CN117709591A - Intelligent analysis method for economic data - Google Patents
Intelligent analysis method for economic data Download PDFInfo
- Publication number
- CN117709591A CN117709591A CN202311726965.4A CN202311726965A CN117709591A CN 117709591 A CN117709591 A CN 117709591A CN 202311726965 A CN202311726965 A CN 202311726965A CN 117709591 A CN117709591 A CN 117709591A
- Authority
- CN
- China
- Prior art keywords
- commodity
- data
- characteristic
- set time
- value
- 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
- 238000004458 analytical method Methods 0.000 title claims abstract description 28
- 239000011159 matrix material Substances 0.000 claims description 51
- 238000000034 method Methods 0.000 claims description 19
- 238000012512 characterization method Methods 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 5
- 238000012216 screening Methods 0.000 claims description 5
- 230000008901 benefit Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013523 data management Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the technical field of business data prediction, in particular to an intelligent analysis method for economic data, which comprises the following steps: acquiring at least two commodity data of each commodity in a commodity field ledger in each set time period; obtaining a profit representation value of each commodity in each set time period according to each commodity data of each commodity in each set time period; according to commodity data and the income representation value, a data characteristic value is obtained; dividing all the set time periods according to the fluctuation condition of the data characteristic values to obtain a characteristic time set, and further determining a preferable time set; and carrying out data prediction by utilizing the data characteristic values of each commodity data in each set time period in the optimized time set to obtain the predicted characteristic values of each commodity data, and determining the purchase quantity of each commodity according to the predicted characteristic values. According to the invention, the data characteristic values in the optimal time set are utilized to conduct data prediction, so that the difficulty of data prediction is reduced.
Description
Technical Field
The invention relates to the technical field of business data prediction, in particular to an intelligent analysis method for economic data.
Background
With the rapid development and wide application of the digitization age, various industries produce a large amount of digitized data. In the economic field, a large amount of economic data such as market trade data, enterprise management data and the like are accumulated, and how to efficiently manage and analyze a large amount of economic data, so that it is important to mine the hidden value. Based on the method, mass economic data are intelligently analyzed to accurately predict commodity transaction data in enterprises. However, the commodity transaction data of enterprises are more in commodity variety, the commodity transaction data are more in commodity variety, the transaction data amount of each commodity in the historical data is huge, the data dimension of the data prediction of each commodity is higher by the existing prediction method, and different sales prediction models are required to be built for the historical sales data of each commodity to predict the sales data, so that the difficulty of data prediction is higher.
Disclosure of Invention
In order to solve the technical problem of high difficulty in data prediction caused by high commodity data dimension, the invention aims to provide an economic data intelligent analysis method, which adopts the following technical scheme:
acquiring at least two commodity data of each commodity in a commodity field ledger in each set time period;
carrying out profit analysis according to each commodity data of each commodity in each set time period to obtain a profit representation value of each commodity in each set time period; obtaining the data characteristic value of each commodity data in each set time period according to the commodity data and the income characteristic value of each commodity in each set time period;
dividing all set time periods according to the fluctuation condition of the data characteristic values to obtain a characteristic time set; determining a preferable time set according to the difference condition between the data characteristic value of each commodity data in each set time period in each characteristic time set and the data characteristic value of each commodity data in all set time periods;
and carrying out data prediction by utilizing the data characteristic values of each commodity data in each set time period in the optimized time set to obtain the predicted characteristic values of each commodity data, and determining the purchase quantity of each commodity according to the predicted characteristic values.
Preferably, the commodity data includes: purchase price, sales price, purchase quantity, and sales quantity.
Preferably, the obtaining the data characteristic value of each commodity data in each set time period according to the commodity data and the profit characterizing value of each commodity in each set time period specifically includes:
for any set time period, constructing a commodity data matrix based on each commodity data of each commodity; each row of elements in the commodity data matrix is respectively the purchase price, the sale price, the purchase quantity and the sale quantity of each commodity, and the number of rows of the commodity data matrix is equal to the commodity type;
constructing a profit data column matrix based on the profit characterization value of each commodity, wherein the number of lines of the profit data column matrix is equal to the commodity type; calculating the product of the transposed matrix of the commodity data matrix and the profit data column matrix to obtain a data characteristic value column matrix; each row of elements in the data characteristic value column matrix corresponds to the data characteristic value of each commodity data.
Preferably, the obtaining the profit representation value of each commodity in each set time period by carrying out the profit analysis according to each commodity data of each commodity in each set time period specifically includes:
for any commodity in any set time period, calculating the profit coefficient of the commodity according to the purchase price, the sale price and the sale quantity of the commodity; calculating the sales coefficients of the commodities according to the purchase quantity and the sales quantity of the commodities; and taking the normalized value of the product of the profit coefficient and the sales coefficient as the profit characterization value of the commodity in the set time period.
Preferably, the calculating the profit coefficient of the commodity according to the purchase price, the sale price and the sale quantity of the commodity specifically includes: and calculating the difference between the selling price and the purchasing price of the commodity, and taking the product of the difference and the selling quantity as the profit coefficient of the commodity.
Preferably, the calculating the sales coefficient of the commodity according to the purchase quantity and the sales quantity of the commodity specifically includes: and taking the ratio of the selling quantity and the purchasing quantity of the commodity as the selling coefficient of the commodity.
Preferably, the determining the preferred time set according to the difference between the data characteristic value of each commodity data in each set time period and the data characteristic value of each commodity data in all set time periods specifically includes:
recording any commodity data as target commodity data, calculating the average value of the data characteristic values of the target commodity data in all set time periods, and recording the average value as the characteristic average value of the target commodity data;
for any one of the characteristic time sets, calculating the square of the difference between the data characteristic value of the target commodity data in each set time period and the characteristic average value, and marking the square as the difference coefficient of the target commodity data in each set time period in the characteristic time set; carrying out negative correlation normalization processing on the average value of the difference coefficients of the target commodity data in all the set time periods in the time feature set to obtain the association degree of the target commodity data in the time feature set;
and screening all the time feature sets according to the association degree of each commodity data to obtain a preferred time set.
Preferably, the screening of the feature sets of all times according to the association degree of each commodity data to obtain a preferred time set specifically includes:
and recording a time feature set corresponding to any commodity data with the association degree larger than a preset association threshold value as a preferable time set.
Preferably, the dividing all the set time periods according to the fluctuation condition of the data characteristic value to obtain a characteristic time set specifically includes:
dividing two adjacent set time periods, of which the difference between the data characteristic values of the commodity data of the same kind meets the preset condition, into the same characteristic time set;
the preset conditions are as follows: for any commodity data, the normalized value of the absolute value of the difference between the data characteristic values of the commodity data in two adjacent set time periods is smaller than or equal to a preset characteristic threshold.
Preferably, the determining the purchase quantity of each commodity according to the prediction characteristic value specifically includes:
for any commodity, recording the ratio between the predicted characteristic value of the sales quantity and the average value of the data characteristic values of the sales quantity of the commodity in all the set time periods in the optimized time set as the adjustment coefficient of the sales quantity;
and (3) upwardly rounding the product between the adjustment coefficient and the average value of the sales quantity of the commodity in all the set time periods in the optimal time set to obtain the predicted purchase quantity of the commodity.
The embodiment of the invention has at least the following beneficial effects:
the method comprises the steps of firstly obtaining at least two commodity data of each commodity in a commodity field ledger in each set time period, so that the commodity and the commodity data are respectively analyzed by taking the set time period as a time unit to obtain prediction data. Then, carrying out profit analysis on each commodity data to obtain profit characterization values of each commodity in each set time period, namely integrating attribute data of a plurality of dimensions of each commodity in a time unit, and obtaining profit information of each commodity; and then, carrying out dimension reduction processing on various commodities in a time unit by utilizing the commodity data and the income representation value, so that the data characteristic value of each commodity data in each set time period can be obtained, and the characteristic information corresponding to the commodity data is represented by utilizing the data characteristic value. Further, the fluctuation condition of the data characteristic value in each time unit is analyzed, the set time period is screened and divided, so that a preferable time set with smaller overall difference can be screened out later, the preferable time set is used for data prediction, the data characteristic value in the preferable time set is used for data prediction, the data dimension is lower, the difficulty of data prediction is reduced, the purchasing data effect of each commodity is finally determined, and the method has higher reference value.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a method of the present invention for intelligent analysis of economic data.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following is a detailed description of specific implementation, structure, characteristics and effects of an intelligent analysis method for economic data according to the invention with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following specifically describes a specific scheme of the intelligent analysis method for economic data provided by the invention with reference to the accompanying drawings.
Referring to fig. 1, a method flowchart of an intelligent analysis method for economic data according to an embodiment of the invention is shown, and the method includes the following steps:
step one, acquiring at least two commodity data of each commodity in a commodity field ledger in each set time period.
In this embodiment, the method for extracting commodity transaction data in the standing book data is known technology, and will not be described too much.
The commodity data represents various attribute data of corresponding commodities, and the various commodity data of each commodity represent multiple attribute information of corresponding same commodity, namely, each commodity has attribute data of multiple dimensions, in this embodiment, transaction information of the commodity is analyzed, so four commodity data are collected in total, and the method respectively comprises the following steps: purchase price, sales price, purchase quantity, and sales quantity. Meanwhile, in this embodiment, the time length of each set time period is set to 1 day, commodity data in 30 set time periods is collected in total, and the implementer can set according to the specific implementation scenario.
Step two, carrying out profit analysis according to each commodity data of each commodity in each set time period to obtain a profit representation value of each commodity in each set time period; and obtaining the data characteristic value of each commodity data in each set time period according to the commodity data and the income characteristic value of each commodity in each set time period.
Because the various commodity data of each commodity respectively represent different attribute information of the commodity, and further each commodity has attribute data of multiple dimensions, the difficulty of predicting transaction data aiming at the commodity of multiple dimensions is high, and in the embodiment, the data dimension reduction operation based on the benefit data is performed by performing benefit analysis on the commodity data of each commodity, so that the subsequent prediction on the transaction information of the commodity is simpler.
Firstly, carrying out profit analysis according to each commodity data of each commodity in each set time period to obtain the profit representation value of each commodity in each set time period. Specifically, for any commodity in any set time period, calculating the profit coefficient of the commodity according to the purchase price, the sale price and the sale quantity of the commodity, namely calculating the difference value between the sale price and the purchase price of the commodity, and taking the product of the difference value and the sale quantity as the profit coefficient of the commodity. Calculating the sales coefficient of the commodity according to the purchase quantity and the sales quantity of the commodity, and taking the ratio of the sales quantity and the purchase quantity of the commodity as the sales coefficient of the commodity; and taking the normalized value of the product of the profit coefficient and the sales coefficient as the profit characterization value of the commodity in the set time period.
In this embodiment, taking the nth set period and the type a commodity as an example for explanation, the calculation formula of the profit characterizing value of the type a commodity in the nth set period may be expressed as:
wherein Q is n,A Representing the income representation value, RL, of the A-th commodity in the nth set time period n,A Represents the selling price of the A-th commodity in the nth set time period, R n,A Indicating the purchase price, NL, of the A-th commodity in the nth set time period n,A Indicating the sales quantity of the A-th commodity in the nth set time period, N n,A Representing the purchase quantity of the A-th commodity in the nth set time period, and Norm () represents a normalization function, in this embodiment, a maximum value and minimum value normalization method is adopted, and an implementer can set according to a specific implementation scenario.
(RL n,A -R n,A )*NL n,A And representing commodity sales profits of the A-th commodity in the nth set time period by using the profits coefficient, wherein the larger the value of the profits coefficient is, the larger the profits of the commodity sold in the nth set time period is.
And representing the commodity sales rate of the A-th commodity in the nth set time period by using the sales coefficient for the A-th commodity in the nth set time period, wherein the larger the sales coefficient is, the larger the commodity sales amount in the set time period is. The profit representation value of the commodity reflects the profit information of the commodity in the set time period.
Then, each commodity has a corresponding profit representation value in each set time period, meanwhile, each commodity has corresponding multiple commodity data in each set time period, the profit representation value reflects the profit information of the commodity, the commodity information of the commodity has four kinds in the same set time period, and in order to achieve the purpose of data dimension reduction, in this embodiment, the data characteristic value of each commodity data in each set time period is obtained according to the commodity data and the profit representation value of each commodity in each set time period.
Specifically, for any one set period of time, a commodity data matrix is constructed based on each commodity data of each commodity; and each row of elements in the commodity data matrix are respectively the purchase price, the sale price, the purchase quantity and the sale quantity of each commodity, and the number of rows of the commodity data matrix is equal to the commodity type.
In this embodiment, taking the nth set period as an example, the commodity data matrix of the nth set period may be expressed as:
wherein RL is a n,1 、R n,1 、NL n,1 And N n,1 Respectively representing the selling price, the purchasing price, the selling quantity and the purchasing quantity of the first commodity in the nth set time period, RL n,M 、R n,M 、NL n,M And N n,M The selling price, the purchasing price, the selling quantity and the purchasing quantity of the Mth commodity in the nth set time period are respectively represented. In this example, there are M kinds of commodity.
For any set time period, a profit data column matrix is constructed based on the profit characterizing value of each commodity, the number of lines of the profit data column matrix is equal to the commodity type, and in this embodiment, the description is given by taking the nth set time period as an example, the profit data column matrix of the nth set time period may be expressed as:
wherein Q is n,1 Representing a profit indicative value, Q, of the first commodity within the nth set period of time n,M Representing the return of the Mth commodity in the nth set time periodAnd (5) characterizing the value.
Further, the product of the transposed matrix of the commodity data matrix and the profit data column matrix is calculated to obtain a data characteristic value column matrix, and the column number of the transposed matrix of the commodity data matrix is equal to the column number of the profit data column matrix, so that the data characteristic value column matrix can be obtained by calculating the product of the transposed matrix of the commodity data matrix and the profit data column matrix, the column number of the data characteristic value column matrix is equal to the column number of the commodity data matrix, and the column number of the data characteristic value column matrix is equal to the column number of the profit data column matrix. Each row of elements in the data characteristic value column matrix corresponds to the data characteristic value of each commodity data.
In this embodiment, taking the nth set period as an example, the data eigenvalue column matrix of the nth set period may be expressed as:
wherein W is n.RL Data characteristic value representing selling price of nth set time period, W n.R Data characteristic value representing purchase price of nth set time period, W n.NL Data characteristic value, W, representing sales quantity for the nth set period of time n.N A data characteristic value indicating the purchase quantity of the nth set period.
The product of the transposed matrix of the commodity data matrix and the profit data array matrix is calculated, namely, in a set time period, the profit representation value of each commodity is taken as the weight corresponding to each commodity, and under the same commodity data, the commodity data of all commodities are weighted and summed by the weight, so that the data characteristic value of the commodity data in the set time period is obtained.
Specifically, taking the purchase price as an example, taking the profit representation value of each commodity as the weight corresponding to the commodity in the nth set time period, and carrying out weighted summation on the purchase prices of all commodities by using the weight to obtain the data characteristic value of the purchase price in the nth set time period. The data characteristic value characterizes the overall characteristic attribute of all commodities in a set time period of commodity data.
For any set time period, the commodity data matrix reflects various commodity attributes of various commodities, the data volume is large, the dimension is high, the obtained data characteristic value column matrix is obtained after the commodity data matrix is processed by utilizing the benefit data column matrix, the multidimensional data is reduced to be a one-dimensional array, each commodity data corresponds to one data characteristic value, and the overall characteristics of the commodity data of the corresponding type are respectively represented.
Dividing all set time periods according to the fluctuation condition of the data characteristic values to obtain a characteristic time set; and determining a preferable time set according to the difference condition between the data characteristic value of each commodity data in each set time period in each characteristic time set and the data characteristic value of each commodity data in all set time periods.
After the dimension reduction processing is carried out on the commodity data in each set time period, four different commodity data in each set time period respectively correspond to one data characteristic value, and then the data characteristic value of the commodity data in each set time period is utilized to screen each set time period so as to obtain better data to participate in the data prediction operation.
For any commodity data, each data characteristic value reflects the integral characteristic of the commodity data in one set time period, a certain time characteristic is provided between different set time periods, the time correlation between adjacent set time periods is strong, and then all set time periods can be divided based on the difference condition between the integral characteristics of the commodity data in the adjacent set time periods.
And dividing all the set time periods according to the fluctuation condition of the data characteristic values to obtain a characteristic time set. Specifically, two adjacent set time periods in which the difference between the data characteristic values of the same kind of commodity data satisfies a preset condition are divided into the same characteristic time set; the preset conditions are as follows: for any commodity data, the normalized value of the absolute value of the difference between the data characteristic values of the commodity data in two adjacent set time periods is smaller than or equal to a preset characteristic threshold.
For any two adjacent set time periods, respectively calculating the corresponding difference between each commodity data in the two set time periods, namely calculating the normalized value of the absolute value of the difference between the selling prices in the two set time periods to obtain a first characteristic difference; calculating the normalized value of the absolute value of the difference between the purchase prices in the two set time periods to obtain a second characteristic difference, calculating the normalized value of the absolute value of the difference between the sales quantities in the two set time periods to obtain a third characteristic difference, calculating the normalized value of the absolute value of the difference between the purchase quantities in the two set time periods to obtain a fourth characteristic difference, and dividing the two adjacent set time periods into the same characteristic time set once any characteristic difference value is smaller than or equal to a characteristic threshold value, which indicates that the characteristic distribution of the data in the two adjacent set time periods is similar. In this embodiment, the value of the feature threshold is 0.3, and the implementer can set according to the specific implementation scenario.
In the same characteristic time set, the set time periods are adjacent in time and similar in data. Based on this, in each local time range, the characteristic information of each commodity data should be relatively similar to the characteristic information of the commodity data in the total time length, that is, the characteristic information has relatively high correlation between the characteristic information and the characteristic information, and the characteristic trend of the meeting part and the characteristic trend of the total commodity data are relatively similar. When the characteristic information of commodity data and the characteristic information of commodity data in the total time length are greatly different in the local time range, abnormal conditions may exist in the characteristic information in the local time range, so that the characteristic data in the local time range is not adopted when the data are predicted.
Based on the difference, the characteristic time sets are screened according to the difference between the data characteristic value of each commodity data in each set time period in each characteristic time set and the data characteristic value of each commodity data in all set time periods, and the optimal time set is determined.
Specifically, any commodity data is recorded as target commodity data, the average value of the data characteristic values of the target commodity data in all set time periods is calculated, and the average value is recorded as the characteristic average value of the target commodity data; for any one of the characteristic time sets, calculating the square of the difference between the data characteristic value of the target commodity data in each set time period and the characteristic average value, and marking the square as the difference coefficient of the target commodity data in each set time period in the characteristic time set; and carrying out negative correlation normalization processing on the average value of the difference coefficients of the target commodity data in all the set time periods in the time feature set to obtain the association degree of the target commodity data in the time feature set.
In this embodiment, taking the purchase price as the target commodity data and taking the mth feature time set as an example for explanation, the calculation formula of the association degree of the target commodity data in the mth feature time set may be expressed as:
wherein L is m,R Representing the association degree of the purchase price in the mth characteristic time set,data characteristic value representing purchase price in the ith set period of time in the mth characteristic time set,/for the data characteristic value>Mean value of data characteristic values representing purchase prices in all set time periods, namely characteristic mean value of purchase prices, S m Representing the total number of set time periods contained in the mth feature time set, exp () represents an exponential function based on a natural constant e.
For the coefficient of difference of the purchase price of the ith set time period in the mth characteristic time set, reflecting the data between the purchase price of the ith set time period and the purchase price of the whole time lengthThe larger the difference of the characteristic values, the larger the value of the characteristic values, which shows that the larger the difference of the data characteristics and the whole in the ith set time period, the smaller the value of the association degree of the corresponding purchase price.
The average value of the difference coefficients of all the set time periods in the characteristic time set is calculated, so that the balance condition of the difference degree in the characteristic time set is reflected, and the larger the overall difference is, the smaller the association degree between the corresponding data characteristics in the characteristic time set and the overall is. The association degree of the target commodity data in the characteristic time set reflects the difference condition and the association degree of the data characteristic of the target commodity data and the whole in the local time range.
Further, screening all time feature sets according to the association degree of each commodity data to obtain a preferred time set. Namely, the time feature set corresponding to the correlation degree of any commodity data being larger than the preset correlation threshold value is recorded as the preferable time set.
Specifically, for any one feature time set, each commodity data calculates the corresponding association degree, and when the association degree of the commodity data is larger than the association threshold value, the stronger the association between the data features and the population of the commodity data in the feature time set is, the more suitable the commodity data is as history data for data prediction. When the association degree of the commodity data is smaller than or equal to the association threshold value, the weaker the association between the data features of the commodity data and the overall in the feature time set is, the larger the difference situation possibly exists, so that the prediction operation of the data is not facilitated.
In this embodiment, if the association degree of one commodity data in any one of the feature time sets satisfies the association threshold, the feature time set is screened out and is used as the preferred time set. Meanwhile, the value of the association threshold is set to be 0.7, and an implementer can set according to a specific implementation scene.
And fourthly, carrying out data prediction by utilizing the data characteristic values of each commodity data in each set time period in the optimized time set to obtain the predicted characteristic values of each commodity data, and determining the purchase quantity of each commodity according to the predicted characteristic values.
And the data characteristic value of each commodity data in each set time period in the preferred time set is utilized to perform data prediction.
In this embodiment, since the data feature values of each commodity data have a strong time sequence, the data feature values of each commodity data can be predicted by using a time sequence prediction method, for example, an exponential smoothing algorithm, etc., and an implementer can select according to a specific implementation scenario. It should be noted that, the data characteristic values of each commodity data in the set time period in all the preferred time sets are respectively constructed into a time sequence, that is, four time sequence sequences corresponding to four commodity data, and then the time sequence sequences are used as historical data to respectively predict the data characteristic values of each commodity data, so as to obtain the predicted characteristic values corresponding to each commodity data.
Finally, instructive decisions may be made on subsequent merchandise purchase schemes or fund preparation, etc., based on the predicted characteristic values of each merchandise data obtained. And determining the purchase quantity of each commodity according to the prediction characteristic value. Specifically, for any commodity, the ratio between the predicted characteristic value of the sales quantity and the average value of the data characteristic values of the sales quantity of the commodity in all the set time periods in the preferred time set is recorded as the adjustment coefficient of the sales quantity. And (3) upwardly rounding the product between the adjustment coefficient and the average value of the sales quantity of the commodity in all the set time periods in the optimal time set to obtain the predicted purchase quantity of the commodity.
And comparing the obtained predicted characteristic value with the data characteristic value in the historical data, and adaptively adjusting the purchase quantity of the commodity according to the comparison result, wherein the purchase quantity of the commodity is properly increased when the predicted characteristic value is larger than the historical data. When the predicted characteristic value is small compared with the history data, the purchase quantity of the commodity should be appropriately reduced.
According to the method, the predicted purchase quantity of each commodity can be obtained, the predicted purchase quantity of each commodity can be used as a reference purchase scheme, and an implementer can obtain the actual purchase quantity according to a specific implementation scene.
According to the invention, the multi-dimensional prediction process of the multi-attribute data of various commodities is converted into the respective prediction process of the multi-attribute data, so that the dimension of data prediction is reduced to a certain extent, the difficulty of data prediction is reduced, the workload of economic data management is reduced, the working efficiency is further improved, and the intelligent analysis of the economic data is completed.
The above embodiments are only for illustrating the technical solution of the present application, and are not limiting; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the scope of the embodiments of the present application, and are intended to be included within the scope of the present application.
Claims (10)
1. An intelligent analysis method for economic data is characterized by comprising the following steps:
acquiring at least two commodity data of each commodity in a commodity field ledger in each set time period;
carrying out profit analysis according to each commodity data of each commodity in each set time period to obtain a profit representation value of each commodity in each set time period; obtaining the data characteristic value of each commodity data in each set time period according to the commodity data and the income characteristic value of each commodity in each set time period;
dividing all set time periods according to the fluctuation condition of the data characteristic values to obtain a characteristic time set; determining a preferable time set according to the difference condition between the data characteristic value of each commodity data in each set time period in each characteristic time set and the data characteristic value of each commodity data in all set time periods;
and carrying out data prediction by utilizing the data characteristic values of each commodity data in each set time period in the optimized time set to obtain the predicted characteristic values of each commodity data, and determining the purchase quantity of each commodity according to the predicted characteristic values.
2. The intelligent analysis method of economic data according to claim 1, wherein the commodity data comprises: purchase price, sales price, purchase quantity, and sales quantity.
3. The method for intelligent analysis of economic data according to claim 2, wherein the obtaining the data characteristic value of each commodity data in each set time period according to the commodity data and the profit characteristic value of each commodity in each set time period specifically comprises:
for any set time period, constructing a commodity data matrix based on each commodity data of each commodity; each row of elements in the commodity data matrix is respectively the purchase price, the sale price, the purchase quantity and the sale quantity of each commodity, and the number of rows of the commodity data matrix is equal to the commodity type;
constructing a profit data column matrix based on the profit characterization value of each commodity, wherein the number of lines of the profit data column matrix is equal to the commodity type; calculating the product of the transposed matrix of the commodity data matrix and the profit data column matrix to obtain a data characteristic value column matrix; each row of elements in the data characteristic value column matrix corresponds to the data characteristic value of each commodity data.
4. The method for intelligent analysis of economic data according to claim 2, wherein the step of obtaining the profit representation value of each commodity in each set time period by performing profit analysis according to each commodity data of each commodity in each set time period specifically comprises the following steps:
for any commodity in any set time period, calculating the profit coefficient of the commodity according to the purchase price, the sale price and the sale quantity of the commodity; calculating the sales coefficients of the commodities according to the purchase quantity and the sales quantity of the commodities; and taking the normalized value of the product of the profit coefficient and the sales coefficient as the profit characterization value of the commodity in the set time period.
5. The intelligent analysis method for economic data according to claim 4, wherein the calculating the profit coefficient of the commodity according to the purchase price, the sale price and the sale quantity of the commodity comprises:
and calculating the difference between the selling price and the purchasing price of the commodity, and taking the product of the difference and the selling quantity as the profit coefficient of the commodity.
6. The intelligent analysis method for economic data according to claim 4, wherein the calculating the sales coefficient of the commodity according to the purchase quantity and the sales quantity of the commodity comprises: and taking the ratio of the selling quantity and the purchasing quantity of the commodity as the selling coefficient of the commodity.
7. The method for intelligent analysis of economic data according to claim 1, wherein the determining the preferred time set according to the difference between the data characteristic value of each commodity data in each set time period and the data characteristic value of each commodity data in all set time periods comprises:
recording any commodity data as target commodity data, calculating the average value of the data characteristic values of the target commodity data in all set time periods, and recording the average value as the characteristic average value of the target commodity data;
for any one of the characteristic time sets, calculating the square of the difference between the data characteristic value of the target commodity data in each set time period and the characteristic average value, and marking the square as the difference coefficient of the target commodity data in each set time period in the characteristic time set; carrying out negative correlation normalization processing on the average value of the difference coefficients of the target commodity data in all the set time periods in the time feature set to obtain the association degree of the target commodity data in the time feature set;
and screening all the time feature sets according to the association degree of each commodity data to obtain a preferred time set.
8. The method for intelligent analysis of economic data according to claim 7, wherein the screening of all time feature sets according to the association degree of each commodity data to obtain a preferred time set specifically comprises:
and recording a time feature set corresponding to any commodity data with the association degree larger than a preset association threshold value as a preferable time set.
9. The method for intelligent analysis of economic data according to claim 1, wherein the dividing all the set time periods according to the fluctuation condition of the characteristic data value to obtain the characteristic time set specifically comprises:
dividing two adjacent set time periods, of which the difference between the data characteristic values of the commodity data of the same kind meets the preset condition, into the same characteristic time set;
the preset conditions are as follows: for any commodity data, the normalized value of the absolute value of the difference between the data characteristic values of the commodity data in two adjacent set time periods is smaller than or equal to a preset characteristic threshold.
10. The method for intelligent analysis of economic data according to claim 2, wherein the determining the purchase quantity of each commodity according to the predicted characteristic value specifically comprises:
for any commodity, recording the ratio between the predicted characteristic value of the sales quantity and the average value of the data characteristic values of the sales quantity of the commodity in all the set time periods in the optimized time set as the adjustment coefficient of the sales quantity;
and (3) upwardly rounding the product between the adjustment coefficient and the average value of the sales quantity of the commodity in all the set time periods in the optimal time set to obtain the predicted purchase quantity of the commodity.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311726965.4A CN117709591A (en) | 2023-12-15 | 2023-12-15 | Intelligent analysis method for economic data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311726965.4A CN117709591A (en) | 2023-12-15 | 2023-12-15 | Intelligent analysis method for economic data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117709591A true CN117709591A (en) | 2024-03-15 |
Family
ID=90143828
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311726965.4A Pending CN117709591A (en) | 2023-12-15 | 2023-12-15 | Intelligent analysis method for economic data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117709591A (en) |
-
2023
- 2023-12-15 CN CN202311726965.4A patent/CN117709591A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CA2366782C (en) | Distributed hierarchical evolutionary modeling and visualization of empirical data | |
| CA2153976C (en) | System and method for the advanced prediction of weather impact on managerial planning applications | |
| US20030200134A1 (en) | System and method for large-scale automatic forecasting | |
| CN107563645A (en) | A kind of Financial Risk Analysis method based on big data | |
| KR20010102452A (en) | Methods and systems for finding value and reducing risk | |
| US20080195431A1 (en) | System and method for correlating business transformation metrics with sustained business performance | |
| CN112700324A (en) | User loan default prediction method based on combination of Catboost and restricted Boltzmann machine | |
| CN116739217A (en) | Retail management method and system based on supply chain big data platform | |
| CN102165442A (en) | Methods and apparatus to calibrate a choice forecasting system for use in market share forecasting | |
| CN113450141B (en) | Intelligent prediction method and device based on electricity sales quantity characteristics of large power customer group | |
| CN109978300B (en) | Customer risk tolerance quantification method and system, and asset configuration method and system | |
| US20240087046A1 (en) | System to automatically categorize | |
| CN114298659A (en) | Data processing method and device for evaluation object index and computer equipment | |
| CN115841345B (en) | Cross-border big data intelligent analysis method, system and storage medium | |
| CN117709591A (en) | Intelligent analysis method for economic data | |
| CN116167646A (en) | Evaluation method, device, equipment and storage medium based on transaction algorithm | |
| CN113421154B (en) | Credit risk assessment method and system based on control chart | |
| AU2004202199B2 (en) | Distributed hierarchical evolutionary modeling and visualization of empirical data | |
| CN115271442A (en) | Modeling method and system for evaluating enterprise growth based on natural language | |
| CN114757495A (en) | Membership value quantitative evaluation method based on logistic regression | |
| CN118013469B (en) | Time-dependent model analysis method for managing multidimensional data by enterprise architecture | |
| CN116664172B (en) | Food sales data mining analysis method based on big data | |
| CN117974190A (en) | Multi-commodity management decision-making method and system based on SARIMA and genetic algorithm | |
| Völcker et al. | Modelling customer lifetime value in the retail banking industry | |
| CN117350856A (en) | Investment portfolio recommendation method based on stock inference network |
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 |