchart. One or more implementations may be deleted from the flowchart.

Fig. 1 is a block diagram of a system architecture of a big data based business management system 300 according to some embodiments of the present application, where the big data based business management system 300 may include a server 100 and a plurality of clients 200 in communication therewith.

The client 200 may be a device used when the target user performs online shopping, and may be, for example, a personal computer, a notebook computer, a tablet computer, a smart phone, or the like having a network interaction function.

In some embodiments, please refer to fig. 2, which is a schematic diagram of an architecture of a server 100, wherein the server 100 includes a big data based service management apparatus 110, a memory 120, a processor 130, and a communication unit 140. The elements of the memory 120, the processor 130, and the communication unit 140 are electrically connected to each other, directly or indirectly, to enable the transmission or interaction of data. For example, the components may be electrically connected to each other via one or more communication buses or signal lines. The big data based traffic management apparatus 110 includes at least one software function module that may be stored in the memory 120 in the form of software or firmware (firmware) or solidified in an Operating System (OS) of the server 100. The processor 130 is used to execute executable modules stored in the memory 120, such as software functional modules and computer programs included in the teleeducation-based business information processing apparatus 110.

The Memory 120 may be, but is not limited to, a Random Access Memory (RAM), a Read Only Memory (ROM), a Programmable Read-Only Memory (PROM), an Erasable Read-Only Memory (EPROM), an electrically Erasable Read-Only Memory (EEPROM), and the like. The memory 120 is used for storing a program, and the processor 130 executes the program after receiving the execution instruction. The communication unit 140 is used for establishing a communication connection between the server 100 and the service interaction device through a network, and for transceiving data through the network.

The processor may be an integrated circuit chip having signal processing capabilities. The Processor may be a general-purpose Processor, including a Central Processing Unit (CPU), a Network Processor (NP), and the like; but may also be a Digital Signal Processor (DSP)), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware components. The various methods, steps and logic blocks disclosed in the embodiments of the present invention may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.

It will be appreciated that the configuration shown in fig. 2 is merely illustrative and that server 100 may include more or fewer components than shown in fig. 2 or have a different configuration than shown in fig. 2. The components shown in fig. 2 may be implemented in hardware, software, or a combination thereof.

Fig. 3 is a flowchart of a big data based service management method according to some embodiments of the present application, where the method is applied to the server 100 in fig. 1, and may specifically include the following steps S1-S4. On the basis of the following steps S1-S4, some alternative embodiments will be explained, which should be understood as examples and should not be understood as technical features essential for implementing the present solution.

Step S1 is to receive the business data of the target product under the plurality of sales attributes and obtain a plurality of similar products similar to the target product.

In this embodiment, the target product is a product being browsed by the user, and for the target product, other related products need to be matched with the target product, for example, the target product is a short-sleeved T-shirt, in the process of browsing the short-sleeved T-shirt by the user, other clothes related to the short-sleeved T-shirt browsed by the user need to be pushed to the user at a proper position and at a proper time, and the user may generate a purchase interest after seeing the related clothes, so that the e-commerce platform generates a greater benefit. As for how to find the proper associated goods, there are many factors to be measured, including the type of goods, the defined tag, the price, the style and the color of goods, even the goodness, the click rate and so on, which are collectively referred to as the sales attribute of goods, and the data displayed by the goods attribute is the business data, for example, the type of goods is clothes, the tag is housekeeping + quadratic element, the price is 100 yuan or less, the style is fast selling wind, the color is dark color system, and the above various detailed descriptions of goods are business data. In this embodiment, a plurality of similar commodities similar to the target commodity are obtained according to the similarity of the commodity attributes, and the similar evaluation mode can be realized by evaluating the proximity. For example, the text similarity can be judged according to the number of the same characters of the texts in the service data of the two products, so that the proximity degree is obtained. In addition, an Artificial Intelligence (AI) model can be used to perform the correlation operation.

In this embodiment, the similarity is determined by AI, specifically, the business data of the target commodity corresponding to a plurality of sales attributes is input into a commodity information vector model trained in advance to obtain a commodity information vector of the target commodity, then at least one similar commodity information vector close to the commodity information vector is obtained in a commodity database, and a similar commodity corresponding to the similar commodity information vector is determined. Because the characteristics of the commodities are represented by the commodity vector information, when judging whether the two commodities are similar, the difference of the vector values can be directly compared, for example, a threshold value is set, and when the difference value is smaller than the threshold value, the commodities corresponding to the commodity vector information are proved to be similar.

The training process of the AI model may refer to the following sub-steps:

firstly, a training sample is obtained, the training sample obtains a plurality of training vectors through feature extraction of business data of commodities under a plurality of sales attributes, and each training vector corresponds to one sales attribute. And performing feature fusion processing on the plurality of training vectors to obtain commodity information vectors corresponding to the commodities. Through feature fusion, information of training vectors of all sales attributes is integrated, and the fusion process can be realized through vector splicing, for example, after the training vectors of characters are spliced with the training vectors of numbers, in addition, the feature fusion can be performed by adopting conventional pooling processing, or the feature fusion can be performed in a full-connection processing mode. The embodiment of the invention does not limit the process of feature fusion.

After fusion, the type prediction corresponding to the commodity is obtained through the commodity information vector. The prediction process can be performed by adopting a classifier, the training process of the commodity information vector model provided by the embodiment is supervised training, and a loss value between the type prediction and the real type of the commodity is determined through a first loss function. And adjusting parameters of the commodity information vector model through the loss value until a preset condition is met. The first loss function may be a conventional loss function, such as a cross-entropy loss function, a 0-1 loss function. The preset condition may be any possible condition, such as reaching a preset number of training times, model convergence.

In addition, the training process may further include:

acquiring similar training samples, wherein the similar training samples comprise service data of two commodities; and acquiring commodity information vectors of the two service data and performing feature fusion to obtain a fusion commodity information vector of the similar training sample. And obtaining the predicted value of the proximity degree of the two commodities by fusing the commodity information vectors. And determining a loss value between the proximity prediction value and the pre-marked real proximity through a second loss function. And adjusting parameters of the commodity information vector model through the loss value until a preset condition is met. The above training process for performing the similarity comparison between the two commodities, the related feature fusion, the result prediction and other implementation processes can be implemented by adopting a conventional deep learning technology, and details are not repeated here.

In step S2, an associated product list in which a plurality of similar products match each other is acquired.

The related commodity in the related commodity list is at least one other commodity related to the similar commodity during information display, the related commodity list may include a plurality of related commodities, and the related commodity is a commodity appearing in the pushed content during previous commodity information display of the similar commodity.

For obtaining a list of associated commodities matched with each of a plurality of similar commodities, the present embodiment provides the following implementation manner, including the following steps:

step S21, a plurality of prior associated commodities for each similar commodity when displayed are obtained.

The goods appearing in the push content when the previously associated goods are displayed, namely similar goods.

Step S22 is to obtain group click rates of a plurality of prior related commodities in different user groups, respectively.

After the pushed content is presented to the user, the user may generate a click behavior, and the general click behavior can reflect that the user catches eyeballs at the first time, and the commodity can generate attraction to the user at the first time, so that the click rate is a key index for measuring whether the commodity is pushed successfully or not. For example, through statistics, after a male user group clicks a pushed commodity, the purchase conversion rate is greater than that of a female user group, so whether the prior associated commodity is suitable for pushing is evaluated, and evaluation needs to be performed according to the click rates of different user groups.

And step S23, obtaining final click rates corresponding to a plurality of prior related commodities according to the obtained click rates of the plurality of groups and the weights corresponding to the user groups respectively.

Because the click conversion rates of different user groups are different, the user group with high conversion rate is given high weight, so that the user group with low click conversion rate is given low weight, and the final click rate of all the user groups for the prior associated commodity can be obtained through weight calculation, such as weighted average.

And step S24, selecting the associated commodities with the final click rates larger than the preset click rate according to the obtained final click rates, wherein the selected associated commodities with the click rates larger than the preset click rate are associated commodities corresponding to similar commodities.

In step S3, the related item whose click rate of each related item in each related item list satisfies a preset condition is taken as a candidate related item.

After the associated commodities are selected, the associated commodities can be subjected to related sorting according to the click rate, and the commodities which are sorted in the front are preferentially selected as alternative associated commodities. As an embodiment, the step S3 may specifically include the following steps:

step S311, obtaining the weighted click rate corresponding to each associated product in the associated product list respectively according to the similarity between the similar product and the target product and the click rate of each associated product in the associated product list of the similar product.

Step S321, selecting at least one associated commodity with the weighted click rate larger than the preset click rate through the weighted click rate of each associated commodity in the associated commodity list, wherein the selected at least one associated commodity with the weighted click rate larger than the preset click rate is a candidate associated commodity.

In actual operation, the same associated commodity may exist, that is, the associated commodity list corresponding to different similar commodities has the same associated commodity, the same associated commodity appears twice or even more times, and the weighted click rate of the corresponding associated commodity appears correspondingly many times, so that the accuracy is reduced. Therefore, in this embodiment, before step S311, the same associated products in the associated product list may be determined, and the weighted click rate selected from the same associated products is obtained by using the weighted click rate of the same associated products corresponding to each similar product, where the selection may be a random selection, or a selection of the highest click rate, or an average of the click rates, and the average click rate is obtained as the click rate. The same associated commodity is the same associated commodity in the associated commodity list corresponding to different similar commodities.

As another embodiment, the method may further include the following steps of taking the associated item, of which the click rate of each associated item in each associated item list satisfies the preset condition, as an alternative associated item:

step S312, a plurality of weighted group click rates of the associated commodities are obtained through the group click rates of the associated commodities corresponding to the plurality of user groups and the similarity corresponding to the associated commodities.

The similarity corresponding to the associated commodity is the similarity between the similar commodity corresponding to the associated commodity and the target commodity.

And step S322, obtaining the weighted click rate of the associated commodity according to the obtained plurality of weighted group click rates.

And 332, selecting at least one associated commodity with the weighted click rate larger than the preset click rate through the weighted click rate of each associated commodity, wherein the selected at least one associated commodity with the weighted click rate larger than the preset click rate is a candidate associated commodity.

Finally, in step S4, a pseudo-related product of the target product at the time of information display is determined from the candidate related products.

Because the associated commodities are required to be pushed to the user finally, the user belongs to a user group, when the associated commodities are pushed to the user, the user can analyze the interest degree of the user on the associated commodities by combining the account information of the user, and the quasi-associated commodities of the target commodities during information display are searched from the alternative associated commodities. The proposed associated merchandise is the associated merchandise ready to be pushed to the user.

As an embodiment, the following steps may be included:

step S41, the business data of the candidate related commodity is input into the vector generation model, and a commodity vector of the candidate related commodity is obtained.

And step S42, obtaining the association degree of the alternative associated commodities and the account number through the commodity vector and the account number vector of the user receiving the alternative associated commodities. The account vector of the user can reflect which user group the user belongs to, and which aspect of the commodity is more emphasized, the commodity vector can reflect the characteristics of the associated commodity, the commodity characteristics reflected by the commodity vector and the commodity characteristics emphasized by the user can be judged by the proximity degree of the vector value, for example, the smaller the difference value is, the higher the association degree is, and the more the associated commodity meets the requirements of the user.

Step S43 is to input the click rate data of the candidate related product into the inference model to obtain the click rate inference result of the candidate related product.

In step S44, the degree of delivery of the candidate related product is obtained from the association degree and click rate estimation result.

Step S45, selecting a pseudo-related commodity from the at least one candidate related commodity according to the obtained pushing degree of each candidate related commodity.

The higher the pushing degree is, the higher the possibility that pushing is possible is indicated, and the more suitable the associated product is to be pushed. It is easy to understand. The vector generation model and the inference model may be obtained by training any possible neural network model, which is not limited in this embodiment.

Referring to fig. 4, which is a schematic structural diagram of a big data based service management apparatus 110 according to an embodiment of the present invention, the big data based service management apparatus 110 may be used to execute a big data based service management method, where the big data based service management apparatus 110 includes:

the query module 111 is configured to receive service data of the target product under multiple sales attributes, and obtain multiple similar products similar to the target product.

The indexing module 112 is configured to obtain an associated commodity list in which the plurality of similar commodities are respectively matched, where an associated commodity in the associated commodity list is at least one other commodity associated with the similar commodity when the information is displayed.

And the candidate module 113 is configured to use, as candidate related commodities, related commodities for which the click rate of each related commodity in each related commodity list satisfies a preset condition.

And the determining module 114 is used for determining the to-be-associated commodities of the target commodities when the information is displayed from the alternative associated commodities.

The query module 111 may be configured to perform the step S1, the index module 112 may be configured to perform the step S2, the alternative module 113 may be configured to perform the step S3, and the determination module 114 may be configured to perform the step S4.

Since the service management method based on big data provided in the embodiment of the present invention has been described in detail in the above embodiment, and the principle of the service management device 110 based on big data is the same as that of the method, the execution principle of each module of the service management device 110 based on big data is not described herein again.

In summary, in the service management method and system based on big data provided in the embodiments of the present invention, a plurality of similar commodities similar to a target commodity are obtained through service data of the received target commodity under a plurality of sales attributes, an associated commodity list matched with each of the similar commodities is obtained, the associated commodity in the associated commodity list is at least one other commodity associated with the similar commodity when information is displayed, then the associated commodity with a click rate of each associated commodity in each associated commodity list meeting a preset condition is used as an alternative associated commodity, and finally a quasi-associated commodity of the target commodity when information is displayed is determined from the alternative associated commodities. The related commodities of the similar commodities are directly called through matching the similar commodities, reasonable matching is carried out through combination judgment of the click rate and the similarity, the related commodities serve as the related commodities of the target commodity, matching is carried out in the database through the related data of the target commodity directly in the prior art, and efficiency and accuracy of searching the related commodities are improved.

In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method can be implemented in other ways. The apparatus embodiments described above are merely illustrative, and for example, the flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions. In addition, the functional modules in the embodiments of the present invention may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes. It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.

The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention. It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures.

Claims (10)

1. A big data-based service management method is applied to a server, and is characterized in that the method comprises the following steps:

receiving business data of a target commodity under a plurality of sales attributes, and obtaining a plurality of similar commodities similar to the target commodity;

acquiring an associated commodity list matched with the similar commodities respectively, wherein the associated commodity in the associated commodity list is at least one other commodity associated with the similar commodities during information display;

taking the associated commodities of which the click rate of each associated commodity in each associated commodity list meets a preset condition as alternative associated commodities;

and determining a quasi-associated commodity of the target commodity when the information is displayed from the alternative associated commodity.

2. The method of claim 1, wherein the obtaining a list of associated items that each match the plurality of similar items comprises:

obtaining a plurality of prior associated commodities of each similar commodity when being displayed;

respectively obtaining group click rates of the plurality of prior associated commodities in different user groups;

obtaining final click rates corresponding to the plurality of prior associated commodities according to the obtained click rates of the plurality of groups and weights respectively corresponding to the user groups;

and selecting the associated commodities with the final click rates larger than the preset click rate according to the obtained final click rates, wherein the selected associated commodities with the click rates larger than the preset click rate are associated commodities corresponding to the similar commodities.

3. The method as claimed in claim 2, wherein the step of regarding the associated commodities, of which the click rate of each associated commodity in each associated commodity list satisfies a preset condition, as alternative associated commodities comprises:

respectively obtaining the weighted click rate corresponding to each associated commodity in the associated commodity list according to the similarity between the similar commodity and the target commodity and the click rate of each associated commodity in the associated commodity list of the similar commodity;

and selecting at least one associated commodity with the weighted click rate larger than a preset click rate through the weighted click rate of each associated commodity in the associated commodity list, wherein the selected at least one associated commodity with the weighted click rate larger than the preset click rate is the alternative associated commodity.

4. The method as claimed in claim 3, wherein before selecting at least one associated item with a weighted click through rate greater than a preset click through rate by weighted click through rate of each associated item in the associated item list, the method further comprises:

determining the same associated commodities in the associated commodity list;

and obtaining the weighted click rate selected from the same associated commodities by the weighted click rate of the same associated commodity corresponding to each similar commodity, wherein the same associated commodity is the same associated commodity in an associated commodity list corresponding to different similar commodities.

5. The method as claimed in claim 2, wherein the step of regarding the associated commodities, of which the click rate of each associated commodity in each associated commodity list satisfies a preset condition, as alternative associated commodities comprises:

obtaining a plurality of weighted group click rates of the associated commodities according to the group click rates of the associated commodities corresponding to a plurality of user groups and the similarity corresponding to the associated commodities; the similarity corresponding to the associated commodity is the similarity between the similar commodity corresponding to the associated commodity and the target commodity;

obtaining the weighted click rate of the associated commodity according to the obtained click rates of the weighted groups;

and selecting at least one associated commodity with the weighted click rate larger than a preset click rate through the weighted click rate of each associated commodity, wherein the selected at least one associated commodity with the weighted click rate larger than the preset click rate is the alternative associated commodity.

6. The method of claim 1, wherein obtaining a plurality of similar goods similar to the target good based on business data of the target good under a plurality of sales attributes comprises:

inputting the business data of the target commodity corresponding to a plurality of sales attributes into a commodity information vector model to obtain a commodity information vector of the target commodity, obtaining at least one similar commodity information vector close to the commodity information vector, and determining the similar commodity corresponding to the similar commodity information vector.

7. The method of claim 6, wherein the merchandise information vector model is trained by:

obtaining a training sample, wherein the training sample is a plurality of training vectors obtained by performing feature extraction on business data of a commodity under a plurality of sales attributes, and each training vector corresponds to one sales attribute;

performing feature fusion processing on the training vectors to obtain commodity information vectors corresponding to the commodities;

obtaining type prediction corresponding to the commodity through the commodity information vector;

determining a loss value between the type prediction and a true type of the good through a first loss function;

and adjusting parameters of the commodity information vector model through the loss value until a preset condition is met.

8. The method of claim 7, wherein the training of the merchandise information vector model further comprises:

acquiring similar training samples, wherein the similar training samples comprise service data of two commodities;

acquiring commodity information vectors of the two service data and performing feature fusion to obtain a fusion commodity information vector of the similar training sample;

obtaining a predicted value of the proximity degree of the two commodities through the fusion commodity information vector;

determining a loss value between the proximity degree predicted value and a pre-labeled real proximity degree through a second loss function;

and adjusting parameters of the commodity information vector model through the loss value until a preset condition is met.

9. The method according to any one of claims 1 to 8, wherein the determining of the quasi-associated commodity of the target commodity in information display from the alternative associated commodities comprises:

inputting the business data of the alternative associated commodity into a vector generation model to obtain a commodity vector of the alternative associated commodity;

obtaining the association degree of the alternative associated commodities and the account number through the commodity vector and the account number vector of the user receiving the alternative associated commodities;

inputting the click rate data of the alternative associated commodities into a conjecture model to obtain click rate conjecture results of the alternative associated commodities;

obtaining the pushing degree of the alternative associated commodity according to the association degree and click rate conjecture result;

and selecting the to-be-associated commodity from the at least one candidate associated commodity according to the obtained pushing degree of each candidate associated commodity.

10. A big data based business management system, comprising a client and a server which are communicated with each other, wherein the client is used for sending business data of target commodities under a plurality of sales attributes to the server, the server comprises a processor and a memory which are communicated with each other, the processor is used for retrieving a computer program from the memory and realizing the method of any one of the above claims 1-9 by running the computer program.

Images