CN115809891A - Multi-product combined demand prediction method and device based on substitution and correlation effects - Google Patents
Multi-product combined demand prediction method and device based on substitution and correlation effects Download PDFInfo
- Publication number
- CN115809891A CN115809891A CN202211655555.0A CN202211655555A CN115809891A CN 115809891 A CN115809891 A CN 115809891A CN 202211655555 A CN202211655555 A CN 202211655555A CN 115809891 A CN115809891 A CN 115809891A
- Authority
- CN
- China
- Prior art keywords
- commodity
- attribute
- substitution
- association
- combined
- 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
Links
- 238000006467 substitution reaction Methods 0.000 title claims abstract description 101
- 238000000034 method Methods 0.000 title claims abstract description 43
- 230000000694 effects Effects 0.000 title claims abstract description 30
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 32
- 238000013277 forecasting method Methods 0.000 claims abstract description 22
- 238000013439 planning Methods 0.000 claims abstract description 19
- 238000007476 Maximum Likelihood Methods 0.000 claims abstract description 17
- 230000009471 action Effects 0.000 claims abstract description 7
- 238000005314 correlation function Methods 0.000 claims abstract description 6
- 238000005065 mining Methods 0.000 claims description 16
- 230000006870 function Effects 0.000 claims description 15
- 230000008569 process Effects 0.000 claims description 6
- 238000007637 random forest analysis Methods 0.000 claims description 5
- 238000012549 training Methods 0.000 claims description 5
- 238000004458 analytical method Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 4
- 230000000737 periodic effect Effects 0.000 claims description 3
- 230000002349 favourable effect Effects 0.000 claims description 2
- 230000006399 behavior Effects 0.000 abstract description 17
- 238000010801 machine learning Methods 0.000 abstract description 5
- 238000007405 data analysis Methods 0.000 abstract 1
- 230000035772 mutation Effects 0.000 description 4
- 239000000654 additive Substances 0.000 description 3
- 230000000996 additive effect Effects 0.000 description 3
- 238000003066 decision tree Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000004590 computer program Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 239000008370 chocolate flavor Substances 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 238000009396 hybridization Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000010845 search algorithm Methods 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
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
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/047—Probabilistic or stochastic networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Biophysics (AREA)
- Health & Medical Sciences (AREA)
- Business, Economics & Management (AREA)
- Life Sciences & Earth Sciences (AREA)
- Data Mining & Analysis (AREA)
- Strategic Management (AREA)
- Finance (AREA)
- Development Economics (AREA)
- General Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- Biomedical Technology (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Biology (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Entrepreneurship & Innovation (AREA)
- Computational Linguistics (AREA)
- Evolutionary Computation (AREA)
- Genetics & Genomics (AREA)
- Physiology (AREA)
- General Business, Economics & Management (AREA)
- Game Theory and Decision Science (AREA)
- Economics (AREA)
- Marketing (AREA)
- Probability & Statistics with Applications (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a multi-product combined demand prediction method and a multi-product combined demand prediction device based on substitution and correlation effects, wherein the prediction method comprises the following steps: determining the core attribute of the commodity based on the sales information of the current commodity on sale; defining substitution and correlation functions among various attribute combined commodities in the purchasing behavior of the consumer; and (3) parameters for measuring substitution and correlation actions are incorporated into an operation planning model established based on a maximum likelihood thought, and a meta-heuristic algorithm is used for solving the actual demand and demand proportion of each attribute combined commodity including the existing commodity and the new product. The forecasting method can be suitable for actual demand forecasting of multiple categories of commodities in retail stores, jointly enables demand forecasting based on data analysis capability of machine learning and planning capability of operational research, determines actual demand for commodities on the basis of considering substitution and correlation effects in consumer purchasing behaviors, and expands a demand forecasting model generally used for a single category to be suitable for two/multiple associated categories.
Description
Technical Field
The invention relates to the technical field of data, in particular to a multi-product combined demand prediction method and device based on substitution and correlation effects.
Background
The commodity demand forecast includes a forecast of an existing commodity and a forecast of a new commodity. The sales performance prediction of the new commodity has a key influence on the successful promotion of the commodity, and can help companies to develop markets and improve the accuracy and real-time performance of the new commodity demand prediction, namely, the new commodity introduction decision can be supported; secondly, the method also has important help to solve the challenges brought by fast iterative commodity combination to the class management and the like in the new retail background, and the obtained actual demand proportion of each attribute combination commodity including the new commodity and the existing commodity is used as the reference for the class quantity configuration.
For the former method, regarding new product sales prediction, the machine learning algorithm is a common method by combining the idea of sample migration learning, but the problem is that the number of features which can be extracted by a new product is small, so that the model training effect and the prediction accuracy are influenced; for the latter, regarding the management problem of categories, because there are relevance and substitution between the commodities, the sales share of some commodities in current stores cannot represent the real demand of the consumer, so that it cannot be directly used as a reference for quantity configuration, and in the field of planning operation, the measurement of substitution is only considered in the usually adopted commodity actual demand prediction model, and the promotion effect brought by the relevance to the commodity demand is rarely considered.
In view of the above, the present invention is especially proposed.
Disclosure of Invention
In order to solve the technical problems, the invention provides a multi-product combined demand forecasting method and device based on substitution and correlation effects.
Specifically, the following technical scheme is adopted:
the multi-category combined demand forecasting method based on substitution and correlation effects comprises the following steps:
determining the core attribute of the commodity based on the sales information of the current commodity on sale, and converting the SKU-level commodity into an attribute-level commodity;
defining substitution and correlation functions among various attribute combined commodities in the purchasing behavior of the consumer;
and (3) parameters for measuring substitution and correlation actions are incorporated into an operation planning model established based on a maximum likelihood thought, and a meta-heuristic algorithm is used for solving the actual demand and demand proportion of each attribute combined commodity including the existing commodity and the new product.
As an optional embodiment of the present invention, in the method for predicting a multi-item combined demand based on substitution and association, the determining a core attribute of a commodity based on sales information of a currently sold commodity, and converting a SKU-level commodity into an attribute-level commodity includes:
extracting attribute features of a current commodity on sale, wherein the attribute features comprise a commodity specification category, a price category, a brand grade category, a brand origin category and a function category;
and taking the corresponding periodic sales volume of the current commodity on sale as a label, training a random forest regression model, determining the core attribute of the commodity through the marginal contribution degree sequencing of the features on the basis of an interpretable SHAP model, and converting the SKU-level commodity into an attribute-level commodity.
As an alternative embodiment of the present invention, in the method for predicting multi-item combined demand based on substitution and association actions, the defining substitution actions among the combined items of each attribute in the purchasing behavior of the consumer includes:
based on the fact that the substitution effect among the commodities of the same class is measured by different attribute combinations of the current commodities on sale, the idea of joint analysis is used for reference, and the substitution probability among the commodities of the corresponding attribute combination is determined to be the product of the substitution probabilities among the attributes.
As an alternative embodiment of the present invention, in the multi-category demand forecasting method based on substitution and correlation, the calculation process of the substitution probability includes:
the commodity set corresponding to a certain category is S, the favorite commodity of the user is i, and when i is not in the category set S, the user may use the favorite commodity to be present in the category set S Article for children Resemblance quotient of i in class set S Article (A) j is substituted with a substitution probability pi i j (ii) a Then the level is j on the a attributes respectively 1 ，j 2 Said, j-merchandise of jA replaces l-merchandise of i1, i2, i, A, respectively, on A attributes with a probability of
As an optional embodiment of the present invention, in the method for predicting a multi-item combined demand based on substitution and association, the defining an association between each attribute combination item in the consumer purchasing behavior includes:
based on A p Determining an association category and an association probability according to an association rule obtained by an r i or i algorithm, and assuming the association probability as a confidence result corresponding to the association rule mining;
assuming that the first category corresponds to the commodities of the m attribute combinations determined by association rule mining, the second category corresponds to the commodities of the n attribute combinations, mn estimated values of association probabilities are obtained according to the records of tickets purchased by users, and the commodity names in the ticket data are changed into attribute level names during association rule mining, so that the corresponding association probabilities among the commodities of all attribute combinations across categories are obtained.
As an alternative embodiment of the present invention, the method for predicting the multi-item combined demand based on the substitution and association effects of the present invention includes:
the attribute combination commodity set of a pair of associated categories is S and Z respectively, the commodity p of each attribute combination in the set Z has an associated action on the commodity j of each attribute combination in the set S, and the association probability is w jp ， Is not limited to Merchandise with combined attributes in set Z q The association probability for each product j of the attribute combinations in the set S is w jp π qp 。
As an optional embodiment of the present invention, in the method for predicting a multi-item combined demand based on substitution and association, the step of incorporating parameters for measuring substitution and association into an operation planning model established based on a maximum likelihood idea, and the step of solving an actual demand and a demand proportion of each attribute combined commodity including an existing commodity and a new commodity by using a meta-heuristic algorithm includes:
the core attribute parameters, the association probability, the substitution probability and the probability of the most favorable attribute combined commodity are brought into an operation planning model established based on a maximum likelihood thought:
the attribute combination commodity set of a pair of related item classes is S and Z respectively, and the probability F of purchasing the attribute combination commodity j in the set S j (S) is
Probability F of purchasing property combination commodity p in set Z p (Z) similarly, the maximum likelihood estimation has the objective function of
Wherein f is j Means probability, y, of the favorite commodity j P 、x j Store sales of goods p, j, respectively;
and solving the actual demand quantity of each attribute combined commodity including the existing commodity and the new commodity by using a meta-heuristic algorithm.
As an alternative embodiment of the present invention, the method for predicting the multi-item combined demand based on the substitution and association effects of the present invention includes:
and determining the association probability among the associated categories based on confidence results obtained by mining association rules by an Apriori algorithm, and establishing the association among a plurality of categories, thereby establishing a multi-category combined demand prediction model, wherein an objective function is the product of powers of purchase probabilities of the categories, and the power items are the store sales volume of the combined commodity with the attributes corresponding to the categories respectively.
The invention also provides a multi-product combined demand prediction device based on substitution and correlation effects, which comprises the following components:
the commodity attribute module is used for determining the core attribute of the commodity based on the sales information of the current commodity on sale and converting the SKU-level commodity into an attribute-level commodity;
the substitution and association module is used for defining substitution and association functions among various attribute combination commodities in the purchasing behavior of the consumer;
and the commodity demand prediction module is used for bringing parameters for measuring substitution and correlation into an operation planning model established based on a maximum likelihood thought, and solving the actual demand and demand proportion of each attribute combined commodity including the existing commodity and the new product by using a meta-heuristic algorithm.
The invention also provides a computer storage medium which stores a computer executable program, and is characterized in that when the computer executable program is executed, the method for predicting the multi-category united demand based on substitution and association effects is realized.
Compared with the prior art, the invention has the beneficial effects that:
the multi-category combined demand forecasting method based on substitution and association is suitable for forecasting actual demand (free from influence of substitution and association) and forecasting demand of new products of existing commodities of retail stores, and the substitution and association in purchasing behaviors of consumers are considered in the multi-category combined demand forecasting method based on substitution and association, and the substitution and association are respectively as follows: selecting a substitute when the current item set does not have the favorite items of the user; for the associated commodities existing in the current store item set at the same time, the user has a certain probability to select association; for the associated commodities which do not exist in the current store item class set simultaneously, the promotion effect of one party not in the set on the initial demand of the other party is represented by the product of the substitution probability and the association probability, namely the part of the association promotion effect transferred to the substitute commodities. Therefore, the multi-product combined demand forecasting method based on substitution and correlation effects can obtain a more accurate commodity actual demand forecasting result by more completely defining the purchasing behavior of the consumers in the model.
The invention provides a multi-category combined demand forecasting method based on substitution and association functions, wherein a demand forecasting model is expanded to be suitable for the situation of two or more associated categories, association probability estimated values obtained through association rule mining are used for establishing the association among all the associated categories, the association is considered more completely, and the accuracy of commodity actual demand forecasting is improved.
Therefore, the multi-category combined demand forecasting method based on the substitution and association effects uses historical sales data of combined commodities with different attributes of various categories of retail stores, and achieves multi-category combined actual demand forecasting by defining the purchasing behaviors of consumers including the substitution and association effects and establishing a demand forecasting model suitable for multiple associated categories. Meanwhile, the demand proportion determined according to the actual demand of the existing commodities and the new commodities can be used for the class planning guidance of retail stores with different scales.
Description of the drawings:
FIG. 1 is a flow chart of a multi-category demand forecasting method based on substitution and correlation disclosed by the embodiment of the invention;
FIG. 2 is a flowchart of the establishment of a multi-item combined demand forecasting model based on substitution and correlation effects, which is disclosed by the embodiment of the invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be described clearly and completely with reference to the accompanying drawings. It is to be understood that the embodiments described are only a few embodiments of the present invention, and not all embodiments.
Thus, the following detailed description of the embodiments of the invention is not intended to limit the scope of the invention as claimed, but is merely representative of some embodiments of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that the embodiments of the present invention and the features and technical solutions thereof may be combined with each other without conflict.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures.
In the description of the present invention, it should be noted that the terms "upper", "lower", and the like refer to orientations or positional relationships based on those shown in the drawings, or orientations or positional relationships that are conventionally arranged when the products of the present invention are used, or orientations or positional relationships that are conventionally understood by those skilled in the art, and such terms are used for convenience of description and simplification of the description, and do not refer to or imply that the devices or elements referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus, should not be construed as limiting the present invention. Furthermore, the terms "first," "second," and the like are used merely to distinguish one description from another, and are not to be construed as indicating or implying relative importance.
Referring to fig. 1, the method for predicting the multi-item combined demand based on substitution and association of the embodiment includes:
screening and determining the core attributes of the commodities: determining the core attribute of the commodity based on the sales information of the current commodity on sale, and converting the SKU-level commodity into an attribute-level commodity;
defining substitution and correlation functions among various attribute combined commodities in the purchasing behavior of the consumer;
and (3) parameters for measuring substitution and correlation actions are incorporated into an operation planning model established based on a maximum likelihood thought, and a meta-heuristic algorithm is used for solving the actual demand and demand proportion of each attribute combined commodity including the existing commodity and the new product.
The multi-category combined demand forecasting method based on substitution and association is suitable for forecasting actual demands (not influenced by substitution and association) and forecasting new demands of existing commodities of a retail store, historical sales data of commodities are combined by using different attributes of various categories of the retail store, and multi-category combined actual demand forecasting is achieved by defining consumer purchasing behaviors including substitution and association and building a demand forecasting model suitable for multiple associated categories. Meanwhile, the demand proportion determined according to the actual demand of the existing commodities and the new commodities can be used for the class planning guidance of retail stores with different scales.
According to the multi-product combined demand forecasting method based on substitution and association, when the core attributes of the commodities are determined, the SHAP model is used for performing interpretable analysis on the random forest regression model, when the association between cross-product commodities is defined, the Apriori algorithm is used, and when the model is solved, the differential evolution algorithm is used.
The random forest regression model is composed of a plurality of regression trees, each decision tree in the forest is not related, and the final output of the model is jointly determined by each decision tree in the forest. The model introduces randomness in the training process of the decision tree, so that the model has excellent anti-overfitting and anti-noise capabilities.
The SHAP model is an interpretable machine learning model, the output of any machine learning model can be interpreted, the name of the SHAPLY additive interpretation model is derived from Shapley additive interpretation, an additive interpretation model is constructed by the SHAP under the inspiration of cooperative game theory, all characteristics are regarded as contributors, for each prediction sample, the model generates a prediction value, and the SHAP value is a numerical value allocated to each characteristic in the sample and can reflect the influence of the characteristic in each sample.
The Apriori algorithm is an association rule mining algorithm, and decomposes an association rule mining task into two main subtasks: subtask 1 is to generate a frequent item set, and subtask 2 is to generate a strong association rule. The support degree and the confidence degree are respectively adopted to quantify the frequent item set and the association rule, and the association function is defined by the embodiment by using an Apriori algorithm to mine the association categories with higher confidence degrees among the association categories in the strong association rule.
The differential evolution algorithm is a heuristic search algorithm based on a group, the evolution process comprises mutation, hybridization and selection operations, the mutation vector of the differential evolution algorithm is generated by a parent differential vector, is crossed with the parent individual vector to generate a new individual vector, and is directly selected with the parent individual, so that the risk of falling into a local optimal solution can be avoided at a higher probability in the mutation operation compared with the random mutation operation of the genetic algorithm, namely, the new solutions for dissociating the original solutions have certain distances through scaling, the exploration capability is ensured, and the demand prediction model is solved by using the differential evolution algorithm, thereby being beneficial to obtaining a better prediction result.
Referring to fig. 2, in the multi-item-class combined demand forecasting method based on substitution and association, determining the core attribute of the commodity based on the sales information of the currently sold commodity, and converting the SKU-level commodity into the attribute-level commodity includes:
extracting attribute features of the current commodity on sale, including a commodity specification category (large, medium and small), and/or a price category (high, medium and low), and/or a brand grade category (whether a brand network is on-list or not), and/or a brand origin category (whether the brand is local commodity), and/or a function category;
and taking the corresponding periodic sales volume of the current commodity on sale as a label, training a random forest regression model, determining the core attribute of the commodity through the marginal contribution degree sequence of the features based on an interpretable SHAP model, and converting the SKU (stock keeping unit) level commodity into an attribute level commodity. For example, selecting the core attribute as a brand rating attribute, a specification category, and a price category, then "a piglet petty dip cup chocolate flavor of 0.12 dollars per g unit price" may be named "not in a leaderboard" according to the specification and price distribution of the biscuit-like product.
In the embodiment, the SKU-level commodities are converted into the attribute-level commodities, and the substitution probability among the attribute combination commodities of the same category and the association probability among the attribute combination commodities of the cross-category are determined on the basis of the attribute-level commodities.
Further, in the method for predicting the multi-item combined demand based on substitution and correlation in this embodiment, the defining the substitution effect among the combined items of each attribute in the purchasing behavior of the consumer includes:
based on the fact that substitution effects among commodities of the same class are measured by different attribute combinations of the current commodities on sale, and the idea of joint analysis is used for reference, the substitution probability among the commodities of the corresponding attribute combination is determined to be the product of the substitution probabilities among all the attributes.
Specifically, in the multi-category demand forecasting method based on substitution and correlation, the calculation process of the substitution probability includes:
the commodity set corresponding to a certain category is S, the favorite commodity of the user is i, when i is not in the category set S, the user may substitute similar commodities j of i in the category set S, and the substitution probability is pi ij (ii) a Then the level is j on the A attributes 1 Day of the year 2 ，...，j A The j commodity substitutes are respectively i on the A attributes 1 ，i 2 ，...，i A I goods of (1) has a probability of
As an optional implementation manner of this embodiment, in the method for predicting a multi-item combined demand based on substitution and association, the defining an association between each attribute combined commodity in a purchasing behavior of a consumer includes:
determining an association class and an association probability based on an association rule obtained by an Apriori algorithm, and assuming the association probability as a confidence result corresponding to the association rule mining;
assuming that the first category determined by association rule mining corresponds to the commodities of the m attribute combinations, the second category corresponds to the commodities of the n attribute combinations, mn estimation values of association probabilities are obtained according to the records of tickets purchased by users, and the commodity names in the ticket data are changed into attribute level names during association rule mining, so that the corresponding association probabilities among the commodities of each attribute combination of the cross-category are obtained.
Specifically, in the multi-item combined demand prediction method based on substitution and association, the calculation method of the association probability includes:
the attribute combination commodity sets of a pair of related categories are S and Z respectively, the commodity p of each attribute combination in the set Z has a correlation function on the commodity j of each attribute combination in the set S, and the correlation probability is w jp The association probability of a product q not having each attribute combination in the set Z to a product j having each attribute combination in the set S is w jp π qp 。
In the method for predicting the multi-product combined demand based on the substitution and association, the parameter for measuring the substitution and association is incorporated into an operation planning model established based on a maximum likelihood thought, and the method for solving the actual demand and demand proportion of each attribute combined commodity including the existing commodity and the new commodity by using a meta-heuristic algorithm includes:
and (3) bringing the core attribute parameters, the association probability, the substitution probability and the probability of the favorite certain attribute combination commodity into an operation planning model established based on the maximum likelihood thought. Assume that past sales have included consideration of substitution and correlation, i.e., the probability of a user purchasing any item in the current set of categories can be divided into three parts: the data samples taken over a period of time, for example, are considered to reflect the overall situation with the greatest probability (Max problem) due to the purchase probability increased by substitution, the purchase probability facilitated by association, and the purchase probability unaffected by the substitution and association.
The attribute combined commodity set of a pair of related categories is S and Z respectively, and the probability F of purchasing the attribute combined commodity j in the set S j (S) is
Probability F of purchasing property combination commodity p in set Z p (Z) similarly, the maximum likelihood estimation has the objective function of
Wherein f is j Means probability, y, of the favorite commodity j P 、x j Respectively, store sales for items p, j.
And solving the actual demand quantity of each attribute combined commodity including the existing commodity and the new commodity by using a meta-heuristic algorithm.
Preferably, solving the actual demand of each attribute combination commodity including the existing commodity and the new commodity by using a differential evolution algorithm;
optionally, the algorithm parameters to be searched are population size (NP), scaling factor (F), and crossover probability (CR).
The method for predicting the multi-product combined demand based on the substitution and association functions comprises the following steps:
and determining the association probability among the associated categories based on confidence results obtained by mining association rules by an Apriori algorithm, and establishing the association among a plurality of categories, thereby establishing a multi-category combined demand prediction model, wherein an objective function is the product of powers of purchase probabilities of the categories, and the power items are the store sales volume of the combined commodity with the attributes corresponding to the categories respectively.
The embodiment is based on the prediction result processing of the multi-product combined demand prediction method with substitution and association effects: and determining respective demand proportions according to the actual demand forecast quantity of the new product and the existing commodity so as to guide the class planning problem of the retail store.
The embodiment provides a multi-product combined demand method based on substitution and association, which fully considers the purchasing behavior of consumers, obtains a more accurate commodity actual demand prediction result by more completely defining the purchasing behavior of the consumers in a model, expands the demand prediction model to be suitable for the situations of two or more associated products, establishes the relation among all the associated products through an association probability estimation value obtained by association rule mining, considers the relevance more completely, and further improves the accuracy of commodity actual demand prediction.
According to the multi-category combined demand forecasting method based on substitution and correlation, historical sales data of retail stores are used, and demand forecasting is achieved through parameter definition, model building and solving. Identifying attributes of key influences on commodity sales through an interpretable machine learning model; the method has the advantages that the purchasing behavior of the consumer, namely the definition substitution and association, is more completely identified in the model, the parameter definition of the model is realized, the association among different categories is established, and the accuracy of demand prediction by adopting the method is improved by the parameter definition process. A model is established by adopting a maximum likelihood estimation principle, the actual demand of each attribute combination commodity including a new commodity and an existing commodity is solved based on a differential evolution algorithm, and the demand proportion determined according to the actual demand of the existing commodity and the actual demand of the new commodity can be used for the class planning guidance of retail stores of different scales.
The embodiment also provides a device for predicting the multi-product combined demand based on substitution and association effects, which comprises:
the commodity attribute module is used for determining the core attribute of the commodity based on the sales information of the current commodity on sale and converting the SKU-level commodity into an attribute-level commodity;
the substitution and association module defines substitution and association functions among various attribute combined commodities in the purchasing behavior of the consumer;
and the commodity demand prediction module is used for bringing parameters for measuring substitution and correlation into an operation planning model established based on a maximum likelihood thought, and solving the actual demand and demand proportion of each attribute combined commodity including the existing commodity and the new product by using a meta-heuristic algorithm.
The embodiment also provides a computer storage medium storing a computer executable program, and when the computer executable program is executed, the multi-item combined demand forecasting method based on the substitution and association effects is realized.
The computer storage medium of the present embodiments may include a propagated data signal with readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable storage medium may also be any readable medium that is not a readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable storage medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
The present embodiment further provides an electronic device, which includes a processor and a memory, where the memory is used to store a computer executable program, and when the computer program is executed by the processor, the processor executes the multi-category demand forecasting method based on substitution and association.
The electronic device is in the form of a general purpose computing device. The processor can be one or more and can work together. The invention also does not exclude that distributed processing is performed, i.e. the processors may be distributed over different physical devices. The electronic device of the present invention is not limited to a single entity, and may be a sum of a plurality of entity devices.
The memory stores a computer executable program, typically machine readable code. The computer readable program may be executed by the processor to enable an electronic device to perform the method of the invention, or at least some of the steps of the method.
The memory may include volatile memory, such as Random Access Memory (RAM) and/or cache memory units, as well as non-volatile memory, such as read-only memory (ROM).
It should be understood that elements or components not shown in the above examples may also be included in the electronic device of the present invention. For example, some electronic devices further include a display unit such as a display screen, and some electronic devices further include a human-computer interaction element such as a button, a keyboard, and the like. Electronic devices are considered to be covered by the present invention as long as the electronic devices are capable of executing a computer-readable program in a memory to implement the method of the present invention or at least a part of the steps of the method.
From the above description of the embodiments, those skilled in the art will readily appreciate that the present invention can be implemented by hardware capable of executing a specific computer program, such as the system of the present invention, and electronic processing units, servers, clients, mobile phones, control units, processors, etc. included in the system. The invention may also be implemented by computer software executing the method of the invention, e.g. by control software executed by a microprocessor, an electronic control unit, a client, a server, etc. It should be noted, however, that the computer software for carrying out the method of the present invention is not limited to being executed by one or a specific number of hardware entities, and may be implemented in a distributed fashion by unspecified specific hardware. For computer software, the software product may be stored in a computer readable storage medium (which may be a CD-ROM, a usb disk, a removable hard disk, etc.) or may be distributed over a network, as long as it enables the electronic device to perform the method according to the present invention.
The above embodiments are only used for illustrating the invention and not for limiting the technical solutions described in the invention, and although the present invention has been described in detail in the present specification with reference to the above embodiments, the present invention is not limited to the above embodiments, and therefore, any modification or equivalent replacement of the present invention is made; all such modifications and variations are intended to be included herein within the scope of this disclosure and the appended claims.
Claims (10)
1. The multi-category combined demand forecasting method based on substitution and correlation effects is characterized by comprising the following steps of:
determining the core attribute of the commodity based on the sales information of the current commodity on sale, and converting the SKU-level commodity into an attribute-level commodity;
defining substitution and correlation functions among various attribute combined commodities in the purchasing behavior of the consumer;
and (3) parameters for measuring substitution and correlation actions are incorporated into an operation planning model established based on a maximum likelihood thought, and a meta-heuristic algorithm is used for solving the actual demand and demand proportion of each attribute combined commodity including the existing commodity and the new product.
2. The method of claim 1, wherein the determining core attributes of the commodity based on sales information of the currently sold commodity, and the converting the SKU-level commodity into an attribute-level commodity comprises:
extracting attribute features of a current commodity on sale, wherein the attribute features comprise a commodity specification category, a price category, a brand grade category, a brand origin category and a function category;
and taking the corresponding periodic sales volume of the current commodity on sale as a label, training a random forest regression model, determining the core attribute of the commodity through the marginal contribution degree sequence of the features based on an interpretable SHAP model, and converting the SKU-level commodity into the attribute-level commodity.
3. The method for forecasting combined demand of multiple categories based on substitution and correlation as claimed in claim 1, wherein the defining substitution roles among the combination commodities with various attributes in the purchasing behavior of the consumer comprises:
based on the fact that the substitution effect among the commodities of the same class is measured by different attribute combinations of the current commodities on sale, the idea of joint analysis is used for reference, and the substitution probability among the commodities of the corresponding attribute combination is determined to be the product of the substitution probabilities among the attributes.
4. The multi-item combined demand forecasting method based on substitution and correlation, characterized in that the calculation process of the substitution probability comprises the following steps:
the commodity set corresponding to a certain category is S, the favorite commodity of the user is i, when i is not in the category set S, the user may replace the similar commodity j of i in the category set S, and the replacement probability is pi ij (ii) a Then the level is j on the a attributes respectively 1 ，j 2 ，…，j A The j commodity substitutes are respectively i on the A attributes 2 ，i 2 ，...，i A I goods of (a) has a probability of
5. The multi-item-class combined demand forecasting method based on substitution and correlation, characterized in that, the defining the correlation between the combined commodities with various attributes in the purchasing behavior of the consumer comprises:
determining association categories and association probabilities based on association rules obtained by an Apriori algorithm, and assuming the association probabilities as confidence results corresponding to the mining of the association rules;
assuming that the first category determined by association rule mining corresponds to the commodities of the m attribute combinations, the second category corresponds to the commodities of the n attribute combinations, mn estimation values of association probabilities are obtained according to the records of tickets purchased by users, and the commodity names in the ticket data are changed into attribute level names during association rule mining, so that the corresponding association probabilities among the commodities of each attribute combination of the cross-category are obtained.
6. The multi-item combined demand forecasting method based on substitution and correlation effects as claimed in claim 5, characterized by comprising:
the attribute combination commodity sets of a pair of related categories are S and Z respectively, the commodity p of each attribute combination in the set Z has a correlation function on the commodity j of each attribute combination in the set S, and the correlation probability is w jp The probability of association of a product q whose attributes are not combined in the set Z with a product j whose attributes are combined in the set S is w jp π qp 。
7. The method according to claim 5, wherein the step of incorporating the parameters for measuring the substitution and correlation into the operation planning model based on the maximum likelihood idea, and the step of solving the actual demand and demand ratio of the commodity with each attribute combination including the existing commodity and the new commodity by using the meta-heuristic algorithm comprises:
the core attribute parameters, the association probability, the substitution probability and the probability of the most favorable attribute combined commodity are brought into an operation planning model established based on a maximum likelihood thought:
attribute combination of a pair of related item classesThe commodity set is S and Z respectively, and the probability F of purchasing the attribute combination commodity j in the set S j (S) is
Probability F of purchasing property group item p in set Z p (Z) similarly, the maximum likelihood estimation has the objective function of
Wherein f is j Probability of finger preferring commodity of 8230 P 、x j Store sales of goods p, j, respectively;
and solving the actual demand quantity of each attribute combined commodity including the existing commodity and the new commodity by using a meta-heuristic algorithm.
8. The multi-item combined demand forecasting method based on substitution and correlation effects as claimed in claim 7, characterized by comprising:
and determining the association probability among the associated categories based on confidence results obtained by mining association rules by an Apriori algorithm, and establishing the association among a plurality of categories, thereby establishing a multi-category combined demand prediction model, wherein an objective function is the product of powers of purchase probabilities of the categories, and the power items are the store sales volume of the combined commodity with the attributes corresponding to the categories respectively.
9. The device for predicting the multi-product combined demand based on substitution and correlation effects is characterized by comprising the following steps:
the commodity attribute module is used for determining the core attribute of the commodity based on the sales information of the current commodity on sale and converting the SKU-level commodity into an attribute-level commodity;
the substitution and association module defines substitution and association functions among various attribute combined commodities in the purchasing behavior of the consumer;
and the commodity demand prediction module is used for bringing parameters for measuring substitution and correlation into an operation planning model established based on a maximum likelihood thought, and solving the actual demand and demand proportion of each attribute combined commodity including the existing commodity and the new product by using a meta-heuristic algorithm.
10. A computer storage medium storing a computer executable program, wherein the computer executable program when executed implements the method for predicting multi-category demand based on substitution and correlation according to any one of claims 1 to 8.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211655555.0A CN115809891A (en) | 2022-12-22 | 2022-12-22 | Multi-product combined demand prediction method and device based on substitution and correlation effects |
| CN202310380958.7A CN116579803A (en) | 2022-12-22 | 2023-04-11 | Multi-class joint demand prediction method and device based on substitution and association effects |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211655555.0A CN115809891A (en) | 2022-12-22 | 2022-12-22 | Multi-product combined demand prediction method and device based on substitution and correlation effects |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115809891A true CN115809891A (en) | 2023-03-17 |
Family
ID=85486659
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211655555.0A Pending CN115809891A (en) | 2022-12-22 | 2022-12-22 | Multi-product combined demand prediction method and device based on substitution and correlation effects |
| CN202310380958.7A Pending CN116579803A (en) | 2022-12-22 | 2023-04-11 | Multi-class joint demand prediction method and device based on substitution and association effects |
Family Applications After (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310380958.7A Pending CN116579803A (en) | 2022-12-22 | 2023-04-11 | Multi-class joint demand prediction method and device based on substitution and association effects |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN115809891A (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117787867A (en) * | 2024-02-27 | 2024-03-29 | 山东财经大学 | Medicine inventory demand analysis method and system |
-
2022
- 2022-12-22 CN CN202211655555.0A patent/CN115809891A/en active Pending
-
2023
- 2023-04-11 CN CN202310380958.7A patent/CN116579803A/en active Pending
Also Published As
| Publication number | Publication date |
|---|---|
| CN116579803A (en) | 2023-08-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108268934A (en) | Recommendation method and apparatus, electronic equipment, medium, program based on deep learning | |
| US10282797B2 (en) | Inference model for traveler classification | |
| US20200234218A1 (en) | Systems and methods for entity performance and risk scoring | |
| CN111027895A (en) | Stock prediction and behavior data collection method, apparatus, device and medium for commodity | |
| CN113157752B (en) | Scientific and technological resource recommendation method and system based on user portrait and situation | |
| JPWO2005024689A1 (en) | Method and apparatus for analyzing consumer purchasing behavior | |
| CN108305181B (en) | Social influence determination method and device, information delivery method and device, equipment and storage medium | |
| CN112365283B (en) | Coupon issuing method and device, terminal equipment and storage medium | |
| CN113742492A (en) | Insurance scheme generation method and device, electronic equipment and storage medium | |
| CN109711917A (en) | Information-pushing method and device | |
| CN111429214B (en) | Transaction data-based buyer and seller matching method and device | |
| KR20140115430A (en) | Method for providing evaluation of online seller | |
| US8738459B2 (en) | Product recommendation | |
| CN113780479A (en) | Periodic prediction model training method and device, and periodic prediction method and equipment | |
| CN115809891A (en) | Multi-product combined demand prediction method and device based on substitution and correlation effects | |
| JPH06119309A (en) | Purchase prospect degree predicting method and customer management system | |
| JP5403809B2 (en) | Customer information management server and customer information management program | |
| CN112950320A (en) | Automatic commodity online and offline method and device based on machine self-learning and electronic equipment | |
| WO2020150597A1 (en) | Systems and methods for entity performance and risk scoring | |
| CN110659947A (en) | Commodity recommendation method and device | |
| CN110197191A (en) | Electronic game recommended method | |
| CN115905558A (en) | Knowledge graph-based XAI model evaluation method, device, equipment and medium | |
| CN111784091B (en) | Method and device for processing information | |
| CN115375219A (en) | Inventory item forecast and item recommendation | |
| JP6809148B2 (en) | Program and combination extraction system |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20230317 |