Search

CN-121981774-A - Method for realizing integral increment and big data consumption of Changbai mountain specialty membership

CN121981774ACN 121981774 ACN121981774 ACN 121981774ACN-121981774-A

Abstract

The invention relates to the technical field of member operation and big data application, and discloses a method for realizing the value added and big data consumption of a Changbai mountain specialty membership system, which lays a foundation through full-dimension acquisition of membership, supply chain, merchant and consumption scene data, encryption storage and labeling treatment; the invention adopts a logistic regression, ARIMA model and greedy algorithm to merge and drive integral layering increment, synchronously controls dynamic balance of a fund pool, relies on K-means clustering and collaborative filtering algorithm to match consumption demands, falls to the whole scene integral consumption and level differentiation single-free mechanism, and forms an operation closed loop through complex optimization of algorithms such as decision trees and the like.

Inventors

  • Tai Lizhu
  • Tai Shoujie

Assignees

  • 蛤灵参(吉林)生鲜连锁超市有限公司

Dates

Publication Date
20260505
Application Date
20260210

Claims (10)

  1. 1. The method for realizing the integral increment and big data consumption of the Changbai mountain specialty membership is characterized by comprising the following steps: The method comprises the steps of S1, full-dimension data acquisition, namely acquiring member basic information, consumption and point behaviors and social fission data after member authorization is acquired, synchronously abutting a supply chain and a merchant system to acquire special product quality, warehouse logistics and merchant operation data, acquiring online and offline consumption scene full-quantity data, and encrypting and storing; S2, calculating and giving out points, namely calculating basic points according to the amount of consumption and special products, calculating red scores and rewards points according to the yield proportion of merchants, and binding membership, merchants and product labels for each point; S3, adopting a logistic regression algorithm to identify member level matching differentiated value-added rules, regulating and controlling the value-added rhythm through an ARIMA model, superposing a greedy algorithm to realize fund pool linkage control, and starting an integral value-added flow of a preset period; s4, enabling consumption by big data, namely matching consumption requirements through K-means clustering and collaborative filtering algorithm, and avoiding a single mechanism of floor full scene integral consumption and level differentiation; And S5, data multiplexing and strategy optimization, namely adopting decision trees, genetic algorithms and multiple linear regression algorithms to multiplex the whole-process data, optimizing integration rules, value-added strategies and supply chain adaptation schemes, and forming an operation closed loop.
  2. 2. The method for realizing the value added and big data consumption of the Changbai mountain specialty membership system according to claim 1, wherein in the step S1, full-dimension data acquisition specifically comprises a member data acquisition sub-step, a supply chain and merchant data acquisition sub-step and a consumption scene data acquisition sub-step, wherein in the member data acquisition sub-step, acquired member basic information comprises member identity identification, region information and member grade association information, consumption and point behavior data comprises consumer preference, consumption frequency, point acquisition channel and point use record, social fission data comprises sharing propagation link, new guest invitation quantity and invitation conversion effect, member clear authorization is acquired through an online popup window and an offline written form before member data acquisition, authorized content clear data acquisition range, use purpose and storage life, encryption processing is carried out on member sensitive information through an AES encryption algorithm after acquisition, and an encryption key is stored in a distributed mode through a blockchain node.
  3. 3. The method for realizing the value added and big data consumption of the Changbai mountain specialty membership system is characterized in that in the step S1, the supply chain and merchant data acquisition substep specifically comprises the steps of abutting against a production management system of a Changbai mountain specialty production enterprise, acquiring specialty quality data such as specialty production place information, variety types, growth periods, quality detection reports and the like, abutting against a warehouse management system, acquiring warehouse data such as warehouse quantity, inventory margins, storage environment parameters and shelf life early warning information and the like of the specialty, abutting against a logistics distribution system, acquiring logistics data such as logistics number, distribution paths, transportation timeliness and signing states and the like, acquiring a commodity shelf list, sales running water, yield proportion, point exchange records and member drainage data of merchants in an abutting against a platform, synchronizing all acquired data to a data center in real time, and establishing a data time stamp to ensure data traceability.
  4. 4. The method for realizing the credit increment and big data consumption of the Changbai mountain specialty member system is characterized in that in the step S2, the credit accounting and issuing specifically comprises the steps of establishing a corresponding mapping relation between the consumption amount and the credit when basic credit accounting is carried out, setting differentiated credit accounting proportions for Changbai mountain core specialty products and non-core specialty products, wherein the core specialty products comprise ginseng, pilose antler, toad oil and lucid ganoderma, the non-core specialty products comprise Changbai mountain area derivative foods and wicresoft products, determining the total amount of credit based on merchant actual yield when the credit and rewarding credit accounting are carried out, carrying out credit allocation according to the consumption amount proportion of the consuming members and the yield proportion of the merchant, adopting a real-time account-entering mechanism for the credit issuing, carrying out credit automatic accounting within 3 seconds after online consumption, carrying out credit accounting within 10 seconds after online consumption is carried out through code scanning and checking, and sending a message APP message notification to the members when the credit accounting is carried out, wherein the notification content comprises the credit type, the credit amount, the effective period and the use range.
  5. 5. The method for realizing the credit value added and big data consumption of the Changbai mountain specialty membership system according to claim 1, wherein in the step S2, each credit binding label comprises a member unique identifier and a member grade label, each binding merchant label comprises a merchant unique identifier and a merchant type label, each binding product label comprises a product unique identifier, a product type label and a product place label, each credit binding credit source label and a credit state label are stored in a format by adopting a key value pair, data screening and statistical analysis are supported according to the dimension of the label, and data support is provided for matching the subsequent credit value added regulation and consumption requirements.
  6. 6. The method for realizing the point increment and big data consumption of the Changbai mountain specialty membership system according to claim 1, wherein in the step S3, the multi-algorithm fusion driving point increment specifically comprises the steps of taking collected membership consumption amount, accumulated consumption frequency, total point holding amount and social fission effect as input characteristics when a logistic regression algorithm identifies membership grades, optimizing model parameters through a gradient descent method, wherein a logistic regression model core formula is as follows: , Wherein the method comprises the steps of The probability is predicted for the membership grade, As a result of the bias term, For the weight parameters of each input feature, Is an input characteristic value; The Sigmoid activation function expression is: , Where z is the function input, corresponding to the linear combination term in the logistic regression model , As a result of the bias term, For the weight parameters of each input feature, For input eigenvalues, e is a natural constant (approximately equal to 2.71828), function output The value range is strictly between (0, 1); by minimizing the loss function: , Where m is the number of samples and where, Optimizing parameters, outputting member level classification results, wherein different member levels correspond to different upper value limits of integral increment and daily increment value amplitudes; When the ARIMA model regulates and controls the value-added rhythm, the model expression is as follows: , Wherein the method comprises the steps of In the form of an autoregressive polynomial, For the moving average polynomial, L is a hysteresis operator, d is a differential order, xt is the total amount data of the pool fund or the pool total amount at the time t, Is a white noise sequence; Taking collected historical fund running water data of the red branch pool and historical integral total data of the integral pool as time series samples, determining optimal parameter combinations of a p autoregressive order, a d differential order and a q moving average order through model training, predicting fund fluctuation trend and integral increment demand in a future preset period, dynamically adjusting daily integral increment amplitude, and ensuring that the integral increment total is matched with the fund stock of the red branch pool; when the greedy algorithm realizes fund pool linkage management and control, the fund balance and integral increment sustainability of the dividing pool are taken as optimization targets, and an optimization objective function is constructed: , wherein Mpool is the pool fund stock, ccost is the integral increment cost, And when the membership uses the integral deduction to consume as the weight coefficient, the processing fee with the preset proportion is preferentially returned to the red pool based on the objective function, and the residual funds are distributed to the merchant incentive account and the platform operation account according to the preset proportion.
  7. 7. The method for realizing the integral increment and big data consumption of the Changbai mountain specialty membership system according to claim 1, wherein in the step S3, the integral increment flow of a preset period further comprises an integral increment monitoring sub-step of recording integral increment full-flow data comprising daily increment amplitude, integral increment total amount and money change condition of a branch pool through a blockchain technology to form an untampereable integral increment account, setting an integral increment abnormal early warning threshold, automatically triggering an early warning mechanism by a system when the daily increment amplitude deviates from a preset range + -5% or the money of the branch pool is lower than an early warning line, pushing an early warning message to an operation management end, suspending an integral increment adjustment flow, and recovering increment after completion of checking treatment by operating personnel to ensure that the integral increment process is stable and controllable.
  8. 8. The method for realizing the credit increment and big data consumption of the Changbai mountain specialty membership system is characterized in that in the step S4, when the big data enabling consumption specifically comprises the step of matching consumption demands by a K-means clustering algorithm, dividing the acquired membership consumption class preference, consumption amount interval and credit use habit into a plurality of subdivision groups such as daily life consumption groups, holiday gift consumption groups, high-end nourishing consumption groups and random consumption groups by taking the acquired membership consumption class preference, consumption amount interval and credit use habit as clustering characteristics, the collaborative filtering algorithm is based on the common consumption characteristics of each subdivision group, and combines the historical consumption records of the similar members, the matched credit exchangeable commodity list and consumption activity information are pushed for the members of different groups, and when the full scene credit consumption falls to the ground, the consumption scenes of online self-service malls, brand malls, offline chain supermarkets and heterogeneous union business are integrated, a unified credit payment interface is built, and two modes of independent payment of the credit and cash combination payment are supported, and real-time check of credit validity and deduction in the payment process are achieved.
  9. 9. The method for realizing the value added and big data consumption of the Changbai mountain specialty membership system is characterized in that in the step S4, a grade differentiation free mechanism specifically comprises the steps of setting differentiated free participation qualification, free probability and free monetary upper limit for different grade members based on the identified membership grade, obtaining free qualification to meet preset conditions, wherein the free qualification comprises the fact that the integral holding amount reaches a preset threshold, a consumption record exists in a near preset period and no integral illegal use record exists, the free screening adopts a weighted random algorithm, the membership grade weight, the integral holding amount weight and the consumption activity weight are used as core influencing factors, calculating the free winning probability of the membership, randomly extracting winning members according to probability, and after the winning members consume in a specified consumption scene, applying for free reimbursement by virtue of an integral deduction certificate, and transferring reimbursement funds from a red pool according to a preset flow.
  10. 10. The method for realizing the integral increment and big data consumption of the Changbai mountain specialty membership system according to claim 1, wherein in the step S5, the data duplication and strategy optimization specifically comprises the steps of constructing a decision tree model by adopting an ID3 algorithm when the decision tree algorithm is used for duplicating full-flow data, wherein the core is the optimal splitting characteristic selected based on the information gain, and the information gain formula is as follows: , S is a sample set, A is a candidate split feature, sv is a sample subset of a v value of the feature A in S, and Entropy (S) is information Entropy of the sample set S; The method comprises the steps of taking the integral utilization rate, the member activity, the trade yield enthusiasm and the supply chain matching degree as core evaluation indexes, analyzing the association relation between each index and operation success through a model, positioning operation pain points such as unreasonable integral accounting proportion, uncontrolled increment rhythm, insufficient consumption scene adaptation and the like, when a genetic algorithm optimizes an integral rule, taking the improvement of integral utilization rate and the activity of a staff as targets, binary coding integral accounting proportion, increment period and handling fee proportion rule parameters to form a chromosome, selecting a father individual through a roulette selection method, performing cross operation through a single-point crossover operator, performing mutation operation through a basic position mutation operator, and performing multi-generation iterative screening optimal parameter combination to form a rule optimization scheme, and when a multi-element linear regression algorithm optimizes the supply chain adaptation scheme, taking the integral exchange rate and the repurchase rate of special products as dependent variables, taking the product specification, supply aging and quality grade as independent variables, establishing a regression model, outputting specific suggestions of product specification adjustment, capacity optimization and supply cycle adaptation to Changbaishan special products, and realizing accurate closed loop operation of the supply chain and consumption requirements.

Description

Method for realizing integral increment and big data consumption of Changbai mountain specialty membership Technical Field The invention relates to the technical field of member operation and big data application, in particular to a method for realizing the value added and big data consumption of a Changbai mountain specialty membership system. Background The special feature of Changbai mountain takes an important role in the market by virtue of unique geographic environment and quality advantages, members are manufactured into key modes for improving user viscosity and promoting special feature popularization, but the traditional special feature of Changbai mountain is widely applied to the industry, but has obvious defects in that the traditional special feature of Changbai mountain is not provided with full-flow data driving and multidimensional value mining capabilities, only points are taken as simple consumption deduction tools, no sustainable value-added mechanism exists, the holding willingness of the points of the members is low, the utilization rate is insufficient, the repurchase behavior is difficult to be effectively excited, meanwhile, the traditional mode does not integrate member consumption preference, supply chain dynamics and merchant operation data, manual recommendation or broad spectrum marketing is relied, so that the special feature supply and demand adaptation is unbalanced, the merchant marketing cost is increased, the member consumption experience is reduced, and a closed loop optimization mechanism is not provided, rules and strategies cannot be timely adjusted according to the operation data, and finally the member loss rate is high, and the large-scale popularization and marketing operation requirements of the special feature of Changbai mountain are difficult to support. Disclosure of Invention The invention aims to provide an integral increment and big data consumption implementation method of a Changbai mountain specialty membership system, so as to solve the problems in the background technology. In order to achieve the purpose, the invention provides the following technical scheme that the method for realizing the integral increment and big data consumption of the Changbai mountain specialty membership comprises the following steps: The method comprises the steps of S1, full-dimension data acquisition, namely acquiring member basic information, consumption and point behaviors and social fission data after member authorization is acquired, synchronously abutting a supply chain and a merchant system to acquire special product quality, warehouse logistics and merchant operation data, acquiring online and offline consumption scene full-quantity data, and encrypting and storing; S2, calculating and giving out points, namely calculating basic points according to the amount of consumption and special products, calculating red scores and rewards points according to the yield proportion of merchants, and binding membership, merchants and product labels for each point; S3, adopting a logistic regression algorithm to identify member level matching differentiated value-added rules, regulating and controlling the value-added rhythm through an ARIMA model, superposing a greedy algorithm to realize fund pool linkage control, and starting an integral value-added flow of a preset period; s4, enabling consumption by big data, namely matching consumption requirements through K-means clustering and collaborative filtering algorithm, and avoiding a single mechanism of floor full scene integral consumption and level differentiation; And S5, data multiplexing and strategy optimization, namely adopting decision trees, genetic algorithms and multiple linear regression algorithms to multiplex the whole-process data, optimizing integration rules, value-added strategies and supply chain adaptation schemes, and forming an operation closed loop. Preferably, in the step S1, the full-dimension data acquisition specifically includes a member data acquisition sub-step, a supply chain and merchant data acquisition sub-step and a consumption scene data acquisition sub-step, wherein in the member data acquisition sub-step, acquired member basic information includes member identification, region information and member class association information, consumption and point behavior data includes consumer preference, consumption frequency, point acquisition channel and point use record, social fission data includes sharing propagation link, new guest invitation number and invitation conversion effect, member clear authorization is acquired through an online popup window and offline form before member data acquisition, authorization content defines a data acquisition range, a use purpose and a storage period, member sensitive information is encrypted by adopting an AES encryption algorithm after acquisition, and an encryption key is stored in a distributed manner through a block chain node. Preferably, in the step