CN117745338A - Wine consumption prediction method based on curvelet transformation, electronic equipment and storage medium - Google Patents
Wine consumption prediction method based on curvelet transformation, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN117745338A CN117745338A CN202410186242.8A CN202410186242A CN117745338A CN 117745338 A CN117745338 A CN 117745338A CN 202410186242 A CN202410186242 A CN 202410186242A CN 117745338 A CN117745338 A CN 117745338A
- Authority
- CN
- China
- Prior art keywords
- consumer
- wine
- data
- curvelet
- attribute
- 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.)
- Granted
Links
- 230000009466 transformation Effects 0.000 title claims abstract description 53
- 238000000034 method Methods 0.000 title claims abstract description 45
- 238000003860 storage Methods 0.000 title claims abstract description 19
- LFQSCWFLJHTTHZ-UHFFFAOYSA-N Ethanol Chemical compound CCO LFQSCWFLJHTTHZ-UHFFFAOYSA-N 0.000 claims abstract description 16
- 230000006399 behavior Effects 0.000 claims description 19
- 235000013334 alcoholic beverage Nutrition 0.000 claims description 16
- 230000006870 function Effects 0.000 claims description 16
- 238000012163 sequencing technique Methods 0.000 claims description 12
- 230000008859 change Effects 0.000 claims description 9
- 238000004590 computer program Methods 0.000 claims description 8
- 238000009826 distribution Methods 0.000 claims description 8
- 230000000694 effects Effects 0.000 claims description 7
- 230000008569 process Effects 0.000 claims description 7
- 230000001629 suppression Effects 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 4
- 238000012952 Resampling Methods 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 3
- 238000012797 qualification Methods 0.000 claims description 3
- 238000000354 decomposition reaction Methods 0.000 claims description 2
- 230000001476 alcoholic effect Effects 0.000 claims 3
- 238000012545 processing Methods 0.000 abstract description 4
- 238000007781 pre-processing Methods 0.000 abstract description 2
- 238000004458 analytical method Methods 0.000 description 3
- 238000013178 mathematical model Methods 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000012854 evaluation process Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 238000005215 recombination Methods 0.000 description 1
- 230000006798 recombination Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
- 239000013598 vector Substances 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- 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
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a wine consumption prediction method based on curvelet transformation, electronic equipment and a storage medium, which belong to the technical field of big data processing, and the technical problem to be solved by the invention is how to accurately predict the purchasing power of wine consumers, so that a merchant can better adjust the stock of wine commodities, thereby improving the selling efficiency of the merchant, and the adopted technical scheme is as follows: selecting and preprocessing the data of wine consumers; alcohol consumer data reconstruction: the original wine consumer data which are randomly and randomly arranged are constructed into ordered two-dimensional data according to two dimensions of consumer information and time; data curvelet transformation: performing data curvelet transformation on the reconstructed wine consumer data to obtain curvelet coefficients capable of describing characteristics of the wine consumer; suppressing interference information; extracting characteristic information: performing inverse curvelet transformation according to the curvelet coefficients to extract consumer group description data with high purchase probability and consumer individuals with high purchase probability; wine consumers purchase predictions.
Description
Technical Field
The invention relates to the technical field of big data processing, in particular to a wine consumption prediction method based on curvelet transformation, electronic equipment and a storage medium.
Background
Consumption level prediction is a method specifically used for predicting the demand of various consumer products. The method mainly utilizes the visual analysis and judgment of consumption level and total number of consumption (or house number) and is aided with simple mathematical calculation to predict the demand of the consumer products. The method is characterized in that the method comprises the steps of providing a market forecast, providing a production enterprise with demand information, facilitating the arrangement of production, facilitating the organization and the delivery of the enterprise, and timely meeting the demand of consumers on the commodity.
With the development of information technology, consumer purchase prediction is increasingly focused on various enterprises. Current wine consumer purchase predictions mainly include predictions based on simple mathematical models and predictions based on big data intelligent models. The prediction based on the simple mathematical model is a method for predicting consumer consumption trend by consumer consumption data, fitting consumption curve, and has the advantages of simple pretreatment of data, low information mining degree and low prediction result precision, and can not provide effective guidance in the wine marketing process. The prediction based on the big data intelligent model is a method for predicting the purchase possibility of consumers by constructing the intelligent model based on big data and carrying out a large amount of learning training through complex network relations, and can solve the defects of the prediction method based on the simple mathematical model, but the situations that the calculation model is solidified, the resource requirement is high, the operation period is long and the calculation model cannot adapt to the rapid change of consumers and commodities can not be met.
Therefore, how to accurately predict the purchasing power of wine consumers, so that merchants can better adjust the inventory of wine commodities, and further improve the selling efficiency of the merchants is a technical problem to be solved at present.
Disclosure of Invention
The invention aims to provide a wine consumption prediction method, electronic equipment and storage medium based on curvelet transformation, which are used for solving the problem of how to accurately predict the purchasing power of wine consumers, so that a merchant can better adjust the inventory of wine commodities, and further improve the selling efficiency of the merchant.
The technical task of the invention is realized in the following way, namely, the method for predicting wine consumption based on curvelet transformation comprises the following steps:
alcohol consumer data selection and pretreatment: selecting basic data, purchase data and behavior data of wine consumers, filtering out the basic data with incomplete registration information and purchase data with incomplete order flow, converting time data in the wine consumer data into a uniform format, and expressing text data in the wine consumer data by using numerical values;
alcohol consumer data reconstruction: the original wine consumer data which are randomly and randomly arranged are constructed into ordered two-dimensional data according to two dimensions of consumer information and time;
data curvelet transformation: performing data curvelet transformation on the reconstructed wine consumer data to obtain curvelet coefficients capable of describing characteristics of the wine consumer;
suppressing interference information;
extracting characteristic information: performing inverse curvelet transformation according to the curvelet coefficients to extract consumer group description data with high purchase probability and consumer individuals with high purchase probability;
wine consumer purchase prediction: and taking the consumer group description data of the purchase possibility as a coefficient of the consumer individual to acquire a prediction result based on the consumer group description consumer individual.
Preferably, the basic data of the alcoholic beverage consumer is data information of age, sex, date of birth, store registration time, store home zone and store to which the consumer selected from the consumer profile table, consumer registration table and store-consumer relation table belongs;
the purchase data of the wine consumers is data information of consumer shopping amount, shopping quantity, commodity purchase, time of payment, time of picking up commodity, time of signing in and purchasing store selected from the payment order table, the purchase order table and the commodity order table;
the behavior data of the wine consumer is data information of a purchase mode, a goods picking mode, a payment mode, an activity participation degree, a purchase qualification giving-up rate and a refund number selected from a purchase order table, an activity information table, a medium ticket table and an after-sales order table.
More preferably, the alcoholic beverage consumer data reconstruction is to sequentially arrange the attribute dimension and the time dimension respectively using the age, sex, date of birth, store registration time, store attribution region, attribute data of the affiliated store, time of order, payment time and time data of pickup time of the affiliated store of the alcoholic beverage consumer as variables, and purchase amount and purchase number in the purchase data and the behavior data of the alcoholic beverage consumer as results.
More preferably, the attribute dimensions are ordered as follows:
different attributes of wine consumers occupy different weights in the sorting process of the wine consumers, and sorting values of the wine consumers are defined as follows:
；
wherein,a ranking value representing wine consumers; />An attribute value representing a consumer of the wine; />Representing wine consumer attribute coefficients (constants); />Representing the number of the attributes of the wine consumers;
the age of the wine consumer and the store to which the wine consumer belongs are taken as core attributes of the attribute dimension of the wine consumer,is changed into->；
Wherein,an age attribute representing wine consumers; />Representing store attributes to which wine consumers belong; />Representing an ordered ranking map; />Attribute coefficients representing the age of wine consumers; />Attribute coefficients representing stores to which wine consumers belong;
the attribute coefficient of the age of the wine consumer and the attribute coefficient of the store to which the wine consumer belongs are solved specifically as follows:
single attribute ordering: calculating the average wine consumption amount of wine consumption corresponding to the age attribute of the wine consumer or the store attribute of the wine consumer, and sorting according to the age attribute of the wine consumer or the average wine consumption amount of the store attribute of the wine consumer;
sorting of double attributes: calculating the age attribute of the wine consumer and the average wine consumption amount of wine consumption corresponding to the affiliated store attribute, and sorting the average wine consumption amount according to the first age attribute, the affiliated store attribute and the first affiliated store attribute and the second age attribute respectively;
comparing the reverse ordinal numbers of the average wine consumption amount of the single attribute sequencing and the double attribute sequencing, and regarding the sequencing mode with small reverse ordinal number as the optimal sequencing; thereby obtainingThe value is 1, & lt + & gt>The value is the number of the age attribute values, namely the maximum sorting value of the age attribute;
and so on,the values of (2) are as follows:
when (when)When (I)>；
When (when)When (I)>；
Wherein,representing the number of consumer attributes; />Representing the maximum ranking of consumer attributes.
More preferably, the ordered arrangement of the time dimensions is specifically as follows:
the only time of the consumer is described as；
Wherein y is c Representing the comprehensive consumption time of the consumer; y is n Representing a consumer time parameter; b n Representing a consumer time coefficient; t represents the number of consumer times;
taking the payment time and the goods taking time as core data, setting time ordering as optimal ordering, and obtaining:；
will beReduced to time solving for the average:
；
wherein y is c Representing the comprehensive consumption time of the consumer; y is pay Representing a consumer payment time; y is get Indicating the time of pickup by the consumer.
Preferably, the data curvelet transform is specifically as follows:
the curvelet change after data reconstruction of the data of the wine consumer is expressed as:
；
wherein,expressed in the scale +.>Direction->Position->A curvelet coefficient; />Representing consumer data in the consumer dimension +.>And time dimension->Is a distribution of (3); />Representing a curvelet function;
alcoholic beverage consumer in Cartesian coordinate systemDataFor input, the discrete form of the curvelet transform is expressed as:
；
wherein,representing the scale +.>Direction->Position->A curvelet coefficient; />Is a discrete curvelet function; />；
Data on wine consumersPerforming discrete Fourier transform to obtain frequency domain expressed asThe method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>；
Resampling is performed in the frequency domain for different scales and angles, and the acquisition of a new frequency domain is expressed as:
；
wherein,；/>representing the support area of the frequency window, the support length and width are respectively +.>And->；
The set of curvelet coefficients is expressed as:
；
wherein,representing an inverse fast fourier transform; />Representing wine consumer data at consumer +.>-time->Window functions in coordinates;
after the data of the wine consumer is converted by the curvelet, the curvelet coefficient capable of describing the characteristics of the wine consumer is obtainedObtaining the numerical distribution of the curvelet coefficient set, namely the curvelet coefficient;
dividing the curvelet coefficients into a Coarse scale layer (Coarse), a Fine scale layer (Detail) and a Fine scale layer (Fine) according to the nature of curvelet transformation and the curvelet coefficients; wherein the coarse scale layer comprises low frequency data information corresponding to alcoholic beverage consumer data-time of dayInterdomain->Overall profile information of all wine consumer data in (a) describing information of a population of wine consumers; the fine scale layer (Detail) comprises medium-high frequency data information and is decomposed in multiple directions by multiple direction parameters, corresponding to different wine consumers +.>Different times->Fine features in the direction, describing individual information of wine consumers; the fine scale includes high frequency data information corresponding to individual random purchase information in the wine consumer data, which is considered as invalid interference information during the wine consumer prediction process.
More preferably, the suppression of interference information is specifically as follows:
in general, the individual purchasing behavior is considered to be random behavior, has no predictability and relatively low possibility of purchasing again outside the consumer group with high purchasing possibility, the curved wave coefficient threshold is set for the high-frequency data information in the curved wave coefficient corresponding to the wine consumer data of the individual purchasing behavior, the high-frequency data with the frequency larger than the curved wave coefficient threshold is assigned as the curved wave coefficient threshold during calculation, namely, the high-frequency data with the frequency larger than the curved wave coefficient threshold is directly assigned as the curved wave coefficient threshold, so that the suppression of the interference information is achieved.
More preferably, the feature information extraction is specifically as follows:
and performing inverse curvelet transformation on the low-frequency data and the medium-low frequency data in the curvelet coefficient to extract consumer group description data with high purchase possibility, wherein the formula is as follows:
；
wherein,for describing consumptionConsumption of the community of people; />Representing inverse curvelet transformation;
and carrying out inverse curvelet transformation on medium-high frequency data in the curvelet coefficient to extract a consumer with high purchase possibility, wherein the formula is as follows:
；
wherein,for describing the consumption of the individual consumer; />Representing inverse curvelet transformation;
the wine consumer purchase forecast is specifically as follows:
taking the data of the group description as the coefficient of the consumer individual, and acquiring a prediction result of the consumer individual based on the consumer group description, wherein the formula is as follows:
；
wherein,indicating a prediction of the likelihood of purchase for the wine consumer, a larger value indicates a higher likelihood of purchase.
An electronic device, comprising: a memory and at least one processor;
wherein the memory has a computer program stored thereon;
the at least one processor executes the computer program stored by the memory, causing the at least one processor to perform the wine consumption prediction method based on curvelet transformation as described above.
A computer readable storage medium having stored therein a computer program executable by a processor to implement a method of wine consumption prediction based on curvelet transformation as described above.
The alcohol consumption prediction method based on curvelet transformation, the electronic equipment and the storage medium have the following advantages:
according to the method, the purchasing potential of the consumer is analyzed, the purchasing possibility of the consumer is predicted, the condition that merchants screen high-quality customers and common customers in the consumer is met, so that the merchants can sell corresponding commodities to corresponding consumers better in marketing selling, selling efficiency is improved, business income is improved, and the method is suitable for an intelligent selling environment better;
secondly, the invention can reasonably combine data, efficiently suppress interference, accurately extract characteristics and complete purchase prediction comprehensively, accurately and rapidly by constructing data processing based on curvelet transformation; meanwhile, the evaluation of consumers can be avoided due to various external factors such as commodity heat, time heat, activity heat and the like; deep attribute information of the consumer is mined, and more accurate prediction is made for the consumer;
thirdly, the invention regards the reaction of each consumer in the purchasing behavior as a signal feedback, combines the time dimension to obtain a piece of information data, constructs two-dimensional data with strong association relations in the horizontal (time) and the vertical (consumers) according to the relation of basic attributes (age, region and the like) of different consumers, and realizes data recombination;
after the consumer data are recombined into the two-dimensional signal data, the interference information can be suppressed by using a time-frequency analysis method with high efficiency, short time consumption and low cost, the consumer information data are converted into a curvelet domain through curvelet transformation, the interference suppression is performed on the consumer data, and the influence of useless information and error information on the consumer evaluation process is reduced;
after data is converted into a curvelet domain to suppress interference by using curvelet change, the characteristic signals can be filtered and extracted on different scales, angles and positions by two-dimensional consumer signal data according to the nature of curvelet change, time domain data is obtained by inverse transformation, and the energy of the signals is calculated by integration, so that the purchase possibility of consumers is described by the intensity of the energy;
after the data reconstruction, the invention ensures that adjacent data have strong association relation, and then according to the property of curvelet change, the characteristics on different scales, angles and positions can be mined, when the purchase possibility of one consumer is predicted, the consumer's own consumption data is used, and the data of other adjacent consumers are used as auxiliary characteristics to comprehensively describe the consumer together, so that the description of the consumer is prevented from being influenced by special factors;
the method comprises the steps of constructing ordered models of consumer basic data, consumer data and behavior data, regarding one consumer data as a signal, mapping the data into a curvelet domain through curvelet change, performing interference suppression and feature extraction, inversely transforming the data into a consumer-time domain, and finally taking group features as coefficients to obtain a prediction result of describing consumer individuals based on consumer groups;
according to the invention, the traditional mode of evaluating the single consumer by using the data of the single consumer is converted into the model of predicting the purchase possibility of one consumer by using the data of all consumers, so that the problem that the single consumer has excessive expectations caused by random consumption and disposable consumption can be solved, the consumption behavior of the consumer can be evaluated more accurately and comprehensively, and a more credible prediction result is given.
Drawings
The invention is further described below with reference to the accompanying drawings.
FIG. 1 is a block flow diagram of a method for predicting wine consumption based on curvelet transformation.
Detailed Description
The alcohol consumption prediction method, the electronic device and the storage medium according to the present invention based on curvelet transformation will be described in detail below with reference to the accompanying drawings and the specific embodiments.
Example 1:
as shown in fig. 1, the embodiment provides a method for predicting wine consumption based on curvelet transformation, which specifically comprises the following steps:
s1, selecting and preprocessing wine consumer data: selecting basic data, purchase data and behavior data of wine consumers, filtering out the basic data with incomplete registration information and purchase data with incomplete order flow, converting time data in the wine consumer data into a uniform format, and expressing text data in the wine consumer data by using numerical values;
s2, reconstructing wine consumer data: the original wine consumer data which are randomly and randomly arranged are constructed into ordered two-dimensional data according to two dimensions of consumer information and time;
s3, data curvelet transformation: performing data curvelet transformation on the reconstructed wine consumer data to obtain curvelet coefficients capable of describing characteristics of the wine consumer;
s4, suppressing interference information;
s5, extracting characteristic information: performing inverse curvelet transformation according to the curvelet coefficients to extract consumer group description data with high purchase probability and consumer individuals with high purchase probability;
s6, purchasing prediction of wine consumers: and taking the consumer group description data of the purchase possibility as a coefficient of the consumer individual to acquire a prediction result based on the consumer group description consumer individual.
The basic data of the alcoholic beverage consumer in step S1 of the present embodiment is data information of age, sex, date of birth, store registration time, store home zone, and affiliated store of the consumer selected from the consumer profile table, consumer registration table, and store-consumer relationship table.
The purchase data of the alcoholic beverage consumer in step S1 of the present embodiment is data information of the consumer shopping amount, shopping quantity, purchased goods, purchase time, payment time, pickup time, sign-in time and purchase store selected from the payment order table, purchase order table and commodity order table.
The behavior data of the alcoholic beverage consumer in step S1 of the present embodiment is data information of a purchase mode, a pick-up mode, a payment mode, an activity participation degree, a purchase qualification giving-up rate, and a refund number selected from a purchase order table, an activity information table, a middle ticket table, and an after-sales order table.
In step S2 of this embodiment, the alcoholic beverage consumer data reconstruction is performed by using the age, sex, date of birth, store registration time, store home zone, attribute data of the store to which the alcoholic beverage consumer belongs, and time data of the order time, payment time, and pickup time as variables, and using the purchase amount and purchase number in the purchase data and the behavior data of the alcoholic beverage consumer as results, and sequentially arranging the attribute dimension and the time dimension, respectively.
The distribution of consumers is unordered, and a ranking order is added to the consumers according to the attributes of the consumers to ensure that adjacent consumers have a certain association relationship.
The ordered arrangement of attribute dimensions in this embodiment is specifically as follows:
(1) Different attributes of wine consumers occupy different weights in the sorting process of the wine consumers, and sorting values of the wine consumers are defined as follows:
；
wherein,a ranking value representing wine consumers; />An attribute value representing a consumer of the wine; />Representing wine consumer attribute coefficients (constants); />Representing the number of the attributes of the wine consumers;
(2) The age of the wine consumer and the store to which the wine consumer belongs are taken as core attributes of the attribute dimension of the wine consumer,is changed into->；
Wherein,an age attribute representing wine consumers; />Representing store attributes to which wine consumers belong; />Representing an ordered ranking map; />Attribute coefficients representing the age of wine consumers; />Attribute coefficients representing stores to which wine consumers belong;
(3) The attribute coefficient of the age of the wine consumer and the attribute coefficient of the store to which the wine consumer belongs are solved specifically as follows:
(1) single attribute ordering: calculating the average wine consumption amount of wine consumption corresponding to the age attribute of the wine consumer or the store attribute of the wine consumer, and sorting according to the age attribute of the wine consumer or the average wine consumption amount of the store attribute of the wine consumer; as shown in tables 1 and 2:
TABLE 1 average wine consumption amount ranking table of wine consumption corresponding to age attributes of wine consumers
Table 2 average alcohol consumption amount ranking table of alcohol consumption corresponding to store attributes to which alcohol consumers belong
(2) Sorting of double attributes: calculating the age attribute of the wine consumer and the average wine consumption amount of wine consumption corresponding to the affiliated store attribute, and sorting the average wine consumption amount according to the first age attribute, the affiliated store attribute and the first affiliated store attribute and the second age attribute respectively; as shown in tables 3 and 4:
TABLE 3 average wine consumption amount ranking table according to age attribute and store attribute
TABLE 4 average wine consumption amount ranking table according to first belonging store attribute and then age attribute
(3) Comparing the reverse ordinal numbers of the average wine consumption amount of the single attribute sequencing and the double attribute sequencing, and regarding the sequencing mode with small reverse ordinal number as the optimal sequencing; the data can be seen to be firstly ordered according to store and then according to age, and can be obtained according to the dataThe value is 1, & lt + & gt>The value is the number of the age attribute values and is also the maximum sorting value of the age attribute;
and so on,the values of (2) are as follows:
when (when)When (I)>；
When (when)When (I)>；
Wherein,representing the number of consumer attributes; />Representing the maximum ranking of consumer attributes.
The time dimension ordering arrangement in this embodiment is specifically as follows:
the only time of the consumer is described as；
Wherein y is c Representing the comprehensive consumption time of the consumer; y is n Representing a consumer time parameter; b n Representing a consumer time coefficient; t represents the number of consumer times;
taking the payment time and the goods taking time as core data, setting time ordering as optimal ordering, and obtaining:；
will beReduced to time solving for the average:
；
wherein y is c Representing the comprehensive consumption time of the consumer; y is pay Representing a consumer payment time; y is get Indicating the time of pickup by the consumer.
The data curvelet transformation in step S3 of this embodiment is specifically as follows:
the curvelet transform is the decomposition of the original signal on a given set of functions or vectors, resulting in a sparse representation of the original data over the transform domain. The curvelet change is a multi-scale analysis tool capable of extracting information of specific scale, specific direction and specific frequency in signals and images.
S301, curvelet change after data reconstruction of the data of wine consumers is expressed as:
；
wherein,expressed in the scale +.>Direction->Position->A curvelet coefficient; />Representing consumer data in the consumer dimension +.>And time dimension->Is a distribution of (3); />Representing a curvelet function;
s302, wine consumer data in Cartesian coordinate systemFor input, the discrete form of the curvelet transform is expressed as:
；
wherein,is shown inScale +.>Direction->Position->A curvelet coefficient; />Is a discrete curvelet function; />；
S303, data on wine consumersPerforming discrete Fourier transform to obtain frequency domain expressed asThe method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>；
S304, resampling is carried out on different scales and angles in the frequency domain, and the new frequency domain is obtained and expressed as:
；
wherein,；/>representing the support area of the frequency window, the support length and width are respectively +.>And->；
S305, a set of curveback coefficients is expressed as:
；
wherein,representing an inverse fast fourier transform; />Representing wine consumer data at consumer +.>-time->Window functions in coordinates;
s306, after the data of the wine consumers are converted through the curvelet, the curvelet coefficient capable of describing the characteristics of the wine consumers is obtainedThe numerical distribution of the set of curvelet coefficients, i.e. curvelet coefficients, is obtained as shown in table 5:
TABLE 5 numerical distribution table of the set of curvelet coefficients
Dividing the curvelet coefficients into a Coarse scale layer (Coarse), a Fine scale layer (Detail) and a Fine scale layer (Fine) according to the nature of curvelet transformation and the curvelet coefficients; wherein the coarse scale layer comprises low frequency data information corresponding to alcoholic beverage consumer data-time domain->Overall profile information of all wine consumer data in (a) describing wineInformation of consumer groups; the fine scale layer (Detail) comprises medium-high frequency data information and is decomposed in multiple directions by multiple direction parameters, corresponding to different wine consumers +.>Different times->Fine features in the direction, describing individual information of wine consumers; the fine scale includes high frequency data information corresponding to individual random purchase information in the wine consumer data, which is considered as invalid interference information during the wine consumer prediction process.
The suppression of the interference information in step S4 of this embodiment is specifically as follows:
in general, the individual purchasing behavior is considered to be random behavior, has no predictability and relatively low possibility of purchasing again outside the consumer group with high purchasing possibility, the curved wave coefficient threshold is set for the high-frequency data information in the curved wave coefficient corresponding to the wine consumer data of the individual purchasing behavior, the high-frequency data with the frequency larger than the curved wave coefficient threshold is assigned as the curved wave coefficient threshold during calculation, namely, the high-frequency data with the frequency larger than the curved wave coefficient threshold is directly assigned as the curved wave coefficient threshold, so that the suppression of the interference information is achieved.
The feature information extraction in step S5 of this embodiment is specifically as follows:
s501, performing inverse curvelet transformation on low-frequency data and medium-low frequency data in curvelet coefficients, and extracting consumer group description data with high purchase possibility, wherein the formula is as follows:
；
wherein,for describing consumption by a consumer group; />Representation ofInverse curvelet transformation;
s502, performing inverse curvelet transformation on medium-high frequency data in curvelet coefficients to extract consumer individuals with high purchase possibility, wherein the formula is as follows:
；
wherein,for describing the consumption of the individual consumer; />Representing an inverse curvelet transform.
The wine consumer purchase forecast in step S6 of this embodiment is specifically as follows:
taking the data of the group description as the coefficient of the consumer individual, and acquiring a prediction result of the consumer individual based on the consumer group description, wherein the formula is as follows:
；/>
wherein,indicating a prediction of the likelihood of purchase for the wine consumer, a larger value indicates a higher likelihood of purchase.
Example 2:
the embodiment also provides an electronic device, including: a memory and a processor;
wherein the memory stores computer-executable instructions;
the processor executes the computer-executable instructions stored in the memory, so that the processor executes the alcohol consumption prediction method based on curvelet transformation in any embodiment of the invention.
The processor may be a Central Processing Unit (CPU), but may also be other general purpose processors, digital Signal Processors (DSPs), application Specific Integrated Circuits (ASICs), off-the-shelf programmable gate arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory may be used to store computer programs and/or modules, and the processor implements various functions of the electronic device by running or executing the computer programs and/or modules stored in the memory, and invoking data stored in the memory. The memory may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function, and the like; the storage data area may store data created according to the use of the terminal, etc. The memory may also include high-speed random access memory, but may also include non-volatile memory, such as a hard disk, memory, plug-in hard disk, memory card only (SMC), secure Digital (SD) card, flash memory card, at least one disk storage period, flash memory device, or other volatile solid state memory device.
Example 3:
the present embodiment also provides a computer-readable storage medium having stored therein a plurality of instructions, the instructions being loaded by a processor, causing the processor to execute the curvelet consumption prediction method based on curvelet transformation in any of the embodiments of the present invention. Specifically, a system or apparatus provided with a storage medium on which a software program code realizing the functions of any of the above embodiments is stored, and a computer (or CPU or MPU) of the system or apparatus may be caused to read out and execute the program code stored in the storage medium.
In this case, the program code itself read from the storage medium may realize the functions of any of the above-described embodiments, and thus the program code and the storage medium storing the program code form part of the present invention.
Examples of storage media for providing program code include floppy disks, hard disks, magneto-optical disks, optical disks (e.g., CD-ROMs, CD-R, CD-RWs, DVD-ROMs, DVD-RYM, DVD-RWs, DVD+RWs), magnetic tapes, nonvolatile memory cards, and ROMs. Alternatively, the program code may be downloaded from a server computer by a communication network.
Further, it should be apparent that the functions of any of the above-described embodiments may be implemented not only by executing the program code read out by the computer, but also by causing an operating system or the like operating on the computer to perform part or all of the actual operations based on the instructions of the program code.
Further, it is understood that the program code read out by the storage medium is written into a memory provided in an expansion board inserted into a computer or into a memory provided in an expansion unit connected to the computer, and then a CPU or the like mounted on the expansion board or the expansion unit is caused to perform part and all of actual operations based on instructions of the program code, thereby realizing the functions of any of the above embodiments.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some or all of the technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit of the invention.
Claims (10)
1. A wine consumption prediction method based on curvelet transformation is characterized by comprising the following steps:
alcohol consumer data selection and pretreatment: selecting basic data, purchase data and behavior data of wine consumers, filtering out the basic data with incomplete registration information and purchase data with incomplete order flow, converting time data in the wine consumer data into a uniform format, and expressing text data in the wine consumer data by using numerical values;
alcohol consumer data reconstruction: the original wine consumer data which are randomly and randomly arranged are constructed into ordered two-dimensional data according to two dimensions of consumer information and time;
data curvelet transformation: performing data curvelet transformation on the reconstructed wine consumer data to obtain curvelet coefficients capable of describing characteristics of the wine consumer;
suppressing interference information;
extracting characteristic information: performing inverse curvelet transformation according to the curvelet coefficients to extract consumer group description data with high purchase probability and consumer individuals with high purchase probability;
wine consumer purchase prediction: and taking the consumer group description data of the purchase possibility as a coefficient of the consumer individual to acquire a prediction result based on the consumer group description consumer individual.
2. The method for predicting wine consumption based on curvelet transformation according to claim 1, wherein the basic data of wine consumers is data information of age, sex, date of birth, time of store registration, region of store attribution and store of the consumer selected from consumer profile table, consumer registry table and store-consumer relationship table;
the purchase data of the wine consumers is data information of consumer shopping amount, shopping quantity, commodity purchase, time of payment, time of picking up commodity, time of signing in and purchasing store selected from the payment order table, the purchase order table and the commodity order table;
the behavior data of the wine consumer is data information of a purchase mode, a goods picking mode, a payment mode, an activity participation degree, a purchase qualification giving-up rate and a refund number selected from a purchase order table, an activity information table, a medium ticket table and an after-sales order table.
3. The alcohol consumption prediction method according to claim 1 or 2, wherein the alcohol consumer data reconstruction is to sequentially arrange the attribute dimension and the time dimension, respectively, using the age, sex, date of birth, store registration time, store home zone and attribute data of the affiliated store, and time data of the order time, payment time and pickup time of the affiliated store of the alcohol consumer as variables, and purchase amount and purchase number in the purchase data and the behavior data of the alcohol consumer as results.
4. The method for predicting alcoholic consumption based on curvelet transformation according to claim 3, wherein the ordered arrangement of attribute dimensions is specifically as follows:
different attributes of wine consumers occupy different weights in the sorting process of the wine consumers, and sorting values of the wine consumers are defined as follows:
；
wherein,a ranking value representing wine consumers; />An attribute value representing a consumer of the wine; />Representing the alcoholic consumer attribute coefficient; />Representing the number of the attributes of the wine consumers;
the age of the wine consumer and the store to which the wine consumer belongs are taken as core attributes of the attribute dimension of the wine consumer,is changed into->；
Wherein,an age attribute representing wine consumers; />Representing store attributes to which wine consumers belong; />Representing an ordered ranking map; />Attribute coefficients representing the age of wine consumers; />Attribute coefficients representing stores to which wine consumers belong;
the attribute coefficient of the age of the wine consumer and the attribute coefficient of the store to which the wine consumer belongs are solved specifically as follows:
single attribute ordering: calculating the average wine consumption amount of wine consumption corresponding to the age attribute of the wine consumer or the store attribute of the wine consumer, and sorting according to the age attribute of the wine consumer or the average wine consumption amount of the store attribute of the wine consumer;
sorting of double attributes: calculating the age attribute of the wine consumer and the average wine consumption amount of wine consumption corresponding to the affiliated store attribute, and sorting the average wine consumption amount according to the first age attribute, the affiliated store attribute and the first affiliated store attribute and the second age attribute respectively;
comparing the reverse ordinal numbers of the average wine consumption amount of the single attribute sequencing and the double attribute sequencing, and regarding the sequencing mode with small reverse ordinal number as the optimal sequencing; thereby obtainingThe value is 1, & lt + & gt>The value is the number of the age attribute values, namely the maximum sorting value of the age attribute;
and so on,is of the value of (2)The method comprises the following steps:
when (when)When (I)>；
When (when)When (I)>；
Wherein,representing the number of consumer attributes; />Representing the maximum ranking of consumer attributes.
5. The method for predicting alcoholic consumption based on curvelet transformation according to claim 3, wherein the ordered arrangement of time dimensions is as follows:
the only time of the consumer is described as；
Wherein y is c Representing the comprehensive consumption time of the consumer; y is n Representing a consumer time parameter; b n Representing a consumer time coefficient; t represents the number of consumer times;
taking the payment time and the goods taking time as core data, setting time ordering as optimal ordering, and obtaining:；
will beReduced to time solving for the average:
；
wherein y is c Representing the comprehensive consumption time of the consumer; y is pay Representing a consumer payment time; y is get Indicating the time of pickup by the consumer.
6. The alcohol consumption prediction method based on curvelet transformation according to claim 1, wherein the data curvelet transformation is specifically as follows:
the curvelet change after data reconstruction of the data of the wine consumer is expressed as:
；
wherein,expressed in the scale +.>Direction->Position->A curvelet coefficient; />Representing consumer data in the consumer dimension +.>And time dimension->Distribution of (3)；/>Representing a curvelet function;
wine consumer data in Cartesian coordinate systemFor input, the discrete form of the curvelet transform is expressed as:
；
wherein,representing the scale +.>Direction->Position->A curvelet coefficient; />Is a discrete curvelet function; />；
Data on wine consumersPerforming discrete Fourier transform to obtain frequency domain expressed as +.>The method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>；
Resampling is performed in the frequency domain for different scales and angles, and the acquisition of a new frequency domain is expressed as:
；
wherein,；/>representing the support area of the frequency window, the support length and width are respectively +.>And->；
The set of curvelet coefficients is expressed as:
；
wherein,representing an inverse fast fourier transform; />Representing wine consumer data at consumer +.>-time ofWindow functions in coordinates;
after the data of wine consumers are converted by curvelet, the curve capable of describing the characteristics of the wine consumers is obtainedWave coefficientObtaining the numerical distribution of the curvelet coefficient set, namely the curvelet coefficient;
dividing the curvelet coefficient into a coarse scale layer, a fine scale layer and a fine scale according to the characteristic of curvelet transformation and the curvelet coefficient; wherein the coarse scale layer comprises low frequency data information corresponding to alcoholic beverage consumer data-time domain->Overall profile information of all wine consumer data in (a) describing information of a population of wine consumers; the fine scale layer comprises medium-high frequency data information and carries out multidirectional decomposition through a plurality of direction parameters, and the information corresponds to different wine consumers +.>Different times->Fine features in the direction, describing individual information of wine consumers; the fine scale includes high frequency data information corresponding to individual random purchase information in the wine consumer data, which is considered as invalid interference information during the wine consumer prediction process.
7. The method for predicting alcoholic beverage consumption based on curvelet transformation according to claim 6, wherein the suppression of the interference information is specifically as follows:
a curved wave coefficient threshold is set for high frequency data information in curved wave coefficients corresponding to wine consumer data of individual purchasing behavior, and the high frequency data with the frequency greater than the curved wave coefficient threshold is assigned as the curved wave coefficient threshold during calculation.
8. The method for predicting alcoholic beverage consumption based on curvelet transformation according to claim 6, wherein the feature information extraction is specifically as follows:
and performing inverse curvelet transformation on the low-frequency data and the medium-low frequency data in the curvelet coefficient to extract consumer group description data with high purchase possibility, wherein the formula is as follows:
；
wherein,for describing consumption by a consumer group; />Representing inverse curvelet transformation;
and carrying out inverse curvelet transformation on medium-high frequency data in the curvelet coefficient to extract a consumer with high purchase possibility, wherein the formula is as follows:
；
wherein,for describing the consumption of the individual consumer; />Representing inverse curvelet transformation;
the wine consumer purchase forecast is specifically as follows:
taking the data of the group description as the coefficient of the consumer individual, and acquiring a prediction result of the consumer individual based on the consumer group description, wherein the formula is as follows:
；
wherein,indicating a prediction of the likelihood of purchase for the wine consumer, a larger value indicates a higher likelihood of purchase.
9. An electronic device, comprising: a memory and at least one processor;
wherein the memory has a computer program stored thereon;
the at least one processor executing the computer program stored by the memory causes the at least one processor to perform the curvelet transform-based wine consumption prediction method according to any one of claims 1 to 8.
10. A computer readable storage medium having stored therein a computer program executable by a processor to implement the curvelet transformation-based wine consumption prediction method according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410186242.8A CN117745338B (en) | 2024-02-20 | 2024-02-20 | Wine consumption prediction method based on curvelet transformation, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410186242.8A CN117745338B (en) | 2024-02-20 | 2024-02-20 | Wine consumption prediction method based on curvelet transformation, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117745338A true CN117745338A (en) | 2024-03-22 |
| CN117745338B CN117745338B (en) | 2024-05-03 |
Family
ID=90251191
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410186242.8A Active CN117745338B (en) | 2024-02-20 | 2024-02-20 | Wine consumption prediction method based on curvelet transformation, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117745338B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20050122164A (en) * | 2004-06-23 | 2005-12-28 | 삼성전자주식회사 | Method and apparatus for deblocking using edge flow-directed filter and curvelet transform |
| CN102270279A (en) * | 2011-07-27 | 2011-12-07 | 华北电力大学 | Short-term power load predicting method |
| KR20170017378A (en) * | 2015-08-06 | 2017-02-15 | 주식회사 메디비앙 | System and Method for Matching of Buyers and Sellers using Purchase History Data |
| US20170148081A1 (en) * | 2015-11-20 | 2017-05-25 | Mastercard International Incorporated | Method and system for recommending relevant merchants for a consumer at a given geographical location by evaluating the strength of the intersect between consumer vectors and merchant vectors |
| KR20210066514A (en) * | 2019-11-28 | 2021-06-07 | (주)양유 | Marketing solution system that can analyze and manage the impact of the influencer and Marketing solution method using the same |
| CN113435936A (en) * | 2021-07-02 | 2021-09-24 | 深圳市好伙计科技有限公司 | Online and offline consumer goods recommendation method and system based on big data |
| CN113469722A (en) * | 2020-03-31 | 2021-10-01 | 广达电脑股份有限公司 | Consumption prediction system and consumption prediction method |
| CN117391820A (en) * | 2023-12-01 | 2024-01-12 | 深圳市思迅网络科技有限公司 | SaaS service comprehensive management method and system |
-
2024
- 2024-02-20 CN CN202410186242.8A patent/CN117745338B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20050122164A (en) * | 2004-06-23 | 2005-12-28 | 삼성전자주식회사 | Method and apparatus for deblocking using edge flow-directed filter and curvelet transform |
| CN102270279A (en) * | 2011-07-27 | 2011-12-07 | 华北电力大学 | Short-term power load predicting method |
| KR20170017378A (en) * | 2015-08-06 | 2017-02-15 | 주식회사 메디비앙 | System and Method for Matching of Buyers and Sellers using Purchase History Data |
| US20170148081A1 (en) * | 2015-11-20 | 2017-05-25 | Mastercard International Incorporated | Method and system for recommending relevant merchants for a consumer at a given geographical location by evaluating the strength of the intersect between consumer vectors and merchant vectors |
| KR20210066514A (en) * | 2019-11-28 | 2021-06-07 | (주)양유 | Marketing solution system that can analyze and manage the impact of the influencer and Marketing solution method using the same |
| CN113469722A (en) * | 2020-03-31 | 2021-10-01 | 广达电脑股份有限公司 | Consumption prediction system and consumption prediction method |
| CN113435936A (en) * | 2021-07-02 | 2021-09-24 | 深圳市好伙计科技有限公司 | Online and offline consumer goods recommendation method and system based on big data |
| CN117391820A (en) * | 2023-12-01 | 2024-01-12 | 深圳市思迅网络科技有限公司 | SaaS service comprehensive management method and system |
Non-Patent Citations (2)
| Title |
|---|
| 杜吉梁;: "基于大数据和层次分析法的商场会员购买力评价模型", 信息技术与信息化, no. 05, 25 May 2019 (2019-05-25) * |
| 董烈乾;李培明;张奎;曾宪龙;雷连民;祝杨;: "利用曲波变换预测多次波模型", 石油地球物理勘探, no. 06, 15 December 2015 (2015-12-15) * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117745338B (en) | 2024-05-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Enders et al. | Threshold cointegration and purchasing power parity in the pacific nations | |
| CN107730389A (en) | Electronic installation, insurance products recommend method and computer-readable recording medium | |
| Brezigar-Masten et al. | CART-based selection of bankruptcy predictors for the logit model | |
| CN108648074A (en) | Loan valuation method, apparatus based on support vector machines and equipment | |
| CN107689008A (en) | A kind of user insures the method and device of behavior prediction | |
| Behar et al. | Exports and international logistics | |
| CN103927615B (en) | Entity is associated with classification | |
| Chen | The gamma CUSUM chart method for online customer churn prediction | |
| CN111401777A (en) | Enterprise risk assessment method and device, terminal equipment and storage medium | |
| CN112148973B (en) | Data processing method and device for information push | |
| CN112001754A (en) | User portrait generation method, device, equipment and computer readable medium | |
| Giesecke et al. | A general framework for measuring VAT compliance rates | |
| JP5914549B2 (en) | Information processing apparatus and information analysis method | |
| Hung et al. | A modified Lotka–Volterra model for competition forecasting in Taiwan’s retail industry | |
| JP2015043167A (en) | Sales prediction system and method | |
| Chit | Exchange rate volatility and exports: Evidence from the ASEAN-China Free Trade Area | |
| CN111179051A (en) | Financial target customer determination method and device and electronic equipment | |
| CN114266640A (en) | Auditing method and device, computer equipment and storage medium | |
| Jain et al. | NFT Appraisal Prediction: Utilizing Search Trends, Public Market Data, Linear Regression and Recurrent Neural Networks | |
| Zhang et al. | How can dmall do better? discussion on new retail marketing mode based on 4Cs theory | |
| CN112435068A (en) | Malicious order identification method and device, electronic equipment and storage medium | |
| CN117745338B (en) | Wine consumption prediction method based on curvelet transformation, electronic equipment and storage medium | |
| CN109636627B (en) | Insurance product management method, device, medium and electronic equipment based on block chain | |
| CN110851708A (en) | Negative sample extraction method and device, computer equipment and storage medium | |
| CN114625975B (en) | Knowledge graph-based customer behavior analysis 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |