CN115794785A

CN115794785A - E-commerce data screening method and system based on big data and cloud platform

Info

Publication number: CN115794785A
Application number: CN202211178993.2A
Authority: CN
Inventors: 韩亚欣; 孙丹
Original assignee: Individual
Current assignee: China Soft International Technology Service Co ltd
Priority date: 2023-02-01
Filing date: 2023-02-01
Publication date: 2023-03-14
Anticipated expiration: 2043-02-01
Also published as: CN115794785B

Abstract

According to the E-commerce data screening method, the E-commerce data screening system and the cloud platform based on the big data, provided by the embodiment of the invention, the target E-commerce data noise field and each group of supervision data noise fields can be converted into a unified expert knowledge system for analysis and processing, and the feature association variable of the relationship network between the first E-commerce big data knowledge element relationship network and each group of second authentication E-commerce data noise field relationship network is determined through the feature association variable, so that the noise field association factor between the target E-commerce data noise field and each group of supervision data noise fields is determined, the complexity of determining the feature association variable between the target E-commerce data noise field and each group of supervision data noise fields is simplified as much as possible, the noise association identification efficiency aiming at different E-commerce data noise fields is ensured, not only can the feature association variable of the relevant E-commerce data noise fields be accurately and reliably calculated, but also a certain cloud platform computing power can be released, and the timeliness of subsequent data screening is improved.

Description

E-commerce data screening method and system based on big data and cloud platform

Technical Field

The invention relates to the technical field of data processing, in particular to an e-commerce data screening method and system based on big data and a cloud platform.

Background

The electronic commerce refers to a commerce activity taking an information network technology as a means and taking commodity exchange as a center, can also be understood as an activity of performing transaction activities and related services in an electronic transaction mode on the internet, an intranet and a value-added network, and is an electronization, networking and informatization upgrade of each link of the traditional commerce activity. At present, the application scale of electronic commerce is becoming wider and wider, and management of electronic commerce data becomes a focus of attention of all parties. In some application scenarios, too much e-commerce data causes processing pressure of a platform system, and noise cleaning of the e-commerce data is not easy, but correlation between noise data is difficult to accurately and efficiently identify by related technologies, so that timeliness of subsequent noise screening is difficult to guarantee.

Disclosure of Invention

In order to solve the technical problems in the related art, the invention provides an e-commerce data screening method and system based on big data and a cloud platform.

In a first aspect, an embodiment of the present invention provides an e-commerce data screening method based on big data, which is applied to an e-commerce data screening cloud platform, and the method includes:

determining a first E-business big data knowledge element relation network in which a characteristic association variable between the E-business big data knowledge elements and a target E-business data noise field reaches a first set characteristic association variable limit value and a second E-business big data knowledge element relation network in which a characteristic association variable between the E-business big data knowledge elements and a supervision data noise field reaches the first set characteristic association variable limit value through each group of E-business big data knowledge elements acquired from a shared service system;

respectively obtaining a relation network characteristic association variable between the first E-commerce big data knowledge element relation network and each group of second E-commerce big data knowledge element relation networks, and determining a noise field association factor between the target E-commerce data noise field and each group of supervision data noise fields;

wherein: the first E-business big data knowledge element relation network is obtained through not less than one group of first E-business big data knowledge elements in each group of E-business big data knowledge elements; the second E-commerce big data knowledge element relation network is obtained through no less than one group of second E-commerce big data knowledge elements in each group of E-commerce big data knowledge elements.

In a possible embodiment, before determining, through the sets of e-commerce big data knowledge elements acquired from the shared service system, a first e-commerce big data knowledge element relationship network where a feature association variable between the e-commerce big data knowledge element relationship network and a target e-commerce data noise field reaches a first set feature association variable limit value, the method further includes:

dividing each group of the supervision data noise fields into a plurality of groups of local supervision data noise fields with the same number one by one, and determining a local supervision data noise field relation network corresponding to each group of the supervision data noise fields; each group of local supervision data noise fields in the local supervision data noise field relation network are updated according to the distribution priority of each group of local supervision data noise fields in the corresponding supervision data noise field;

and determining corresponding big data knowledge elements not lower than one group of E-commerce big data knowledge elements based on the local supervision data noise fields under the same distribution variable in each group of local supervision data noise field relation network, and determining each group of E-commerce big data knowledge elements acquired from the shared service system.

In a possible embodiment, determining, based on the local supervised data noise fields under the same distribution variable in each group of local supervised data noise field relationship networks, corresponding not less than one group of e-commerce big data knowledge elements, and determining each group of e-commerce big data knowledge elements acquired from the shared service system, includes:

in each group of local supervision data noise field relation networks, taking local supervision data noise fields under the same distribution variable as a local power company big data noise relation network, carrying out noise clustering processing on each group of local power company big data noise relation networks, and determining that no less than one group of power company big data knowledge elements respectively corresponding to each group of local power company big data noise relation networks are determined;

and determining each group of E-commerce big data knowledge elements acquired from the shared service system through not less than one group of E-commerce big data knowledge elements corresponding to each group of local E-commerce big data noise relationship networks.

In a possible embodiment, on the premise that the local e-commerce big data noise relationship network and the distribution variables of the local supervision data noise fields included in the local e-commerce big data noise relationship network in the corresponding supervision data noise fields have correlation, a first e-commerce big data knowledge element relationship network that determines, through each group of e-commerce big data knowledge elements acquired from the shared service system, that a feature association variable between the target e-commerce big data noise field and the feature association variable reaches a first set feature association variable limit value includes:

dividing the target e-commerce data noise field into a plurality of groups of target local e-commerce data noise fields, and determining a target local e-commerce data noise field relation network of the target e-commerce data noise field; each group of target local e-commerce data noise fields in the target local e-commerce data noise field relation network are updated according to the distribution priority of each group of target local e-commerce data noise fields in the target e-commerce data noise fields;

determining a first e-commerce big data knowledge element, which corresponds to the local e-commerce big data noise relationship network and is not lower than a group of e-commerce big data knowledge elements, and of which the characteristic association variable between the first e-commerce big data knowledge element and the target local e-commerce data noise field reaches a second set characteristic association variable limit value; the distribution variable of the target local e-commerce data noise field in the target local e-commerce data noise field relation network is consistent with the linkage distribution variable of the local e-commerce big data noise relation network;

when each group of target local e-commerce data noise fields has a corresponding first e-commerce big data knowledge element, determining the first e-commerce big data knowledge element relation network obtained by the first e-commerce big data knowledge elements of each group, and determining the first e-commerce big data knowledge element relation network corresponding to the target e-commerce data noise fields when the characteristic association variable between the target e-commerce big data noise fields and the first e-commerce big data knowledge element relation network reaches a first set characteristic association variable limit value.

In a possible embodiment, when each group of e-commerce big data knowledge elements acquired from the shared service system is annotated with a knowledge element keyword, and the knowledge element keyword is used for independently reflecting each group of e-commerce big data knowledge elements, determining a first e-commerce big data knowledge element relationship network corresponding to the target e-commerce data noise field, including: and determining a first E-commerce big data knowledge element relation network corresponding to the target E-commerce data noise field through each group of knowledge element key words of the first E-commerce big data knowledge element.

In a possible embodiment, on the premise that the local e-commerce big data noise relationship network and the distributed variables of the local supervision data noise fields included in the local e-commerce big data noise relationship network in the corresponding supervision data noise fields have correlation, determining, by each group of e-commerce big data knowledge elements acquired from the shared service system, each group of second e-commerce big data knowledge element relationship networks in which the feature association variables between each group of supervision data noise fields acquired from the shared service system respectively reach the first set feature association variable limit value, including:

determining a second E-commerce big data knowledge element, of which the characteristic association variable between the local supervision data noise field reaches a second set characteristic association variable limit value, in the E-commerce big data knowledge elements which correspond to the local E-commerce big data noise relationship network and are not lower than a group of E-commerce big data knowledge elements; the distribution variable of the local supervision data noise field in the local supervision data noise field relation network is consistent with the linkage distribution variable of the local power grid big data noise relation network;

when each group of local supervision data noise fields has a corresponding second E-business big data knowledge element, determining a second E-business big data knowledge element relation network obtained by each group of second E-business big data knowledge elements, and determining the second E-business big data knowledge element relation network corresponding to the supervision data noise fields when a characteristic association variable between the second E-business big data knowledge element relation network and the supervision data noise fields reaches a first set characteristic association variable limit value.

In a possible embodiment, on the basis that the number of first e-commerce big-data knowledge elements in the first e-commerce big-data knowledge element relationship network is consistent with the number of second e-commerce big-data knowledge elements in the second e-commerce big-data knowledge element relationship network, obtaining a relationship network characteristic association variable between the first e-commerce big-data knowledge element relationship network and each group of the second e-commerce big-data knowledge element relationship networks respectively comprises:

respectively obtaining local knowledge characteristic association variables between each group of first e-commerce big data knowledge elements in the first e-commerce big data knowledge element relation network and second e-commerce big data knowledge elements of corresponding distribution variables in the second e-commerce big data knowledge element relation network;

and weighting the obtained local knowledge characteristic association variables according to a credible factor to determine the relation network characteristic association variables between the first E-commerce big data knowledge element relation network and the second E-commerce big data knowledge element relation network.

In a second aspect, the present invention further provides an e-commerce data screening system based on big data, where the system includes an e-commerce data screening cloud platform and an e-commerce interaction terminal, where the e-commerce data screening cloud platform is used for: determining a first E-business big data knowledge element relation network in which a characteristic association variable between each group of E-business big data knowledge elements acquired from a shared service system reaches a first set characteristic association variable limit value and a second E-business big data knowledge element relation network in which a characteristic association variable between each group of E-business big data knowledge elements and a target E-business data noise field reaches the first set characteristic association variable limit value; respectively obtaining a relation network characteristic association variable between the first E-commerce big data knowledge element relation network and each group of second E-commerce big data knowledge element relation networks, and determining a noise field association factor between the target E-commerce data noise field and each group of supervision data noise fields; wherein: the first E-business big data knowledge element relation network is obtained through not less than one group of first E-business big data knowledge elements in each group of E-business big data knowledge elements; the second E-commerce big data knowledge element relation network is obtained through not less than one group of second E-commerce big data knowledge elements in each group of E-commerce big data knowledge elements.

In a third aspect, the invention further provides an e-commerce data screening cloud platform, which comprises a processor and a storage; the processor is connected with the memory in communication, and the processor is used for reading the computer program from the memory and executing the computer program to realize the method.

In a fourth aspect, the present invention also provides a readable storage medium, on which a program is stored, which when executed by a processor implements the method described above.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and together with the description, serve to explain the principles of the invention.

Fig. 1 is a schematic flow chart of an e-commerce data screening method based on big data according to an embodiment of the present invention.

Fig. 2 is a schematic communication architecture diagram of an e-commerce data screening system based on big data according to an embodiment of the present invention.

Detailed Description

Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the invention, as detailed in the appended claims.

It should be noted that the terms "first," "second," and the like in the description and claims of the present invention and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order.

The method provided by the embodiment of the invention can be executed in an e-commerce data screening cloud platform, a computer device or a similar operation device. Taking the example of operating on an e-commerce data screening cloud platform, the e-commerce data screening cloud platform 10 may include one or more processors 102 (the processors 102 may include but are not limited to processing devices such as a microprocessor MCU or a programmable logic device FPGA), and a memory 104 for storing data, and optionally, the e-commerce data screening cloud platform may further include a transmission device 106 for communication functions. It will be understood by those of ordinary skill in the art that the above structure is merely illustrative, and does not limit the structure of the above e-commerce data screening cloud platform. For example, the e-commerce data screening cloud platform 10 may also include more or fewer components than shown above, or have a different configuration than shown above.

The memory 104 may be used to store computer programs, for example, software programs and modules of application software, such as a computer program corresponding to a big data-based e-commerce data screening method in an embodiment of the present invention, and the processor 102 executes various functional applications and data processing by running the computer programs stored in the memory 104, so as to implement the above-mentioned method. The memory 104 may include high speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include memory remotely located from the processor 102, which may be connected to the e-commerce data screening cloud platform 10 over a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.

The transmission device 106 is used for receiving or transmitting data via a network. Specific examples of the network described above may include a wireless network provided by a communication provider of the e-commerce data screening cloud platform 10. In one example, the transmission device 106 includes a Network adapter (NIC), which can be connected to other Network devices through a base station so as to communicate with the internet. In one example, the transmission device 106 may be a Radio Frequency (RF) module, which is used for communicating with the internet in a wireless manner.

In the embodiment of the invention, in each group of E-business big data knowledge elements acquired from a shared service system, a first E-business big data knowledge element relation network corresponding to a target E-business data noise field is determined by determining that a characteristic association variable between the E-business big data knowledge element relation network and a supervision data noise field reaches a first set characteristic association variable limit value and is not lower than a group of first E-business big data knowledge elements, and a second E-business big data knowledge element relation network corresponding to each group of supervision data noise fields is determined by determining that a characteristic association variable between the E-business big data knowledge element relation network and a supervision data noise field reaches a first set characteristic association variable limit value and is not lower than a group of second E-business big data knowledge elements. Therefore, the target e-commerce data noise field and each group of supervision data noise fields are converted into a uniform expert knowledge system for analysis and processing, and the feature association variable between the first e-commerce big data knowledge element relationship network and each group of second authentication e-commerce data noise field relationship network is determined through the feature association variable between each group of e-commerce big data knowledge elements acquired from the shared service system, so that the noise field association factor between the target e-commerce data noise field and each group of supervision data noise fields is determined, the step of determining the feature association variable between the target e-commerce data noise field and each group of supervision data noise fields is simplified as much as possible, the efficiency of noise association identification for different e-commerce data noise fields is ensured, the feature association variable of the related e-commerce data noise fields can be accurately and reliably calculated, certain cloud platform computing power can be released, and the timeliness of subsequent noise data screening is improved.

For example, a CPU of the e-commerce data screening cloud platform acquires, from a memory unit of the e-commerce data screening cloud platform, each set of e-commerce big data knowledge elements acquired from the shared service system. The CPU of the e-commerce data screening cloud platform determines a first e-commerce big data knowledge element relation network, wherein the feature association variables between the e-commerce big data knowledge elements and the target e-commerce data noise fields reach a first set feature association variable limit value, through each group of e-commerce big data knowledge elements acquired from the shared service system. The CPU of the e-commerce data screening cloud platform determines a second e-commerce big data knowledge element relation network, wherein the second e-commerce big data knowledge element relation network enables feature association variables between each group of monitored data noise fields acquired from the shared service system to reach a first set feature association variable limit value, through each group of e-commerce big data knowledge elements acquired from the shared service system. The first E-commerce big data knowledge element relation network is obtained through at least one group of first E-commerce big data knowledge elements in each group of E-commerce big data knowledge elements, and the second E-commerce big data knowledge element relation network is obtained through at least one group of second E-commerce big data knowledge elements in each group of E-commerce big data knowledge elements.

And respectively obtaining a relation network characteristic association variable between the first E-commerce big data knowledge element relation network and each group of second E-commerce big data knowledge element relation networks by a CPU (Central processing Unit) of the E-commerce data screening cloud platform, and determining a noise field association factor between a target E-commerce data noise field and each group of supervision data noise fields.

Based on this, please refer to fig. 1, fig. 1 is a schematic flowchart of a big data-based e-commerce data screening method provided by an embodiment of the present invention, the method is applied to an e-commerce data screening cloud platform, and may further include the technical solutions described in the following NODEs 201 to 204.

NODE201, determining each group of E-commerce big data knowledge elements.

For the embodiment of the invention, each group of E-commerce big data knowledge elements acquired from the shared service system in the memory unit of the E-commerce data screening cloud platform can be determined by each group of supervision data noise fields after a CPU of the E-commerce data screening cloud platform determines each group of supervision data noise fields; or the data is determined by each group of supervision data noise fields when the CPU of the e-commerce data screening cloud platform is not busy; or the CPU of the E-commerce data screening cloud platform acquires guide data for determining each group of E-commerce big data knowledge elements, and then determines the guide data through each group of supervision data noise fields. The supervision data noise field (reference e-commerce data noise field) is an e-commerce data noise field (which may be, for example, a corresponding feature field, a feature array, etc.) of a reference e-commerce interaction session determined after the reference e-commerce interaction session is identified by the key information.

After the CPU of the e-commerce data screening cloud platform determines each group of e-commerce big data knowledge elements (such as standard expert knowledge features), each group of e-commerce big data knowledge elements can be stored in the memory unit of the e-commerce data screening cloud platform, so that each group of e-commerce big data knowledge elements can be conveniently searched for by the memory unit of the e-commerce data screening cloud platform when a target e-commerce data noise field needs to be determined by each group of e-commerce big data knowledge elements.

For some examples, a mapping list (corresponding relationship) between each group of reference e-commerce interaction sessions and a supervision data noise field corresponding to the reference e-commerce interaction sessions may be saved in the memory unit of the e-commerce data screening cloud platform, and by determining the supervision data noise field, a reference e-commerce interaction session corresponding to the supervision data noise field may be correspondingly determined.

For some examples, the saved supervision data noise field may be determined by performing feature normalization on the obtained e-commerce data noise field after the e-commerce data noise field determined by performing noise field mining on the reference e-commerce interaction session, so that negative effects on the feature association variable may be avoided, and the accuracy and the reliability of the determined feature association variable may be improved.

For some examples, the stored supervision data noise field may be determined after pooling the obtained e-commerce data noise field after the e-commerce data noise field determined by performing noise field mining on the reference e-commerce interaction session, and further, the pooling may be implemented, for example, based on a relevant feature processing algorithm model, by first determining a noise attention matrix V1 of the e-commerce data noise field, and then determining a pooled field matrix V2 of the e-commerce data noise field in combination with a V1= V0V 2 manner, where V0 is the e-commerce data noise field. By avoiding the relation among noises, the accuracy and the reasonableness in the subsequent determination of the characteristic association variables among the E-commerce data noise fields are improved.

In some possible embodiments, detailed technical solutions for determining the knowledge elements of the e-commerce big data of each group can be exemplarily described.

And the NODE1-1 is used for dividing each group of the supervision data noise fields into a plurality of groups of local supervision data noise fields with the same number one by one and determining a local supervision data noise field relation network corresponding to each group of the supervision data noise fields.

For the embodiment of the present invention, after the noise fields of the supervision data are determined, each group of the noise fields of the supervision data may be divided into a plurality of groups of local noise fields of the supervision data with the same number, and the local noise field relationship network corresponding to each group of the noise fields of the supervision data is determined. For example, the surveillance data noise field is [ noiseA, noiseB, noiseC, noiseD, noiseE, noiseF, noiseG, noiseH, and noiseI ], and if the surveillance data noise field is divided into three groups, the three groups of local surveillance data noise fields may be [ noiseA, noiseB, noiseC ], noiseD, noiseE, noiseF ], and [ noiseG, noiseH, and noiseI ], respectively, so that the determined local surveillance data noise field relation network may be [ noiseA, noiseB, noiseC ], [ noiseD, noiseE, noiseF ], noiseG, noiseH, and noiseI ]. And updating each group of local supervision data noise fields in the local supervision data noise field relation network according to the distribution priority of each group of local supervision data noise fields in the supervision data noise fields.

For some examples, in segmenting the supervised data noise field, the supervised data noise field may be segmented averagely by segmenting the number (quantity) of local supervised data noise fields; alternatively, the supervised data noise field may be non-averagely partitioned by individual field feature values in the supervised data noise field.

For some examples, the number of split local supervisory data noise fields may be determined by a size of the supervisory data noise field, which may be an integer multiple of the number of split local supervisory data noise fields, e.g., the scale of the supervisory data noise field is 64 dimensions, then the number of split local supervisory data noise fields may be 2, etc. The number of the local supervised data noise fields may be set in advance by a preset value, or may be determined by referring to the number of divisions.

By way of example, a locally supervised data noise field may be understood as a data noise field used as a reference. A local supervised data noise field relational network may be understood as a set/matrix of supervised data noise fields used as a reference.

And NODE1-2, taking the local supervision data noise fields under the same distribution variable in each group of local supervision data noise field relation networks as a local power company big data noise relation network, carrying out noise clustering processing on each group of local power company big data noise relation networks, and determining that no less than one group of power company big data knowledge elements respectively corresponding to each group of local power company big data noise relation networks.

For the embodiment of the present invention, after each group of the monitoring data noise fields is segmented, and the local monitoring data noise field relationship network of each group of the monitoring data noise fields is determined, the local monitoring data noise fields in each group of the local monitoring data noise field relationship network under the same distribution variable (which can be understood as under the same position) are used as a local power grid big data noise relationship network. For example, a local supervision data noise field relation network is [ noise a, noise b, noise c ] [ noise d, noise e, noise f ] [ noise g, noise h, noise i ] and another local supervision data noise field relation network is [ noise b, noise c, noise d ] [ noise e, noise g, noise f ] [ noise h, noise i, noise a ], then the local supervision data noise fields under the first distribution variable are noise a, noise b, noise e ] and [ noise b, noise d ] as well as the local supervision data noise fields under the first distribution variable are noise a, noise b, noise e ] and the noise b, noise e, noise d ] as a local electrical quotient big data noise network, and the local supervision data under the second distribution variable are noise e, noise e ] and the noise field under the third distribution variable are as the local supervision data noise field under the first distribution variable.

It can be understood that the number of the determined local power grid big data noise relationship networks is consistent with the number of the local supervision data noise fields, each group of local power grid big data noise relationship networks is not lower than one group of local supervision data noise fields, each group of local power grid big data noise relationship networks is subjected to noise clustering (such as Kmeans classification), and similar local supervision data noise fields are induced and integrated. After the noise clustering processing is carried out on the local power grid big data noise relationship network, it can be determined that not less than one group of noise sets corresponding to the local power grid big data noise relationship network, local supervision data noise fields in the integrated grouping are induced in the same way, and feature association variables (which can be understood as feature similarity) between the noise sets of the induced integrated grouping and the local supervision data noise fields are in a set value interval. Therefore, not less than one group of noise sets corresponding to each group of local power grid big data noise relationship networks can be determined, and one noise set can be understood as a power grid big data knowledge element.

For some examples, the more the monitoring data noise fields, the greater the difference between the monitoring data noise fields may be, so that when the local power grid big data noise relationship network is subjected to noise clustering processing, the number of power grid big data knowledge elements may be increased according to the increase of the monitoring data noise fields or decreased according to the decrease of the monitoring data noise fields, so that the accuracy and the reliability of determining the power grid big data knowledge elements can be improved, and the accuracy and the reliability of determining the characteristic associated variables can be improved. For example, the number of the e-commerce big data knowledge elements may be set to 200 for a large number of the supervision data noise fields, set to 400 for a large number of the supervision data noise fields, and so on.

For example, the local power-provider big data noise relationship net can be understood as a power-provider big data noise vector set. The e-commerce big data knowledge element can be understood as a standard expert knowledge vector (such as e-commerce big data knowledge characteristics extracted through an existing expert system model).

And the NODE1-3 determines the various groups of E-commerce big data knowledge elements acquired from the shared service system through not less than one group of E-commerce big data knowledge elements corresponding to each group of local E-commerce big data noise relationship networks.

For the embodiment of the invention, after determining that each group of local e-commerce big data noise relationship networks respectively corresponds to not less than one group of e-commerce big data knowledge elements, the CPU of the e-commerce data screening cloud platform can issue each group of e-commerce big data knowledge elements to the memory unit of the e-commerce data screening cloud platform, and after the memory unit of the e-commerce data screening cloud platform acquires each group of e-commerce big data knowledge elements issued by the CPU of the e-commerce data screening cloud platform, each group of e-commerce big data knowledge elements is recorded, and each group of e-commerce big data knowledge elements acquired from the sharing service system is determined.

And the NODE202 determines each group of second E-commerce big data knowledge element relation networks, wherein the second E-commerce big data knowledge element relation networks respectively reach a first set characteristic association variable limit value with the characteristic association variable between each group of supervision data noise fields acquired from the shared service system through each group of E-commerce big data knowledge elements acquired from the shared service system.

For the embodiment of the invention, the local power grid big data noise relationship network is matched with the distribution variables of the local supervision data noise fields in the corresponding supervision data noise field relationship network, wherein the local supervision data noise fields are included in the local power grid big data noise relationship network.

And determining a second E-commerce big data knowledge element, wherein the characteristic association variable between the second E-commerce big data knowledge element and the local supervision data noise field under the matching distribution variable of the local E-commerce big data noise relationship network reaches a second set characteristic association variable limit value, in the E-commerce big data noise relationship network corresponding to the group of E-commerce big data knowledge elements not lower than the group of E-commerce big data knowledge elements. For example, in the electricity business big data knowledge elements not lower than a group of electricity business big data knowledge elements corresponding to the local electricity business big data noise relation network which satisfies the matching condition and has the first distribution variable, the second electricity business big data knowledge element is determined, wherein the feature related variable between the local supervision data noise fields under the first distribution variable (the relative position in the feature coordinate system corresponding to the expert knowledge system) in the supervision data noise field relation network reaches the second set feature related variable limit value.

The second set characteristic-related variable limit value may be that the characteristic-related variable is within the set numerical value interval, or that the characteristic-related variable is greater than the set determination value, or the like. Thus, the number of the determined second e-commerce big data knowledge elements can be a plurality of groups. The second set characteristic-related variable limit value may also be a characteristic-related variable maximum, based on which the number of the determined second e-commerce big data knowledge elements may be one. When the number of the second e-commerce big data knowledge elements is a plurality of groups, each group of the second e-commerce big data knowledge elements can be sorted according to the size of the characteristic association variable and the actual situation.

The second E-commerce big data knowledge elements corresponding to each group of local supervision data noise fields in the local supervision data noise field relation network can be determined by not less than one group of E-commerce big data knowledge elements corresponding to each group of local supervision data noise relation network, when each group of local supervision data noise fields has one corresponding second E-commerce big data knowledge element, the second E-commerce big data knowledge element relation network obtained by each group of second E-commerce big data knowledge elements is determined, and the feature association variable between the second E-commerce big data knowledge elements and the supervision data noise fields reaches the first set feature association variable limit value. And when the characteristic associated variable between the second E-business big data knowledge element relation network and the supervision data noise field reaches a first set characteristic associated variable limit value, determining the second E-business big data knowledge element relation network corresponding to the supervision data noise field.

After the second e-commerce big data knowledge element relation network corresponding to each group of the supervision data noise fields is determined, the CPU of the e-commerce data screening cloud platform can send the second e-commerce big data knowledge element relation network corresponding to each group of the supervision data noise fields to the memory unit of the e-commerce data screening cloud platform, the memory unit of the e-commerce data screening cloud platform obtains the second e-commerce big data knowledge element relation network corresponding to each group of the supervision data noise fields sent by the CPU of the e-commerce data screening cloud platform and records the second e-commerce big data knowledge element relation network, so that the CPU of the e-commerce data screening cloud platform can directly obtain the second e-commerce big data knowledge element relation network corresponding to each group of the supervision data noise fields to determine when determining the feature association variables between the target e-commerce data noise fields and each group of the supervision data noise fields, the second e-commerce big data knowledge element relation network corresponding to each group of the supervision data noise fields does not need to determine, and timeliness of determining the feature association variables between the target e data noise fields and each group of the supervision data noise fields can be improved.

For some examples, there may be one knowledge element keyword per group of e-commerce big data knowledge elements that may independently reflect each group of e-commerce big data knowledge elements; or each group of E-commerce big data knowledge elements can have knowledge element keywords which can independently reflect the E-commerce big data knowledge elements in the corresponding local E-commerce big data noise relationship network. A second E-business big data knowledge element relation network for monitoring data noise fields can record through a knowledge element keyword relation network, a CPU of the E-business data screening cloud platform does not need to correspondingly process each group of E-business big data knowledge elements by conveying each group of E-business big data knowledge elements in the processing process, and the E-business big data knowledge elements needing to be processed can be determined only by conveying knowledge element keywords, so that the calculation capacity utilization rate is improved.

And the NODE203 determines a first E-commerce big data knowledge element relation network in which the characteristic associated variable between the E-commerce big data knowledge element relation network and the target E-commerce data noise field reaches a first set characteristic associated variable limit value through each group of E-commerce big data knowledge elements acquired from the shared service system.

For the embodiment of the invention, when the noise field correlation factor between the target e-commerce data noise field and each group of supervision data noise fields needs to be determined, the target e-commerce data noise field can be firstly segmented, and each group of target local e-commerce data noise fields corresponding to the target e-commerce data noise field are determined. The number of the local supervision data noise fields of the target local e-commerce data noise field and the supervision data noise field is kept consistent, the target e-commerce data noise field is divided into all groups of target local e-commerce data noise fields, the thought of the target local e-commerce data noise field relation network is determined, and the thought of the local supervision data noise field relation network is consistent with the thought of dividing the supervision data noise field into all groups of local supervision data noise fields and determining the local supervision data noise field relation network.

The process of determining a first E-business big data knowledge element relation network in which the characteristic associated variable between the E-business big data knowledge elements and the target E-business data noise field reaches a first set characteristic associated variable limit value in each group of E-business big data knowledge elements acquired from the shared service system is consistent with the idea of determining a second E-business big data knowledge element relation network in which the characteristic associated variable between the E-business big data knowledge elements and the supervision data noise field reaches the first set characteristic associated variable limit value in each group of E-business big data knowledge elements acquired from the shared service system.

For some examples, the first e-commerce big data knowledge element relation network of the target e-commerce data noise field can be recorded through a knowledge element keyword relation network, and in the processing process of the CPU of the e-commerce data screening cloud platform, the e-commerce big data knowledge elements to be processed can be determined by only transmitting knowledge element keywords without transmitting the e-commerce big data knowledge elements to correspondingly process the e-commerce big data knowledge elements, so that the computational efficiency is improved.

And the NODE204 is used for respectively obtaining the relation network characteristic association variables between the first E-commerce big data knowledge element relation network and each group of second E-commerce big data knowledge element relation networks and determining the noise field association factor between the target E-commerce data noise field and each group of supervision data noise fields.

For embodiments of the present invention, after determining the first e-commerce big-data knowledge element relationship network of the target e-commerce data noise field, a relationship network characteristic association variable between the first e-commerce big-data knowledge element relationship network and each set of second e-commerce big-data knowledge element relationship networks may be determined. In view of the fact that the first e-commerce big data knowledge element relationship network and each group of the second e-commerce big data knowledge element relationship networks are obtained through each group of e-commerce big data knowledge elements, it can be seen that the relationship network feature association variables of the first e-commerce big data knowledge element relationship network and the second e-commerce big data knowledge element relationship network can be determined by determining the local knowledge feature association variables between the first e-commerce big data knowledge elements and the second e-commerce big data knowledge elements of the corresponding distribution variables in the first e-commerce big data knowledge element relationship network and the second e-commerce big data knowledge element relationship network.

Further, when determining the local knowledge characteristic associated variable between the first e-commerce big data knowledge element and the second e-commerce big data knowledge element corresponding to the distributed variable, the local knowledge characteristic associated variable between the first e-commerce big data knowledge element and the second e-commerce big data knowledge element can be found from the local knowledge characteristic associated variables between the sets of e-commerce big data knowledge elements acquired from the shared service system through the knowledge element key words of the first e-commerce big data knowledge element and the knowledge element key words of the second e-commerce big data knowledge element corresponding to the distributed variable. When local knowledge characteristic associated variables corresponding to the knowledge element keywords of the first e-commerce big data knowledge element and the knowledge element keywords of the second e-commerce big data knowledge element are searched, the knowledge element keywords can be directly searched in the memory unit of the e-commerce data screening cloud platform, or each group of knowledge element keywords of the first e-commerce big data knowledge element relationship network and the corresponding relation between each group of knowledge element keywords of each group of e-commerce big data knowledge elements in the local e-commerce big data noise relationship network can be generated, and a local knowledge characteristic associated variable relation table of each group of e-commerce big data knowledge elements in the first e-commerce big data knowledge element relationship network and the local e-commerce big data noise relationship network is determined, so that the local knowledge characteristic associated variables between each group of the first e-commerce big data knowledge element and each group of the second e-commerce big data knowledge element can be searched and determined through the knowledge element keywords. After the local knowledge characteristic associated variables between the first e-commerce big data knowledge element and each group of second e-commerce big data knowledge elements are determined, weighting according to a credible factor can be carried out on the obtained local knowledge characteristic associated variables, the thought of the first e-commerce big data knowledge element relationship network and the second e-commerce big data knowledge element relationship network is determined through the local knowledge characteristic associated variable mean value, the contribution values (influence degree and importance) of the local knowledge characteristic associated variables with differences can be corrected, and the accuracy of the first e-commerce big data knowledge element relationship network and the second e-commerce big data knowledge element relationship network is determined to be higher.

For some examples, if one target local e-commerce data noise field corresponds to several groups of first e-commerce big data knowledge elements, for example, the target local e-commerce data noise field corresponds to T first e-commerce big data knowledge elements, the mapping list of the local knowledge characteristic association variables of the first e-commerce big data knowledge element relationship network and each group of e-commerce big data knowledge elements in the local e-commerce big data noise relationship network may be recorded through a table according to actual requirements.

For example, a target local e-commerce data noise field corresponds to a plurality of groups of first e-commerce big data knowledge elements, each group of first e-commerce big data knowledge elements may carry a weight, and the larger the feature association variable between the first e-commerce big data knowledge element and the corresponding target local e-commerce data noise field is, the larger the weight of the first e-commerce big data knowledge element is. Local knowledge characteristic association variables between each group of first E-business big data knowledge elements and second E-business big data knowledge elements in a plurality of groups of first E-business big data knowledge elements corresponding to the target local E-business data noise field are respectively obtained, the obtained local knowledge characteristic association variables are weighted according to the credible factors through the weight of each group of first E-business big data knowledge elements, and the local knowledge characteristic association variables between the first E-business big data knowledge elements and the second E-business big data knowledge elements are determined. After the local knowledge characteristic association variables between the first e-commerce big data knowledge element and each group of second e-commerce big data knowledge elements are determined, weighting according to the confidence factor can be performed on each group of obtained local knowledge characteristic association variables, and a relationship network characteristic association variable between the first e-commerce big data knowledge element relationship network and the second e-commerce big data knowledge element relationship network is determined.

After the relation network characteristic association variable between the first e-commerce big data knowledge element relation network and the second e-commerce big data knowledge element relation network is determined, the relation network characteristic association variable between the first e-commerce big data knowledge element relation network and the second e-commerce big data knowledge element relation network is used as a noise field association factor between the target e-commerce data noise field and the supervision data noise field, and therefore the noise field association factor between the target e-commerce data noise field and each group of supervision data noise fields can be determined.

For some examples, after determining the noise field correlation factor between the target e-commerce data noise field and each set of supervisory data noise fields, the sets of supervisory data noise fields may be sorted by a magnitude of the noise field correlation factor, and the supervisory data noise fields in the sets of supervisory data noise fields that precede the sort location are derived by the sort location.

For some examples, the feature association variable determination in the embodiment of the present invention may be determined by determining cosine similarity between two noise fields, and the like.

The big data-based e-commerce data screening method provided below can also be implemented as follows.

NODE301, the CPU of the e-commerce data screening cloud platform determines sets of supervised data noise fields. And the NODE302 is that a CPU of the e-commerce data screening cloud platform performs field normalization and pooling processing on each group of supervision data noise fields through each group of supervision data noise fields, and determines each group of processed supervision data noise fields. And NODE303, a CPU of the e-commerce data screening cloud platform determines each group of e-commerce big data knowledge elements through each group of processed supervision data noise fields. And NODE304, recording each group of E-commerce big data knowledge elements in a memory unit of the E-commerce data screening cloud platform by a CPU of the E-commerce data screening cloud platform. NODE305, the CPU of the e-commerce data screening cloud platform determines feature association variables between every two e-commerce big data knowledge elements. And NODE306, recording feature association variables among all groups of E-commerce big data knowledge elements in a memory unit of the E-commerce data screening cloud platform by a CPU of the E-commerce data screening cloud platform. And NODE307, determining a second E-commerce big data knowledge element relation network corresponding to each group of supervision data noise fields by a CPU (central processing unit) of the E-commerce data screening cloud platform through each group of E-commerce big data knowledge elements. NODE308, a CPU of the e-commerce data screening cloud platform may record each group of second e-commerce big data knowledge element relationship networks in a memory unit of the e-commerce data screening cloud platform. And NODE309, when a supervision data noise field similar to the target e-commerce data noise field needs to be determined in each group of supervision data noise fields, the CPU of the e-commerce data screening cloud platform determines the target e-commerce data noise field. And NODE310, a CPU of the e-commerce data screening cloud platform determines a first e-commerce big data knowledge element relation network corresponding to a target e-commerce data noise field through each group of e-commerce big data knowledge elements. And the NODE311 is that a CPU of the e-commerce data screening cloud platform searches a second e-commerce big data knowledge element relation network corresponding to each group of supervision data noise fields in the memory unit of the e-commerce data screening cloud platform and characteristic association variables among each group of e-commerce big data knowledge elements. NODE312, a CPU of the e-commerce data screening cloud platform determines a relation network characteristic association variable between a first e-commerce big data knowledge element relation network and a second e-commerce big data knowledge element relation network.

Generally, the supervisory data noise field is divided into three local supervisory data noise fields, and correspondingly, the target e-commerce data noise field is divided into three target local e-commerce data noise fields. Each group of local supervision data noise fields corresponds to two second e-commerce big data knowledge elements, correspondingly, each group of target local e-commerce data noise fields corresponds to two first e-commerce big data knowledge elements, local knowledge characteristic association variables between the first e-commerce big data knowledge elements and the second e-commerce big data knowledge elements corresponding to each group are determined, weighted summation is carried out on the local knowledge characteristic association variables, relation network characteristic association variables between a first e-commerce big data knowledge element relation network and a second e-commerce big data knowledge element relation network are determined, and noise field association factors between the target e-commerce data noise fields and the supervision data noise fields are determined.

The relation network can be constructed based on a knowledge graph technology, can be generated by combining a computer vision technology, is high in universality, and can improve the efficiency and flexibility of noise screening and analyzing.

In some embodiments, after determining the noise field correlation factor between the target e-commerce data noise field and each set of supervisory data noise fields, the method may further include: determining a to-be-processed supervision data noise field with a noise field association factor (noise similarity) higher than a set noise screening factor (set noise similarity) of the target e-commerce data noise field from each set of supervision data noise fields, and determining a noise cleaning strategy for target e-commerce big data based on the to-be-processed supervision data noise field and the target e-commerce data noise field; and carrying out noise cleaning on the target E-business big data through the noise cleaning strategy.

The value of the set noise screening factor can be flexibly set according to the actual computing power of the e-commerce data screening cloud platform, if the actual computing power of the e-commerce data screening cloud platform is high, the value of the set noise screening factor can be adjusted to be low, and if the actual computing power of the e-commerce data screening cloud platform is low, the value of the set noise screening factor can be increased, so that the computing power utilization rate is adaptively improved, and overload of the e-commerce data screening cloud platform is avoided.

In addition, determining a noise cleaning strategy for target e-commerce big data based on the to-be-processed supervision data noise field and the target e-commerce data noise field may include the following: performing feature fusion on the to-be-processed supervision data noise field and the target e-commerce data noise field to obtain a fusion noise field; carrying out noise influence analysis on the fusion noise field to obtain a noise influence analysis result; determining a noise cleaning strategy for the target E-business big data based on the noise influence analysis result.

The noise influence analysis result may represent the influence of noise on other data before and after cleaning, for example, for noise 1, if the normal use of other data is not influenced after cleaning, a forced cleaning policy may be provided for noise 1. For noise 2, which may affect normal use of other data after being flushed, a delay flushing policy may be provided for noise 2 (for example, flexible noise flushing may be performed during a use blank window of data associated with noise 2, including but not limited to some series of processes such as noise data compression buffering and subsequent noise data recovery). Therefore, the stability of noise cleaning can be improved, and interference on normal use of other data is avoided.

In some embodiments that may be independent, performing noise influence analysis on the fused noise field to obtain a noise influence analysis result, which may include the following steps: loading the fused noise field to a noise influence feature identification layer in a feature pyramid model, and obtaining a first noise influence description phrase and a second noise influence description phrase of the fused noise field, which are generated by the noise influence feature identification layer, wherein the noise influence feature identification layer comprises a plurality of sliding filtering units connected in series, the first noise influence description phrase is a noise influence description phrase (a feature vector for reflecting the influence of noise data on other data) generated by a tail sliding filtering unit in the plurality of sliding filtering units connected in series except for the tail sliding filtering unit (a convolution unit), and the second noise influence description phrase is a noise influence description phrase generated by a tail sliding filtering unit in the plurality of sliding filtering units connected in series; loading the second noise influence description phrase to an influence trend primary identification layer in the feature pyramid model to obtain an initial influence trend identification result generated by the influence trend primary identification layer; loading the first noise influence description phrase, the second noise influence description phrase, a third noise influence description phrase and the initial influence trend identification result to an influence trend depth identification layer in the feature pyramid model, and obtaining a noise influence analysis result generated by the influence trend depth identification layer, wherein the third noise influence description phrase is a noise influence description phrase generated by a sliding filtering unit in the influence trend initial identification layer according to a reference description phrase, and the reference description phrase is a description phrase obtained by adjusting the second noise influence description phrase.

By the design, the trend prediction precision of the initial trend influencing identification layer is different from that of the trend influencing depth identification layer, the trend influencing identification layer is high in prediction speed and relatively low in precision, the trend influencing depth identification layer is relatively low in prediction speed and relatively high in precision, and in the world, the timeliness and the accuracy of trend prediction can be ensured as far as possible through the combined processing of the initial trend influencing identification layer and the trend influencing depth identification layer, so that the subsequent noise cleaning quality is improved.

Based on the same or similar inventive concepts, please refer to fig. 2 in combination, and a schematic structural diagram of an e-commerce data screening system 30 based on big data is further provided, which includes an e-commerce data screening cloud platform 10 and an e-commerce interaction terminal 20 that communicate with each other, and the e-commerce data screening cloud platform 10 and the e-commerce interaction terminal 20 implement or partially implement the technical solutions described in the above method embodiments when running.

Further, a readable storage medium is provided, on which a program is stored, which when executed by a processor implements the method described above.

In the embodiments provided in the present invention, it should be understood that the disclosed apparatus and method can be implemented in other ways. The apparatus and method embodiments described above are illustrative only, as the flowcharts 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 that 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 may be stored in a computer-readable storage medium if they are implemented in the form of software functional modules and sold or used as separate products. Based on such understanding, the technical solution of the present invention or a part thereof which substantially contributes to the prior art may be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a media service server 10, 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 various media capable of storing program codes. It should be noted that, in this document, 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 phrases "comprising one of 8230; \8230;" 8230; "does not exclude the presence of additional like 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.

Claims

1. The E-commerce data screening method based on big data is applied to an E-commerce data screening cloud platform, and comprises the following steps:

determining a first E-business big data knowledge element relation network when a characteristic association variable between the E-business big data knowledge elements and a target E-business data noise field reaches a first set characteristic association variable limit value and a second E-business big data knowledge element relation network when the characteristic association variable between the E-business big data knowledge elements and a supervision data noise field reaches the first set characteristic association variable limit value through each group of E-business big data knowledge elements acquired from the shared service system;

wherein: the first E-business big data knowledge element relation network is obtained through not less than one group of first E-business big data knowledge elements in each group of E-business big data knowledge elements; the second E-commerce big data knowledge element relation network is obtained through not less than one group of second E-commerce big data knowledge elements in each group of E-commerce big data knowledge elements.

2. The method of claim 1, wherein prior to determining a first e-commerce big data knowledge element relationship network where a feature association variable between a target e-commerce data noise field reaches a first set feature association variable limit by each set of e-commerce big data knowledge elements obtained from the shared services system, further comprising:

dividing each group of the supervision data noise fields into a plurality of groups of local supervision data noise fields with the same number one by one, and determining a local supervision data noise field relation network corresponding to each group of the supervision data noise fields; wherein, each group of local supervision data noise fields in the local supervision data noise field relation network are updated according to the distribution priority of each group of local supervision data noise fields in the corresponding supervision data noise fields;

3. The method of claim 2, wherein determining not less than one set of E-commerce big data knowledge elements based on the local supervised data noise fields under the same distributed variable in each set of the local supervised data noise field relationship network, determining each set of E-commerce big data knowledge elements obtained from the shared service system, comprises:

and determining each group of E-commerce big data knowledge elements acquired from the shared service system through not less than one group of E-commerce big data knowledge elements respectively corresponding to each group of local E-commerce big data noise relationship network.

4. The method of claim 3, wherein on the premise that the local e-commerce big data noise relationship network and the local supervised data noise fields included in the local e-commerce big data noise relationship network have correlation with the distributed variables in the corresponding supervised data noise fields, determining a first e-commerce big data knowledge element relationship network in which the feature association variable between the target e-commerce data noise field and the feature association variable reaches a first set feature association variable limit by each group of e-commerce big data knowledge elements acquired from the shared service system, comprises:

5. The method of claim 1, wherein when each group of E-commerce big data knowledge elements acquired from the shared service system is annotated with knowledge element keywords, and the knowledge element keywords are used for independently reflecting each group of E-commerce big data knowledge elements, determining a first E-commerce big data knowledge element relation network corresponding to the target E-commerce data noise field comprises: and determining a first E-commerce big data knowledge element relation network corresponding to the target E-commerce data noise field through each group of knowledge element key words of the first E-commerce big data knowledge element.

6. The method according to claim 3, wherein, on the premise that the local power grid big data noise relationship network and the local supervision data noise fields included in the local power grid big data noise relationship network have correlation with the distribution variables in the corresponding supervision data noise fields, determining, through the sets of power grid big data knowledge elements acquired from the shared service system, sets of second power grid big data knowledge element relationship networks, each set of second power grid big data knowledge element relationship networks having characteristic correlation variables reaching first set characteristic correlation variable limits respectively with the sets of supervision data noise fields acquired from the shared service system, comprises:

when each group of local supervision data noise fields has a corresponding second e-commerce big data knowledge element, determining a second e-commerce big data knowledge element relation network obtained by each group of second e-commerce big data knowledge elements, and determining the second e-commerce big data knowledge element relation network corresponding to the supervision data noise fields when the feature association variable between the second e-commerce big data knowledge element relation network and the supervision data noise fields reaches a first set feature association variable limit value.

7. The method of claim 6, wherein obtaining the relationship network characteristic association variables between the first e-commerce big-data knowledge element relationship network and each set of the second e-commerce big-data knowledge element relationship networks on the basis that the number of first e-commerce big-data knowledge elements in the first e-commerce big-data knowledge element relationship network is consistent with the number of second e-commerce big-data knowledge elements in the second e-commerce big-data knowledge element relationship network respectively comprises:

8. The big data-based e-commerce data screening system is characterized by comprising an e-commerce data screening cloud platform and an e-commerce interaction terminal which are communicated with each other, and further comprising: determining a first E-business big data knowledge element relation network in which a characteristic association variable between the E-business big data knowledge elements and a target E-business data noise field reaches a first set characteristic association variable limit value and a second E-business big data knowledge element relation network in which a characteristic association variable between the E-business big data knowledge elements and a supervision data noise field reaches the first set characteristic association variable limit value through each group of E-business big data knowledge elements acquired from a shared service system; respectively obtaining a relation network characteristic association variable between the first E-commerce big data knowledge element relation network and each group of second E-commerce big data knowledge element relation networks, and determining a noise field association factor between the target E-commerce data noise field and each group of supervision data noise fields; wherein: the first E-commerce big data knowledge element relation network is obtained through not less than one group of first E-commerce big data knowledge elements in each group of E-commerce big data knowledge elements; the second E-commerce big data knowledge element relation network is obtained through not less than one group of second E-commerce big data knowledge elements in each group of E-commerce big data knowledge elements.

9. An e-commerce data screening cloud platform is characterized by comprising a processor and a storage; the processor is communicatively connected to the memory, and the processor is configured to read the computer program from the memory and execute the computer program to implement the method of any one of claims 1 to 7.

10. A readable storage medium, characterized in that a program is stored thereon, which program, when being executed by a processor, carries out the method of any one of claims 1-7.