CN114118610A - Product sales prediction method and system based on relevance big data - Google Patents

Product sales prediction method and system based on relevance big data Download PDF

Info

Publication number
CN114118610A
CN114118610A CN202111470377.XA CN202111470377A CN114118610A CN 114118610 A CN114118610 A CN 114118610A CN 202111470377 A CN202111470377 A CN 202111470377A CN 114118610 A CN114118610 A CN 114118610A
Authority
CN
China
Prior art keywords
relevance
big data
prediction
product sales
data
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202111470377.XA
Other languages
Chinese (zh)
Inventor
肖勇民
傅俊
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Fawang Cloud Logistics Technology Co ltd
Original Assignee
Shanghai Fawang Cloud Logistics Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shanghai Fawang Cloud Logistics Technology Co ltd filed Critical Shanghai Fawang Cloud Logistics Technology Co ltd
Priority to CN202111470377.XA priority Critical patent/CN114118610A/en
Publication of CN114118610A publication Critical patent/CN114118610A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G06F18/232—Non-hierarchical techniques
    • G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
    • G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0201—Market modelling; Market analysis; Collecting market data
    • G06Q30/0202—Market predictions or forecasting for commercial activities

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Strategic Management (AREA)
  • Human Resources & Organizations (AREA)
  • Development Economics (AREA)
  • Physics & Mathematics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Economics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Marketing (AREA)
  • Finance (AREA)
  • Accounting & Taxation (AREA)
  • General Business, Economics & Management (AREA)
  • Game Theory and Decision Science (AREA)
  • Operations Research (AREA)
  • Educational Administration (AREA)
  • Tourism & Hospitality (AREA)
  • Quality & Reliability (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Probability & Statistics with Applications (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • General Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention provides a product sales prediction method and system based on relevance big data. The method comprises the following steps: according to the big data online classification and estimation function based on the prediction behavior, online estimation and classification are carried out on the big data of the controllable relevance index influencing the prediction, and the big data mining direction is determined; finding out the association regularity among the controllable association indexes according to the association rule function so as to sort the available data sources; dividing data blocks by utilizing an online k-Means clustering and describing function with stronger clustering function to enable the similarity of the classified controllable relevance indexes to be as large as possible; product sales forecasts are performed. The product sales prediction method and the product sales prediction system based on the big relevance data can establish a quantifiable, dynamic and intelligent export product sales dynamic prediction model based on the big cross-border electronic commerce controllable relevance data under the drive of 'internet + foreign trade'.

Description

Product sales prediction method and system based on relevance big data
Technical Field
The invention relates to the technical field of big data, in particular to a product sales prediction method and system based on relevance big data.
Background
The existing popular foreign trade product sales prediction method simply researches and predicts problems from the perspective of a third-party platform or big data, and the consideration of the application of the integration of an internet platform, cross-border e-commerce and big data to the dynamic evolution prediction of product sales is insufficient. In order to improve the export product sales prediction effect and realize the flexibility and dynamic generalization of a prediction system, based on the research on the large data mining of the controllable relevance of the cross-border electric business export product sales, the personalized prediction mechanism and the intelligent prediction algorithm in the internet and foreign trade environment and the improvement on the corresponding algorithms such as distributed quantitative and centralized qualitative calculation, an export product sales dynamic prediction model based on the large controllable relevance data of the cross-border electric business under the internet and foreign trade drive is provided, and application experiments are carried out to contrastively analyze the experimental results of various models. Experimental results show that the model fully integrates the advantages of openness and extensibility of the internet plus and dynamic prediction of big data, and dynamic, intelligent and quantitative qualitative prediction of export product sales based on cross-border e-commerce controllable relevance big data in the internet and foreign trade environment is achieved. The comprehensive prediction effect of the model is obviously superior to that of the traditional model, and the model has stronger dynamic performance and higher practical value.
Disclosure of Invention
The invention aims to solve the technical problem of providing a product sales prediction method and system based on relevance big data, and can establish a quantifiable, dynamic and intelligent export product sales dynamic prediction model based on cross-border e-commerce controllable relevance big data under the drive of 'internet + foreign trade' from the aspects of cross-border e-commerce export product sales controllable relevance big data mining, personalized prediction mechanism, intelligent prediction algorithm and the like under the 'internet + foreign trade' environment.
In order to solve the technical problem, the invention provides a product sales prediction method and system based on relevance big data, wherein the method comprises the following steps: according to the big data online classification and estimation function based on the prediction behavior, online estimation and classification are carried out on the big data of the controllable relevance index influencing the prediction, and the big data mining direction is determined; finding out the association regularity among the controllable association indexes according to the association rule function so as to sort the available data sources; dividing data blocks by utilizing an online k-Means clustering and describing function with stronger clustering function to enable the similarity of the classified controllable relevance indexes to be as large as possible; the provided internet big data matching principle, semantic analysis and behavior analysis method reasonably matches and retains the highly-evolved characteristics of controllable relevance sales data aiming at main users, introduces the key controllable relevance big data influencing prediction in an online or offline mode, integrates the key controllable relevance big data into an internet big data warehouse, a cross-border e-commerce platform management background and an external application program interface, realizes the integration of key factors, and executes product sales prediction.
In some embodiments, the big data online classification and estimation function has the form:
Figure BDA0003391717790000021
wherein u and v are two kinds of products respectively, discourse area Iuv＝Iu∪IvIs a quantitative space, and data respectively excavates a discourse domain C and a non-empty subset A thereof, wherein S is { S ═ S }i＝(Si1,Si2,Si3)|i∈N+Is defined as a set of confidence measures, S ═ Si＝(Si1,Si2,Si3)|i∈N+Represents a collection of documents and logs on a third party platform in an "internet + foreign trade" environment, comprising n elements; SAM ═ SAMi|i∈N+And SAB ═ SABi|i∈N+Respectively representing a data fuzzy phenomenon and an uncertainty phenomenon set; classify the subject and object mapping function as F1＝(FSi→FCi) (ii) a If it is
Figure BDA0003391717790000033
In some embodiments, the association rule function has the form:
Figure BDA0003391717790000031
wherein, K1,K2Respectively a likelihood space, a certainty measure space; siConstraint time t in A setiXi is defined as K1A fuzzy variable of (d); current degree of association JiExpected value of correlation e (x) with uncertainty variable x, threshold ω0(ii) a Satisfy the mapping "G: K1+K2→K1·K2"the constrained time represents an optimal search time Ti(ii) a The correlation error is recorded as: e (S)i,Sj)＝b-∑xijE(x)e(Si,Sj)＝b-∑xijE(x)。
In some embodiments, the online k-Means clustering and describing function has the form:
Figure BDA0003391717790000032
wherein, the weight recommended by the predicted subject and the object is respectively omega1，ω2(ii) a The cluster recommendation set is TJ ═ tj1,tj2,…,tjnJ, expected value Q of associated descriptionx(tji,tjj)＝TS(SAMii,SABi)·e(Si,Sj) Entropy value Qn(tji,tjj)＝TS(SAMii,SABi)+e(Si,Sj) Entropy of Qe(tji,tjj)＝TS(SAMii,SABi)/e(Si,Sj)。
In some embodiments, the controllable relevance indicators include: product category, customer total demand, customer purchase psychology, product cycle, inventory, price, logistics, risk.
In addition, the invention also provides a product sales forecasting system based on the relevance big data, and the system comprises: one or more processors; storage means for storing one or more programs; when executed by the one or more processors, cause the one or more processors to implement a method for product sales prediction based on big relevance data as described above.
After adopting such design, the invention has at least the following advantages:
the DPMES prediction algorithm and model established by the big data online classification and estimation method based on the prediction behavior, the association rule function, the online k-Means clustering and description function, the personalized prediction mechanism, the distributed quantitative, centralized qualitative and parallel comprehensive prediction method are scientific and reasonable, some theoretical and practical problems are solved, the openness, the extensibility, the online and big data dynamic performances and the prediction advantages of the internet + are fully fused, the dynamic, intelligent, quantitative and qualitative predictions of export product sales based on the cross-border electricity business controllable association big data under the driving environment of the internet + foreign trade are realized, and the method has reference value for efficient marketing and efficient inventory planning of foreign trade enterprises.
Drawings
The foregoing is only an overview of the technical solutions of the present invention, and in order to make the technical solutions of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and the detailed description.
FIG. 1 is a cross-border e-commerce export product sales controllable relevance big data mining model under an "internet + foreign trade" environment;
FIG. 2 is a DPMES dynamic prediction model;
FIG. 3 is an intuitive scatter plot of improved algorithm validation results.
Detailed Description
The preferred embodiments of the present invention will be described in conjunction with the accompanying drawings, and it will be understood that they are described herein for the purpose of illustration and explanation and not limitation.
The cross-border e-commerce activity is complex in the environment of internet and foreign trade, the export product sales prediction is influenced by various factors such as demand, customs, logistics, risks and the like, but the core of the cross-border e-commerce is big data with prediction advantages, so that the export product sales prediction is relatively easy, and therefore, the design of an accurate, safe and efficient export product sales prediction model based on the cross-border e-commerce controllable relevance big data under the drive of the internet and the foreign trade becomes a popular subject of attention. At present, the research on product sales prediction is based on a big data classification method, a correlation grouping rule, an online clustering method, an internet big data matching principle, semantic analysis and behavior analysis methods to mine massive cross-border e-commerce data in a third-party platform and establish a product demand regression prediction model; establishing a product sales forecasting model by using the cross-border e-commerce historical sales data; a big data-based prediction model is provided according to historical data, popularity and new product customer value through a quantitative and qualitative analysis method. The research has a reference function, but the advantages of strong online big data prediction of the internet + are not shown due to the short proposing time of the internet + strategy, and the research does not provide a deeper and effective research for the prediction of how to fuse and apply the big data to dynamic wisdom of the product sales of cross-border e-commerce exports in the internet + foreign trade environment.
Aiming at the current research situation, from the aspects of mining large data with controllable relevance of cross-border e-commerce export product sales, personalized prediction mechanism, intelligent prediction algorithm and the like under the environment of 'internet + foreign trade', the invention tries to design a quantitative, dynamic and intelligent dynamic prediction model of export product sales based on the large data with controllable relevance of cross-border e-commerce under the drive of 'internet + foreign trade', emphasizes on realizing the dynamic prediction target of export product sales, so as to better guide the marketing and optimization inventory strategies of foreign trade enterprises, and simultaneously promotes the innovative research and application development of internet +, large data and cross-border e-commerce technology.
Export product big data mining based on cross-border e-commerce under the environment of 'internet + foreign trade' is firstly based on an 'internet + foreign trade' environment platform (the platform is drawn by the foreign trade economic cooperation department and is intensively deployed in the national internet center), and mining data such as policies, product types, total product demands of customers, trading groups, customer purchasing psychology, payment, quotations, customs duties, inventory, logistics, orders, contracts and reputation, commodity quality risks, return or change rates, counterfeit and shoddy products, false propaganda and the like. Key factors influencing the outlet product sales prediction and controllable relevance big data need to be mined. The excavation model is shown in fig. 1.
The specific mining process and method are as follows:
step 1, designing a big data online classification and estimation function based on the prediction behavior, performing online estimation and classification on the big data of the controllable relevance indexes influencing the prediction, and determining the mining direction of the big data. The existing big data classification related method is applied, but the fuzzy phenomenon of data is not clearly described, the thinking of the internet is not blended, the online classification function of the big data of the internet cannot be realized, the improvement is needed, and the big data online classification and estimation method based on the prediction behavior is provided.
Suppose that: given a data mining discourse domain C and its non-empty subset A, (C, A) { (Ci,Ai) I ∈ N + } is defined as a set of confidence measures, S ═ S [ + ]i＝(Si1,Si2,Si3) I belongs to N + } represents a set of documents and logs on a third-party platform under the environment of 'Internet + foreign trade', and comprises N elements; SAM ═ SAMiI ∈ N + } and SAB ═ SAB [ (SAB) ]iI belongs to N + } respectively represents a data fuzzy phenomenon and an uncertainty phenomenon set; classify the subject and object mapping function as F1＝(FSi→FCi). If Ci∈C∧(Ci,Ai)∧(Si1,Si2,Si3) Λ SAM ≠ False, then the large data online classification and estimation method based on predicted behavior can be represented by the function cf (i) as shown in equation (1):
Figure BDA0003391717790000061
improved CF (i) maps ambiguity and uncertainty to different values of (C)i,Ai) In (1). Compared with the method before improvement, the method can better characterize the ambiguity and uncertainty of the online classification of the internet big data.
And 2, designing an association rule function, finding out association regularity among the controllable association big data, and further sorting available data sources. A correlation grouping rule is provided, but the rule is only preliminary correlation on data, the correlation precision is low, and therefore improvement is needed, and a more accurate correlation rule function is provided.
On the basis of the formula (1), K is set1、K2Respectively a likelihood space, a certainty measure space; siConstraint time t in A setiXi is defined as K1A fuzzy variable of (d); current degree of association JiExpected value of correlation e (x) with uncertainty variable x, threshold ω0(ii) a Satisfy the mapping "G: K1+K2→K1·K2"the constraint time represents an optimal search time Ti; the correlation error is recorded as: e (S)i,Sj)＝b-∑xijE(x)e(Si,Sj)＝b-∑xijE (x), the improved association rule can be expressed by a function shown in equation (2):
Figure BDA0003391717790000071
the improved formula (2) has the function of accurately realizing the association regularity between large sales data of controllable association export products
And 3, designing an online k-Means clustering and describing function with stronger clustering function, enabling the similarity of the classified controllable relevance big data to be as large as possible, and dividing the data blocks. An online clustering method is provided, however, the data blocks divided by the method obviously have a boundary crossing phenomenon. Since the k-Means clustering method has the function of associated clustering, the method is improved by combining the two methods, and an online k-Means clustering and describing method is designed.
Based on the definition of formula (2), the weight recommended by the predicted subject and the object is respectively assumed to be omega1，ω2(ii) a Clustering recommendation set as TJ＝{tj1,tj2,…,tjnJ, expected value Q of associated descriptionx(tji,tjj)＝TS(SAMii,SABi)·e(Si,Sj) Entropy value Qn(tji,tjj)＝TS(SAMii,SABi)+e(Si,Sj) Entropy of excessQe(tji,tjj)＝TS(SAMii,SABi)/e(Si,Sj) Then, the improved online k-Means clustering and describing method can be dynamically represented by the function shown in formula (3):
Figure BDA0003391717790000072
the internet big data matching principle, the semantic analysis method and the behavior analysis method provided in the step 4 are used for reasonably matching and retaining the highly-evolved characteristics of the controllable relevance sales data aiming at the main users, importing the key controllable relevance big data influencing the prediction in an online or offline mode, integrating the key controllable relevance big data into an internet big data warehouse, a cross-border e-commerce platform management background and an external application program interface, realizing the integration of key factors, and better solving the problems of confusion of subject object attributes and inter-domain mapping of the semantic control matrix embodying the predicted behavior.
Through the personalized prediction mechanism and the intelligent prediction algorithm, a DPMES dynamic prediction model is constructed by using a decision tree, as shown in FIG. 2.
The construction path is as follows:
firstly, according to the step 1 of the intelligent prediction algorithm, a prediction target is determined, a data sample range used for modeling is selected, and a plurality of factors with prediction characteristics and capability are obtained through screening and filtering.
Secondly, according to the steps 2-3 of the intelligent prediction algorithm, key factors with controllable relevance are induced, consistency, controllable relevance and regularity of key feature data sequences are verified by using an incremental evolution-integration prediction mechanism and a random distribution-relevance prediction mechanism, incremental evolution and integration relations among key influence factors of the sales volume of various randomly distributed export products are quantized, and prediction components are dynamically added or deleted by using lambda. Key factors are integrated, data is prepared, evaluated, cleaned, non-linear transformed and verified.
Then, collaborative filtering recommendation is realized according to the algorithm step 4, and the prediction behavior subsets are divided.
And thirdly, tracking the controllable relevance big data stream in a centralized and real-time manner according to the formulas (10) to (12), synthesizing and aligning the search indexes of the internet big data in a staggered manner, selecting the key data with the maximum search index as a reference index, constructing a model, reconfiguring on line, giving the weight and the correlation coefficient of the parallel big data, predicting which potential customers are most likely to become consumers and traders in a centralized and qualitative manner, performing significance inspection on possible trading leads, predicting the sales volume of the next period in a distributed and quantitative manner, and predicting the future sales volume structure trend of the export product in a real-time manner.
And finally, evaluating and applying the model. And (4) carrying out error analysis on the prediction result of each prediction method, wherein if the average error of the prediction results in the previous stages is larger when the methods are integrated, the influence degree of the method on the integrated prediction result is smaller when the methods are integrated. And further dividing leaf nodes by the decision tree according to the weight and the correlation coefficient filtered by screening, obtaining a comprehensive predicted value of the sales volume after the model is stable, and accordingly optimizing the inventory strategy and realizing model application.
The test results are shown in table 1.
TABLE 1
Figure BDA0003391717790000091
As can be seen from fig. 3: the reconfiguration error scatter diagram, the characteristic point error scatter diagram and the mismatch rate scatter diagram are all in a monotonous and parallel increasing mode, and when the scatter diagram is in a smooth approach, the personalized prediction mechanism is closer to an actual prediction result; when the prediction efficiency of the advanced normalized fitting data indexes on the fluctuation of the sales volume of the outlet products is analyzed, the prediction efficiency by taking the improved algorithm as a guide method is the highest. This shows that the design of the improved method and the personalized prediction mechanism and the intelligent prediction algorithm based on the improved method is scientific and reasonable.
The invention also provides a product sales prediction system based on the relevance big data. For example, the product sales prediction system based on the relevance big data can be used for a product sales prediction host in a big data analysis system. As described herein, a product sales prediction system based on relevance big data can be used to implement a prediction function for product sales in a big data analytics system. The product sales volume prediction system based on the big relevance data can be implemented in a single node, or the functions of the product sales volume prediction system based on the big relevance data can be implemented in a plurality of nodes in the network. Those skilled in the art will appreciate that the term big data relevance based product sales prediction system includes devices in a broad sense, and that the big data relevance based product sales prediction system mentioned herein is only one example. The inclusion of a big-association-data-based product sales prediction system is for clarity and is not intended to limit the application of the present invention to a particular big-association-data-based product sales prediction system embodiment or to a class of big-association-data-based product sales prediction system embodiments. At least some of the features/methods described herein may be implemented in a network device or component, such as a product sales prediction system based on big relevance data. For example, the features/methods of the present invention may be implemented in hardware, firmware, and/or software running installed on hardware. The product sales prediction system based on the big relevance data can be any device for processing, storing and/or forwarding data frames through a network, such as a server, a client, a data source and the like. A processor may include one or more multi-core processors and/or memory devices, which may serve as data stores, buffers, and the like. The processor may be implemented as a general-purpose processor, or may be part of one or more Application Specific Integrated Circuits (ASICs) and/or Digital Signal Processors (DSPs).
The DPMES prediction algorithm and model established by the big data online classification and estimation method based on the prediction behavior, the association rule function, the online k-Means clustering and description function, the personalized prediction mechanism, the distributed quantitative, centralized qualitative and parallel comprehensive prediction method are scientific and reasonable, some theoretical and practical problems are solved, the openness, the extensibility, the online and big data dynamic performances and the prediction advantages of the internet + are fully fused, the dynamic, intelligent, quantitative and qualitative predictions of export product sales based on the cross-border electricity business controllable association big data under the driving environment of the internet + foreign trade are realized, and the method has reference value for efficient marketing and efficient inventory planning of foreign trade enterprises.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the present invention in any way, and it will be apparent to those skilled in the art that the above description of the present invention can be applied to various modifications, equivalent variations or modifications without departing from the spirit and scope of the present invention.

Claims (6)

1. A product sales prediction method based on relevance big data is characterized by comprising the following steps:
according to the big data online classification and estimation function based on the prediction behavior, online estimation and classification are carried out on the big data of the controllable relevance index influencing the prediction, and the big data mining direction is determined;
finding out the association regularity among the controllable association indexes according to the association rule function so as to sort the available data sources;
dividing data blocks by utilizing an online k-Means clustering and describing function with stronger clustering function to enable the similarity of the classified controllable relevance indexes to be as large as possible;
the provided internet big data matching principle, semantic analysis and behavior analysis method reasonably matches and retains the highly-evolved characteristics of controllable relevance sales data aiming at main users, introduces the key controllable relevance big data influencing prediction in an online or offline mode, integrates the key controllable relevance big data into an internet big data warehouse, a cross-border e-commerce platform management background and an external application program interface, realizes the integration of key factors, and executes product sales prediction.
2. The product sales prediction method based on the relevance big data as claimed in claim 1, wherein the big data online classification and estimation function has the form:
Figure FDA0003391717780000011
wherein u and v are two kinds of products respectively, discourse area Iuv＝Iu∪IvIs a quantitative space, and data respectively excavates a discourse domain C and a non-empty subset A thereof, wherein S is { S ═ S }i＝(Si1,Si2,Si3)|i∈N+Is defined as a set of confidence measures, S ═ Si＝(Si1,Si2,Si3)|i∈N+Represents a collection of documents and logs on a third party platform in an "internet + foreign trade" environment, comprising n elements; SAM ═ SAMi|i∈N+And SAB ═ SABi|i∈N+Respectively representing a data fuzzy phenomenon and an uncertainty phenomenon set; classify the subject and object mapping function as F1＝(FSi→FCi) (ii) a If it is
Figure FDA0003391717780000023
3. The product sales prediction method based on relevance big data according to claim 1, wherein the relevance rule function has the following form:
Figure FDA0003391717780000021
wherein, K1,K2Respectively a likelihood space, a certainty measure space; siConstraint time t in A setiXi is defined as K1A fuzzy variable of (d); current degree of association JiExpected value of correlation e (x) with uncertainty variable x, threshold ω0(ii) a Satisfy the mapping "G: K1+K2→K1·K2"the constrained time represents an optimal search time Ti(ii) a The correlation error is recorded as: e (S)i,Sj)＝b-∑xijE(x)e(Si,Sj)＝b-∑xijE(x)。
4. The product sales prediction method based on relevance big data according to claim 1, wherein the online k-Means clustering and describing function has the following form:
Figure FDA0003391717780000022
wherein, the weight recommended by the predicted subject and the object is respectively omega1，ω2(ii) a The cluster recommendation set is TJ ═ TJ { TJ1,tj2,...,tjnJ, expected value Q of associated descriptionx(tji,tjj)＝TS(SAMii,SABi)·e(Si,Sj) Entropy value Qn(tji,tjj)＝TS(SAMii,SABi)+e(Si,Sj) Entropy of Qe(tji,tjj)＝TS(SAMii,SABi)/e(Si,Sj)。
5. The method for predicting product sales based on big relevance data according to claim 1, wherein the controllable relevance index comprises: product category, customer total demand, customer purchase psychology, product cycle, inventory, price, logistics, risk.
6. A product sales prediction system based on relevance big data is characterized by comprising:
one or more processors;
storage means for storing one or more programs;
when executed by the one or more processors, cause the one or more processors to implement the method for predicting product sales based on big relevance data according to any one of claims 1 to 5.
CN202111470377.XA 2021-12-03 2021-12-03 Product sales prediction method and system based on relevance big data Pending CN114118610A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111470377.XA CN114118610A (en) 2021-12-03 2021-12-03 Product sales prediction method and system based on relevance big data

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111470377.XA CN114118610A (en) 2021-12-03 2021-12-03 Product sales prediction method and system based on relevance big data

Publications (1)

Publication Number Publication Date
CN114118610A true CN114118610A (en) 2022-03-01

Family

ID=80366747

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111470377.XA Pending CN114118610A (en) 2021-12-03 2021-12-03 Product sales prediction method and system based on relevance big data

Country Status (1)

Country Link
CN (1) CN114118610A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117350791A (en) * 2023-12-04 2024-01-05 北京六一六信息技术有限公司 Marketing campaign template customization method and system based on personalized popularization

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117350791A (en) * 2023-12-04 2024-01-05 北京六一六信息技术有限公司 Marketing campaign template customization method and system based on personalized popularization
CN117350791B (en) * 2023-12-04 2024-02-06 北京六一六信息技术有限公司 Marketing campaign template customization method and system based on personalized popularization

Similar Documents

Publication Publication Date Title
Cheriyan et al. Intelligent sales prediction using machine learning techniques
Dhote et al. Hybrid geometric sampling and AdaBoost based deep learning approach for data imbalance in E-commerce
Pan et al. Study on convolutional neural network and its application in data mining and sales forecasting for E-commerce
Cheng et al. An ontology-based business intelligence application in a financial knowledge management system
Omar et al. A hybrid neural network model for sales forecasting based on ARIMA and search popularity of article titles
Gharehchopogh et al. A linear regression approach to prediction of stock market trading volume: a case study
Saha et al. A survey of the application of graph-based approaches in stock market analysis and prediction
CN104321794B (en) A kind of system and method that the following commercial viability of an entity is determined using multidimensional grading
Chen et al. A topic-based sentiment analysis model to predict stock market price movement using Weibo mood
Le et al. A hybrid model for business process event and outcome prediction
Arun et al. Big data: review, classification and analysis survey
Hasheminejad et al. Data mining techniques for analyzing bank customers: A survey
Khalili-Damghani et al. Solving customer insurance coverage recommendation problem using a two-stage clustering-classification model
CN112070126A (en) Internet of things data mining method
CN114693409A (en) Product matching method, device, computer equipment, storage medium and program product
CN114612239A (en) Stock public opinion monitoring and wind control system based on algorithm, big data and artificial intelligence
Feng Data Analysis and Prediction Modeling Based on Deep Learning in E‐Commerce
Sarma et al. Stock market analysis with the usage of machine learning and deep learning algorithms
Rao et al. BMSP-ML: big mart sales prediction using different machine learning techniques
CN114118610A (en) Product sales prediction method and system based on relevance big data
Fakharchian Designing a forecasting assistant of the Bitcoin price based on deep learning using market sentiment analysis and multiple feature extraction
Chashmi et al. Predicting customer turnover using recursive neural networks
Zhang et al. Multi-view dynamic heterogeneous information network embedding
Sebt et al. Implementing a data mining solution approach to identify the valuable customers for facilitating electronic banking
Hung Using Cloud Services to Develop Marketing Information System Applications

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination