CN116777345A - Stock quantity prediction method, system, device and storage medium - Google Patents
Stock quantity prediction method, system, device and storage medium Download PDFInfo
- Publication number
- CN116777345A CN116777345A CN202310722937.9A CN202310722937A CN116777345A CN 116777345 A CN116777345 A CN 116777345A CN 202310722937 A CN202310722937 A CN 202310722937A CN 116777345 A CN116777345 A CN 116777345A
- Authority
- CN
- China
- Prior art keywords
- parameter information
- data
- prediction
- prediction model
- coefficient
- 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
- 238000000034 method Methods 0.000 title claims abstract description 89
- 238000003860 storage Methods 0.000 title claims abstract description 20
- 238000010219 correlation analysis Methods 0.000 claims abstract description 58
- 238000012545 processing Methods 0.000 claims abstract description 41
- 238000003379 elimination reaction Methods 0.000 claims abstract description 24
- 230000008030 elimination Effects 0.000 claims abstract description 21
- 238000004458 analytical method Methods 0.000 claims abstract description 13
- 238000012549 training Methods 0.000 claims description 33
- 230000006870 function Effects 0.000 claims description 22
- 239000011159 matrix material Substances 0.000 claims description 19
- 238000012360 testing method Methods 0.000 claims description 14
- 238000012098 association analyses Methods 0.000 claims description 10
- 230000008859 change Effects 0.000 claims description 9
- 238000012795 verification Methods 0.000 claims description 8
- 238000010200 validation analysis Methods 0.000 claims description 2
- 230000009286 beneficial effect Effects 0.000 abstract description 5
- 238000013528 artificial neural network Methods 0.000 description 24
- 230000008569 process Effects 0.000 description 12
- 238000010586 diagram Methods 0.000 description 9
- 238000003062 neural network model Methods 0.000 description 7
- 238000007726 management method Methods 0.000 description 6
- 238000010220 Pearson correlation analysis Methods 0.000 description 5
- 238000013473 artificial intelligence Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 5
- 238000004519 manufacturing process Methods 0.000 description 5
- 239000013598 vector Substances 0.000 description 5
- 238000004891 communication Methods 0.000 description 3
- 230000000670 limiting effect Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000002829 reductive effect Effects 0.000 description 3
- 238000006467 substitution reaction Methods 0.000 description 3
- 238000003491 array Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 238000012417 linear regression Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000013178 mathematical model Methods 0.000 description 2
- 210000002569 neuron Anatomy 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000036961 partial effect Effects 0.000 description 2
- 239000002994 raw material Substances 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- 238000012384 transportation and delivery Methods 0.000 description 2
- 241000282412 Homo Species 0.000 description 1
- 241001465754 Metazoa Species 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000008713 feedback mechanism Effects 0.000 description 1
- 238000013277 forecasting method Methods 0.000 description 1
- BTCSSZJGUNDROE-UHFFFAOYSA-N gamma-aminobutyric acid Chemical compound NCCCC(O)=O BTCSSZJGUNDROE-UHFFFAOYSA-N 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 230000003278 mimic effect Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000002085 persistent effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000000714 time series forecasting Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a stock quantity prediction method, a system, a device and a storage medium. The method comprises the following steps: acquiring a plurality of parameter information related to the stock quantity; carrying out correlation analysis on the plurality of parameter information, and carrying out first rejection processing on the plurality of parameter information according to the result of the correlation analysis to obtain first data; the first elimination is used for eliminating repeated parameters in a plurality of parameter information; carrying out parameter relevance analysis on the first data to obtain a relevance coefficient; performing second elimination processing on the plurality of parameter information according to the association coefficient to obtain second data; inputting the second data into a time sequence prediction model to obtain a required stock quantity; the time series prediction model is used to characterize a model that predicts the inventory based on the time series of the relevant parameters in the second data. The embodiment of the invention can predict the stock quantity and is beneficial to improving the prediction accuracy; can be widely applied to the technical field of computers.
Description
Technical Field
The present invention relates to the field of computer technologies, and in particular, to a method, a system, an apparatus, and a storage medium for inventory prediction.
Background
Safety inventory is a buffer inventory prepared for future supply or demand uncertainty, at the expense of inventory capital occupation, in exchange for the ability to produce stability and delivery timeliness over time. If the safety stock is too large, warehouse management cost is increased, mobile funds are occupied, and finished products and raw materials are lost; if the safety stock is too small, the service level is reduced, sales profits and enterprise reputation are affected, and normal operation of the production process is affected. Therefore, a reasonably designed safety stock forecasting scheme is of great significance to the production and management of enterprises. In the related art, the secure inventory is generally predicted by a model; the most common are the autoregressive model (AR) and autoregressive moving average model (ARMA) in linear regression methods. However, the model predicts the stock quantity based on a linear mode, so that the accuracy is low, and the enterprise requirement cannot be met.
Disclosure of Invention
The present invention aims to solve at least one of the technical problems existing in the prior art to a certain extent.
Accordingly, an object of the present invention is to provide a fast and practical stock quantity prediction method, system, device and storage medium.
In order to achieve the technical purpose, the technical scheme adopted by the embodiment of the invention comprises the following steps:
in one aspect, an embodiment of the present invention provides a method for predicting an inventory, including the steps of:
the stock quantity prediction method of the embodiment of the invention comprises the following steps: acquiring a plurality of parameter information related to the stock quantity; performing correlation analysis on the plurality of parameter information, and performing first rejection processing on the plurality of parameter information according to the result of the correlation analysis to obtain first data; the first elimination is used for eliminating repeated parameters in a plurality of parameter information; carrying out parameter relevance analysis on the first data to obtain a relevance coefficient; performing second elimination processing on the plurality of parameter information according to the association coefficient to obtain second data; the association coefficient is used for representing the influence degree of each parameter information on the stock quantity; inputting the second data into a time sequence prediction model to obtain a required stock quantity; the time series prediction model is used for representing a model for predicting the stock quantity based on the time series of the related parameters in the second data. According to the embodiment of the invention, the time sequence prediction model for predicting the time sequence based on the related parameters is used for predicting the inventory, so that the influence of the time sequence on the inventory is fully considered, and the prediction accuracy of the model is improved; meanwhile, through correlation analysis and correlation analysis, data with smaller influence in parameter information is removed, and prediction accuracy of the model is improved.
In addition, the stock quantity prediction method according to the above embodiment of the present invention may further have the following additional technical features:
further, according to the stock quantity prediction method provided by the embodiment of the invention, the time sequence prediction model is obtained through training of the following steps:
obtaining a plurality of parameter samples;
performing correlation analysis on the plurality of parameter samples, and performing first rejection processing on the plurality of parameter samples according to the result of the correlation analysis to obtain a first characteristic sample;
carrying out parameter relevance analysis on the first characteristic sample to obtain a sample coefficient; performing second rejection processing on the plurality of parameter samples according to the sample coefficients to obtain second characteristic samples;
and inputting the second characteristic sample into a time sequence prediction model to obtain a prediction result, calculating a loss value by adopting a loss function according to the prediction result and a real result, and updating parameters of the time sequence prediction model according to the loss value to obtain the trained time sequence prediction model.
Further, in one embodiment of the present invention, the first feature sample includes a training sample, a validation sample, and a test sample, the method further comprising:
Inputting the training sample into a first prediction model to obtain a first result; the first prediction model is used for representing a time sequence prediction model of an open loop state;
inputting the training sample and the first result into the first prediction model, and performing closed-loop training on the first prediction model to obtain a second prediction model;
training the second prediction model through a verification sample and a real result corresponding to the verification sample to obtain a third prediction model;
and adjusting the parameters of the third prediction model through the test sample and the real result corresponding to the test sample until the prediction result reaches a set threshold value to obtain a time sequence prediction model.
Further, in an embodiment of the present invention, the time series prediction model comprises an input layer, the method further comprising the steps of:
determining the number of input nodes in the input layer according to the second characteristic sample;
determining the number of output nodes in the input layer according to the first result; wherein the number of output nodes is less than the number of input nodes.
Further, in an embodiment of the present invention, the step of performing correlation analysis on the plurality of parameter information and performing a first culling process on the plurality of parameter information according to a result of the correlation analysis includes:
Acquiring a first correlation coefficient between the first parameter information and the second parameter information; the first parameter information and the second parameter information are any two parameter information in the plurality of parameter information;
if the first correlation coefficient is a preset coefficient, eliminating the first parameter information; or if the first correlation coefficient is a preset coefficient, eliminating the second parameter information.
Further, in an embodiment of the present invention, the parameter relevance analysis is performed on the first data to obtain a relevance coefficient; and performing a second elimination process on the plurality of parameter information according to the association coefficient to obtain second data, including:
establishing an original data matrix according to the first data; the original data matrix is used for representing data based on time sequence change;
based on the data of the first time period, carrying out initial change on the original data matrix to obtain a first data matrix;
performing absolute difference processing on the first data matrix to obtain a first sequence;
carrying out association coefficient solving on the first sequence to obtain association coefficients;
based on the association coefficient, arranging the parameter information and carrying out final bit elimination processing; or, eliminating the parameter information corresponding to the association coefficient smaller than a preset threshold value.
Further, in one embodiment of the present invention, the method further comprises:
classifying the plurality of parameter information to obtain third parameter information, fourth parameter information and fifth parameter information; factors influencing the classification include enterprise aspects, market aspects and consumption aspects;
performing correlation analysis and association analysis on the third parameter information, and performing rejection processing on the third parameter information according to a preset rejection proportion;
performing correlation analysis and association analysis on the fourth parameter information, and performing rejection processing on the fourth parameter information according to a preset rejection proportion;
performing correlation analysis and association analysis on the fifth parameter information, and performing rejection processing on the fifth parameter information according to a preset rejection proportion.
In another aspect, an embodiment of the present invention provides a stock quantity prediction system, including:
a first module for acquiring a plurality of parameter information related to the inventory quantity;
the second module is used for carrying out correlation analysis on the plurality of parameter information, and carrying out first elimination processing on the plurality of parameter information according to the result of the correlation analysis to obtain first data; the first elimination is used for eliminating repeated parameters in a plurality of parameter information;
The third module is used for carrying out parameter relevance analysis on the first data to obtain a relevance coefficient; performing second elimination processing on the plurality of parameter information according to the association coefficient to obtain second data; the association coefficient is used for representing the influence degree of each parameter information on the stock quantity;
a fourth module, configured to input the second data into a time sequence prediction model, to obtain a required inventory; the time series prediction model is used for representing a model for predicting the stock quantity based on the time series of the related parameters in the second data.
In another aspect, an embodiment of the present invention provides an inventory prediction apparatus, including:
at least one processor;
at least one memory for storing at least one program;
the at least one program, when executed by the at least one processor, causes the at least one processor to implement the inventory prediction method described above.
In another aspect, an embodiment of the present invention provides a storage medium in which a processor-executable program is stored, which when executed by a processor is configured to implement the above-described inventory level prediction method.
According to the embodiment of the invention, the time sequence prediction model for predicting the time sequence based on the related parameters is used for predicting the inventory, so that the influence of the time sequence on the inventory is fully considered, and the prediction accuracy of the model is improved; meanwhile, through correlation analysis and correlation analysis, data with smaller influence in parameter information is removed, and prediction accuracy of the model is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following description is made with reference to the accompanying drawings of the embodiments of the present invention or the related technical solutions in the prior art, and it should be understood that the drawings in the following description are only for convenience and clarity of describing some embodiments in the technical solutions of the present invention, and other drawings may be obtained according to these drawings without the need of inventive labor for those skilled in the art.
FIG. 1 is a flow chart of an embodiment of a stock quantity prediction method provided by the present invention;
FIG. 2 is a flow chart of another embodiment of the inventory prediction method provided by the present invention;
FIG. 3 is a schematic diagram illustrating a time series prediction model according to an embodiment of the present invention;
FIG. 4 is a flow chart of one embodiment of the time series prediction model training provided by the present invention;
FIG. 5 is a flow chart of another embodiment of the time series prediction model training provided by the present invention;
FIG. 6 is a flow chart of a second culling embodiment of the present invention;
FIG. 7 is a schematic diagram of a processing result of an embodiment of the first culling and the second culling provided in the present invention;
FIG. 8 is a schematic diagram illustrating the architecture of one embodiment of a stock quantity prediction system provided by the present invention;
FIG. 9 is a schematic diagram showing the construction of an embodiment of the stock quantity predicting apparatus according to the present invention;
fig. 10 is a schematic structural diagram of another embodiment of the stock quantity predicting device provided by the present invention.
Detailed Description
Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to like or similar elements or elements having like or similar functions throughout. The embodiments described below by referring to the drawings are illustrative only and are not to be construed as limiting the invention. The step numbers in the following embodiments are set for convenience of illustration only, and the order between the steps is not limited in any way, and the execution order of the steps in the embodiments may be adaptively adjusted according to the understanding of those skilled in the art.
Before describing embodiments of the present invention in further detail, the terms and terminology involved in the embodiments of the present invention will be described, and the terms and terminology involved in the embodiments of the present invention are suitable for the following explanation:
artificial intelligence (Artificial Intelligence, AI), also known as smart machine, machine intelligence, refers to machines manufactured by humans that can exhibit intelligence. Artificial intelligence generally refers to the technology of presenting human intelligence through a common computer program; the system is a theory, a method, a technology and an application system which simulate, extend and extend human intelligence by using a digital computer or a machine controlled by the digital computer, sense environment, acquire knowledge and acquire an optimal result by using the knowledge. In other words, artificial intelligence is an integrated technology of computer science that attempts to understand the essence of intelligence and to produce a new intelligent machine that can react in a similar way to human intelligence. Artificial intelligence, i.e. research on design principles and implementation methods of various intelligent machines, enables the machines to have functions of sensing, reasoning and decision.
Artificial neural networks (Artificial Neural Networks, ANNs) are algorithmic mathematical models that mimic the behavioral characteristics of animal neural networks and perform distributed parallel information processing. The network is dependent on the complexity of the system, and the aim of processing information is achieved by adjusting the relation of interconnection among a large number of nodes, and the network has self-learning and self-adapting capabilities.
Safety stock is a buffer stock prepared to prevent future supply or demand uncertainty factors, at the expense of stock capital occupation, in exchange for guarantee of production stability and delivery timeliness for a period of time. If the safety stock is too large, warehouse management cost is increased, mobile funds are occupied, and finished products and raw materials are lost; if the safety stock is too small, the service level is reduced, sales profits and enterprise reputation are affected, and normal operation of the production process is affected. Therefore, a reasonably designed safety stock forecasting scheme is of great significance to the production and management of enterprises.
Inventory demand forecasting problems can be attributed to time series forecasting problems. A general description of such problems is: future time t + p point values are predicted from the time value before t time. In early studies, there were many algorithms for the problem of time series prediction. The most common are the autoregressive model (AR) and autoregressive moving average model (ARMA) in linear regression methods. However, most inventory forecasting problems are nonlinear in nature and therefore good results are difficult to obtain using conventional linear methods.
However, most inventory forecasting problems are nonlinear, and thus a large number of nonlinear methods are currently developed, among which neural networks are considered to be the most effective inventory forecasting methods. Neural networks not only have a variety of known nonlinear function capabilities, but also can obtain an approximation of an unknown function by training samples. Therefore, neural networks are widely used in the problem of time series data prediction.
In the present overall situation, inventory demand prediction still stays in the traditional linear research method, and has a great limitation, for this reason, the embodiment of the invention provides a method for predicting inventory demand based on gray correlation analysis and a time sequence neural network hybrid model, and inventory demand prediction is performed by analyzing the current warehouse management and application data of an enterprise and selecting an NARX neural network in the time sequence neural network as a prediction algorithm model.
The inventory level prediction method and system according to the embodiment of the present invention will be described in detail below with reference to the accompanying drawings, and the inventory level prediction method according to the embodiment of the present invention will be described first.
Referring to fig. 1, an embodiment of the present invention provides a method for predicting an inventory amount, which may be applied to a terminal, a server, software running in a terminal or a server, and the like. The terminal may be, but is not limited to, a tablet computer, a notebook computer, a desktop computer, etc. The server may be an independent physical server, a server cluster or a distributed system formed by a plurality of physical servers, or a cloud server providing cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDNs, basic cloud computing services such as big data and artificial intelligent platforms. The stock quantity prediction method in the embodiment of the invention mainly comprises the following steps:
S100: acquiring a plurality of parameter information related to the stock quantity;
s200: carrying out correlation analysis on the plurality of parameter information, and carrying out first rejection processing on the plurality of parameter information according to the result of the correlation analysis to obtain first data; the first elimination is used for eliminating repeated parameters in a plurality of parameter information;
s300: carrying out parameter relevance analysis on the first data to obtain a relevance coefficient; performing second elimination processing on the plurality of parameter information according to the association coefficient to obtain second data; the association coefficient is used for representing the influence degree of each parameter information on the stock quantity;
s400: inputting the second data into a time sequence prediction model to obtain a required stock quantity; the time series prediction model is used to characterize a model that predicts the inventory based on the time series of the relevant parameters in the second data.
In some possible implementations, the present embodiments make predictions of inventory levels by building a mixture model of gray correlation analysis and time series neural networks. It will be appreciated that the parameter information may be a different kind of classification information characterizing the impact of inventory quantity, illustratively information related to business, information related to market, information related to consumer. Meanwhile, the parameter information can also characterize historical information affecting the stock quantity. The specific content and expression form of the parameter information are not particularly limited in the present application. Referring to fig. 2, first, the collected data is preprocessed, normalized, and influence factors are simplified using correlation analysis (pearson correlation analysis may be employed, for example). Next, a final influencing factor is determined using a correlation analysis (illustratively, a gray correlation analysis method may be employed), and a time-series prediction model (illustratively, the time-series prediction model may be a NARX model) is built using the influencing factor as an input to the neural network. And finally, establishing an NARX neural network model, and adjusting model parameters to enable an output result to meet the requirement.
It can be understood that when data is predicted by the time sequence prediction model, certain requirements are required for data quality, the data needs to be standardized, and the accuracy of the prediction result can be ensured only by the processed data. Illustratively, according to the characteristics of the NARX neural network, the input layer data in the embodiment of the application needs a certain degree of independence, which means that the influence factors of the input layer cannot have a functional relationship with each other, otherwise, data repetition is caused, and the complexity of the network is increased; on the other hand, the embodiment of the application should ensure that the correlation exists between the input layer and the output layer of the influencing factors, otherwise, the network is not converged or the prediction is not good and invalid. Therefore, before training, the correlation of the data needs to be tested. Among the original data, many are redundant. This means that many influencing factors have strong correlation and have functional relation, so that the normalized data can be the same or very close. And because the influence of the demand prediction is the same or close, deleting one of the demands does not influence the prediction result. Therefore, the original data needs to be processed to optimize the data quality, and the repeated parameters in the parameter information are removed through the first elimination. According to the embodiment of the application, the related parameters with smaller influence on the inventory in the parameter information are removed through the second elimination, so that the prediction accuracy of the model is improved.
Optionally, in one embodiment of the present invention, the time series prediction model is trained by:
carrying out correlation analysis on a plurality of parameter samples, and carrying out first rejection processing on the plurality of parameter samples according to the result of the correlation analysis to obtain a first characteristic sample;
carrying out parameter relevance analysis on the first characteristic sample to obtain a sample coefficient; performing second rejection processing on a plurality of parameter samples according to the sample coefficients to obtain second characteristic samples;
and inputting the second characteristic sample into a time sequence prediction model to obtain a prediction result, calculating a loss value by adopting a loss function according to the prediction result and a real result, and updating parameters of the time sequence prediction model according to the loss value to obtain a trained time sequence prediction model.
In some possible implementations, referring to fig. 3, the time series prediction model in the embodiment of the present invention may be a NARX neural network model. It will be appreciated that the NARX neural network is a nonlinear autoregressive model with external inputs (nonlinear autoregressive model with external inputs). The NARX neural network model is different from the static neural network model without output feedback. As a dynamic recursive network, neural networks introduce a delay module and an output feedback model. It introduces output vector delay feedback into the network training to form new input vectors. In training and simulation, the method can not only consider past input, but also consider past output, and remarkably improves the generalization capability of network capacity. It can be understood that in the model training process, model prediction is performed through the feature samples, and the model is updated according to the difference between the predicted result and the real result, so as to complete the model training process.
Optionally, referring to fig. 4, in an embodiment of the present application, the first feature sample includes a training sample, a verification sample, and a test sample, and the method further includes:
s410: inputting the training sample into a first prediction model to obtain a first result; the first prediction model is used for representing a time sequence prediction model of an open loop state;
s420: inputting the training sample and the first result into a first prediction model, and performing closed-loop training on the first prediction model to obtain a second prediction model;
s430: training the second prediction model through the verification sample and the real result corresponding to the verification sample to obtain a third prediction model;
s440: and adjusting parameters of the third prediction model through the test sample and the real result corresponding to the test sample until the prediction result reaches a set threshold value to obtain a time sequence prediction model.
In some possible embodiments, it will be appreciated that based on the above description, the NARX neural network is a system for describing nonlinear discrete systems, which enhances the memory of historical data by adding delay and feedback mechanisms, belonging to dynamic neural networks. According to the embodiment of the application, the first characteristic sample is divided into a training sample, a verification sample and a test sample according to the characteristics of the time sequence, and the sample segmentation mode is carried out according to the time sequence. Then, performing open-loop training on the model by using a training sample, and converting the trained model into a closed-loop model; then testing the performance of the closed loop model by verifying the sample; after the performance reaches a feasible level, the accuracy is adjusted through the test sample, and the establishment and test process of the model are completed. Specifically, referring to fig. 5, the mathematical model of the NARX neural network model can be expressed as:
y(t)＝f(y(t-1),y(t-2),...,y(t-n),x(t-1),x(t-2),...,x(t-n),w) (1)
Referring to fig. 3, in formula (1), x ()' is a point of an input vector x (t) (i.e., an input layer), where x (t) in the embodiment of the present application represents a demand coefficient of t months; y (-) is the output vector, which is some point of the output vector y (t) (i.e., the output layer), which represents the t month requirement in the embodiment of the present application. n is n x (n x Not less than 1) input order of nonlinear system, n y (n y Not less than 1) is the output order, and n x ≥n y W is a network weight matrix; f is a nonlinear function formed during training. Illustratively, as in the training process of fig. 5, the delay time of the model is determined to be 4, which takes into account the inputs of the first four time periods and feeds back the current output to the network to predict the demand of the fifth time period.
Illustratively, in the model training process, 70% of data is randomly selected for the NARX network model to train, 15% is verified, and the other 15% is tested. The output result is closer to the actual demand value by repeatedly training the network adjustment parameters. If the output error is large, the weights of the underlying neurons need to be modified to keep the error at an acceptable level.
In addition, after determining the influencing factors, building the time series prediction model, it is necessary to check whether the training network has good generalization capability and is applicable. In general, if there is a significant correlation between the time period prediction errors, the characteristics of the neural network are poor and cannot be used for prediction. In general, other error-independent function values are assumed to be within the 95% confidence interval, and the network can be considered to have good characteristics. And finally determining the number of hidden neurons and the delay order in the NARX neural network model structure by continuously adjusting the parameters.
Optionally, in an embodiment of the present invention, the time series prediction model comprises an input layer, the method further comprising the steps of:
determining the number of input nodes in the input layer according to the second characteristic sample;
determining the number of output nodes in the input layer according to the first result; wherein the number of output nodes is smaller than the number of input nodes.
Optionally, in an embodiment of the present invention, the step of performing correlation analysis on the plurality of parameter information and performing a first culling process on the plurality of parameter information according to a result of the correlation analysis includes:
acquiring a first correlation coefficient between the first parameter information and the second parameter information; the first parameter information and the second parameter information are any two parameter information in a plurality of parameter information;
if the first correlation coefficient is a preset coefficient, eliminating the first parameter information; or if the first correlation coefficient is a preset coefficient, eliminating the second parameter information.
In some possible implementations, the embodiments of the present invention may use pearson correlation analysis in the correlation analysis, where the preset coefficient may be 1, consider the correlation coefficient of the data item with the result of 1 as a data repetition, and then delete one of the data items. The pearson correlation analysis principle is as follows, the joint distribution of the random variables X and Y is a two-dimensional normal distribution, xi and yi are n independent observations, and the pearson correlation coefficient is as shown in formula (2).
After the collected data is consolidated, because NARX neural network hidden layer transfer functions are typically used for log sig and Tansig, which map real numbers to the [0,1] interval, some method is needed to transform the data before the network is formally trained. In the embodiment of the invention, the data is normalized according to the formula (3) so as to improve the prediction accuracy of the model.
Wherein x in formula (3) mon For characterizing minima, x in observations max For characterizing the maximum value in the observed values.
Optionally, referring to fig. 6, in an embodiment of the present invention, parameter relevance analysis is performed on the first data to obtain a relevance coefficient; and performing a second elimination process on the plurality of parameter information according to the association coefficient to obtain second data, including:
s310: establishing an original data matrix according to the first data; the original data matrix is used for representing data based on time sequence change;
s320: based on the data of the first time period, carrying out initial change on the original data matrix to obtain a first data matrix;
s330: performing absolute difference processing on the first data matrix to obtain a first sequence;
s340: carrying out association coefficient solving on the first sequence to obtain association coefficients;
S350: based on the association coefficient, arranging the parameter information and carrying out final bit elimination processing; or rejecting the parameter information corresponding to the association coefficient smaller than the preset threshold value.
In some possible implementations, the correlation analysis in the embodiments of the present application may be analyzed based on the influence factors of the gray correlation analysis. Specifically, gray correlation analysis is a method for measuring the degree of correlation between factors, which analyzes the degree of correlation by recognizing the trend of development between factors. The gray correlation analysis method can analyze the relationship between subsystems (or factors) in the system. Therefore, the gray correlation analysis can quantitatively measure the development trend of the system, and is very suitable for dynamic process analysis. Illustratively, this analysis is as follows:
s61: establishing a raw data matrix x for each index (i.e. first data) i
x i ＝(x i (1),x i (2),x i (3),...,x i (k),...) (4)
In formula (4), x i (k) Representing raw data of factor i over a k period of time, e.g. x 1 (1) The data for the first factor over the first month may be represented.
S62: solving for the initial change matrix (i.e., the first data matrix) x by equation (5) i ′
x i ′ ＝(x i (1)/x i (1),x i (2)/x i (1),...,x i (k)/x i (1),...)
＝(x i ′ (1),x i ′ (2),...,x i ′ (k),...) (5)
S63: the absolute difference sequence (i.e., the first sequence) is determined by equation (6)
S64: calculating grey correlation coefficient, i.e. correlation coefficient xi, by formula (7) 0i (k)
In the formula (7) of the present application,as a resolution factor, the effect is to increase the correlation factor between significant differences +.>In the embodiment of the application, it is preferable that +.>
The gray correlation is calculated by the formula (8):
specifically, there are many factors that affect inventory requirements, and the relationship between the factors is complex. The embodiment of the application mainly analyzes main influencing factors from three aspects of enterprises, markets and users. From an enterprise perspective, the size of the provider, reputation, advantages of supplying the product, price of the product, and shipping style all affect the number of purchases. Inventory management, inventory costs, and inventory forecast levels for enterprises also greatly impact demand; from a market perspective, the impact of the market on demand is indirect. For example, a change in market price will inevitably result in a change in product price, affecting demand; from the user's perspective, the user's consumption level, personalized needs, order volume, etc., can greatly impact the needs.
It can be understood that many factors influencing inventory requirements are included, but many factors have little influence on requirements, and in order to determine main influence factors, the embodiment of the application uses a gray correlation analysis model to determine main requirement influence factors from three directions of enterprises, markets and users, and the specific process is as follows.
Firstly, the pearson correlation analysis is adopted in the data preprocessing stage, so that the randomness of factor selection is eliminated, and the complexity of the problem is reduced. The main demand factors for obtaining Xi market inventory are as follows: the correlation between nine influencing factors and the demand is shown in table 1, and the time (x 1), the forecast stock demand (x 2), the stock cost (x 3), the order quantity (x 4), the order price (x 5), the total order quantity (x 6), the actual cost (x 7), the supply price (x 8) and the defective rate (x 9) are all nine factors.
TABLE 1
Then, gray correlation analysis method is used to analyze gray correlation between the demand and the above 9 influencing factors. And (3) calculating gray correlation degree by using the formulas (4) to (8), removing factors with smaller correlation degree, and selecting a key factor prediction requirement.
Finally, seven influencing factors of inventory cost, order quantity, order price, total order quantity, actual cost, supply price and quantity of unqualified products are determined and used as input of NARX neural network.
Optionally, in one embodiment of the present invention, the method further comprises:
classifying the plurality of parameter information to obtain third parameter information, fourth parameter information and fifth parameter information; factors that influence classification include enterprise aspects, market aspects, and consumption aspects;
Carrying out correlation analysis and association analysis on the third parameter information, and carrying out rejection processing on the third parameter information according to a preset rejection proportion;
performing correlation analysis and association analysis on the fourth parameter information, and performing rejection processing on the fourth parameter information according to a preset rejection proportion;
carrying out correlation analysis and association analysis on the fifth parameter information, and carrying out rejection processing on the fifth parameter information according to a preset rejection proportion.
In some possible implementation manners, the rejection manner in the embodiment of the application can be classified and rejected according to parameters of different types or fields, so that the rejection processing is performed on the data, and the participation degree of each type of parameters is reserved, so that the prediction accuracy of the model is improved. Illustratively, as shown in fig. 7, the parameter information is divided into several parameter information according to influence factors. Of course, it is understood that the parameter information classified into 3 types shown in fig. 7 is an exemplary example, and those skilled in the art can classify the parameter information according to actual needs. When the parameter screening is performed for removing, the parameter information identified by the region 710 can be removed according to each type of parameters, namely, in the third parameter information; in the fourth parameter information, the parameter information identified by the area 720 is removed; in the fifth parameter information, the parameter information identified by the region 730 is culled. The method of the eliminating process can select a mode of presetting eliminating proportion to eliminate parameter information; the end reject mode may also be selected for reject, and the present application is not particularly limited.
The embodiment of the application provides a stock quantity prediction method based on gray correlation analysis and a time series neural network hybrid model. Firstly, screening the most relevant factors influencing inventory requirements by using pearson correlation analysis, then evaluating the screened factors by using gray correlation analysis, removing the factors with smaller correlation degree, and selecting core key influence factors. And finally, predicting the selected core key factors as the input of the NARX neural network, training 70% of data, verifying 15% of data, testing the rest 15% of data, continuously optimizing the model according to the result, the 95% confidence space principle and the like, and finally determining the NARX neural network model structure. Compared with the currently mainstream weighted average prediction method and BP neural network prediction scheme, the inventory demand prediction method based on gray correlation analysis and time sequence neural network mixed model is more accurate in result. Experiments show that the error autocorrelation coefficient of the proposal is in a 95% confidence interval, which shows that the model generalization capability is better.
In summary, according to the embodiment of the application, the time sequence prediction model for predicting the time sequence based on the related parameters is used for predicting the inventory, so that the influence of the time sequence on the inventory is fully considered, and the prediction accuracy of the model is improved; meanwhile, through correlation analysis and correlation analysis, data with smaller influence in parameter information is removed, and prediction accuracy of the model is improved.
Next, a stock quantity prediction system according to an embodiment of the present invention will be described with reference to fig. 8.
FIG. 8 is a schematic diagram of a stock quantity prediction system according to an embodiment of the present invention, the system specifically includes:
a first module 810 for obtaining a number of parameter information related to the inventory quantity;
a second module 820, configured to perform correlation analysis on the plurality of parameter information, and perform a first rejection process on the plurality of parameter information according to a result of the correlation analysis, so as to obtain first data; the first elimination is used for eliminating repeated parameters in a plurality of parameter information;
a third module 830, configured to perform parameter association analysis on the first data to obtain an association coefficient; performing second elimination processing on the plurality of parameter information according to the association coefficient to obtain second data; the association coefficient is used for representing the influence degree of each parameter information on the stock quantity;
a fourth module 840, configured to input the second data into the time-series prediction model, to obtain a required inventory; the time series prediction model is used to characterize a model that predicts the inventory based on the time series of the relevant parameters in the second data.
It can be seen that the content in the above method embodiment is applicable to the system embodiment, and the functions specifically implemented by the system embodiment are the same as those of the method embodiment, and the beneficial effects achieved by the method embodiment are the same as those achieved by the method embodiment.
The embodiment of the application provides a stock quantity prediction device, which comprises the following components:
at least one processor;
at least one memory for storing at least one program;
the at least one program, when executed by the at least one processor, causes the at least one processor to implement a stock quantity prediction method.
Similarly, the content in the above method embodiment is applicable to the embodiment of the present device, and the functions specifically implemented by the embodiment of the present device are the same as those of the embodiment of the above method, and the beneficial effects achieved by the embodiment of the above method are the same as those achieved by the embodiment of the above method.
The stock quantity prediction device for executing the stock quantity prediction method provided by the embodiment of the present application may be a terminal, and referring to fig. 9, fig. 9 is a partial block diagram of a terminal provided by the embodiment of the present application, where the terminal includes: radio Frequency (RF) circuitry 1010, memory 1020, input unit 1030, display unit 1040, sensor 1050, audio circuitry 1060, wireless fidelity (wireless fidelity, wiFi) module 1070, processor 1080, and power source 1090. It will be appreciated by those skilled in the art that the terminal structure shown in fig. 9 is not limiting of the handset and may include more or fewer components than shown, or may combine certain components, or a different arrangement of components.
The RF circuit can be used for receiving and transmitting signals in the process of receiving and transmitting information or communication, particularly, after receiving downlink information of the base station, the downlink information is processed by the processor; in addition, the data of the design uplink is sent to the base station.
The memory may be used to store software programs and modules, and the processor executes the software programs and modules stored in the memory to perform various functional applications and data processing of the handset.
The input unit may be used to receive input numeric or character information and to generate key signal inputs related to the settings and function control of the handset. In particular, the input unit may include a touch panel 1031 and other input devices 1032.
The display unit may be used to display input information or provided information and various menus of the mobile phone. The display unit 1040 may include a display panel 1041.
Audio circuitry 1060, a speaker 1061, and a microphone 1062 may provide an audio interface.
In this embodiment, the processor included in the terminal may perform the stock quantity prediction method of the previous embodiment.
The inventory prediction device for performing the inventory prediction method according to the embodiment of the present application may also be a server, and referring to fig. 10, fig. 10 is a partial block diagram of a server according to the embodiment of the present application, where the server 1100 may have a relatively large difference due to different configurations or performances, and may include one or more central processing units (abbreviated as CPUs, i.e. Central Processing Units), 1122 (e.g. one or more processors) and a memory 1132, and one or more storage media 1130 (e.g. one or more mass storage devices) storing the application 1142 or the data 1144. Wherein the memory 1132 and the storage medium 1130 may be transitory or persistent. The program stored on the storage medium 1130 may include one or more modules (not shown), each of which may include a series of instruction operations on the server 1100. Still further, the central processor 1122 may be provided in communication with a storage medium 1130, executing a series of instruction operations in the storage medium 1130 on the server 1100.
The server(s) may also include one or more power supplies 1126, one or more wired or wireless network interfaces 1150, one or more input-output interfaces 1158, and/or one or more operating systems 1141, such as Windows Server, mac OS XTM, unixTM, linuxTM, freeBSDTM, etc.
A processor in the server may be used to perform the above inventory prediction method.
Similarly, the content in the above method embodiment is applicable to the embodiment of the present device, and the functions specifically implemented by the embodiment of the present device are the same as those of the embodiment of the above method, and the beneficial effects achieved by the embodiment of the above method are the same as those achieved by the embodiment of the above method.
The embodiment of the present invention also provides a computer-readable storage medium in which a processor-executable program is stored, which when executed by a processor is configured to perform the above-described stock quantity prediction method.
Similarly, the content in the above method embodiment is applicable to the present storage medium embodiment, and the specific functions of the present storage medium embodiment are the same as those of the above method embodiment, and the achieved beneficial effects are the same as those of the above method embodiment.
In some alternative embodiments, the functions/acts noted in the block diagrams may occur out of the order noted in the operational illustrations. For example, two blocks shown in succession may in fact be executed substantially concurrently or the blocks may sometimes be executed in the reverse order, depending upon the functionality/acts involved. Furthermore, the embodiments presented and described in the flowcharts of the present invention are provided by way of example in order to provide a more thorough understanding of the technology. The disclosed methods are not limited to the operations and logic flows presented by embodiments of the invention. Alternative embodiments are contemplated in which the order of various operations is changed, and in which sub-operations described as part of a larger operation are performed independently.
Furthermore, while the invention is described in the context of functional modules, it should be appreciated that, unless otherwise indicated, one or more of the functions and/or features may be integrated in a single physical device and/or software module or may be implemented in separate physical devices or software modules. It will also be appreciated that a detailed discussion of the actual implementation of each module is not necessary to an understanding of the present invention. Rather, the actual implementation of the various functional modules in the apparatus disclosed in the embodiments of the present invention will be understood within the ordinary skill of the engineer in view of their attributes, functions and internal relationships. Accordingly, one of ordinary skill in the art can implement the invention as set forth in the claims without undue experimentation. It is also to be understood that the specific concepts disclosed are merely illustrative and are not intended to be limiting upon the scope of the invention, which is to be defined in the appended claims and their full scope of equivalents.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in the form of a software product stored in a storage medium, including several programs for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
Logic and/or steps represented in the flowcharts or otherwise described herein, e.g., a ordered listing of executable programs for implementing logical functions, can be embodied in any computer-readable medium for use by or in connection with a program execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the programs from the program execution system, apparatus, or device and execute the programs. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the program execution system, apparatus, or device.
More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). In addition, the computer readable medium may even be paper or other suitable medium on which the program is printed, as the program may be electronically captured, via, for instance, optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner, if necessary, and then stored in a computer memory.
It is to be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable program execution system. For example, if implemented in hardware, as in another embodiment, may be implemented using any one or combination of the following techniques, as is well known in the art: discrete logic circuits having logic gates for implementing logic functions on data signals, application specific integrated circuits having suitable combinational logic gates, programmable Gate Arrays (PGAs), field Programmable Gate Arrays (FPGAs), and the like.
In the foregoing description of the present specification, reference has been made to the terms "one embodiment/example", "another embodiment/example", "certain embodiments/examples", and the like, means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiments or examples. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
While embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that: many changes, modifications, substitutions and variations may be made to the embodiments without departing from the spirit and principles of the invention, the scope of which is defined by the claims and their equivalents.
While the preferred embodiment of the present invention has been described in detail, the present invention is not limited to the embodiments described above, and various equivalent modifications and substitutions can be made by those skilled in the art without departing from the spirit of the present invention, and these equivalent modifications and substitutions are intended to be included in the scope of the present invention as defined in the appended claims.
Claims (10)
1. A method for predicting an inventory amount, comprising the steps of:
acquiring a plurality of parameter information related to the stock quantity;
performing correlation analysis on the plurality of parameter information, and performing first rejection processing on the plurality of parameter information according to the result of the correlation analysis to obtain first data; the first elimination is used for eliminating repeated parameters in a plurality of parameter information;
carrying out parameter relevance analysis on the first data to obtain a relevance coefficient; performing second elimination processing on the plurality of parameter information according to the association coefficient to obtain second data; the association coefficient is used for representing the influence degree of each parameter information on the stock quantity;
inputting the second data into a time sequence prediction model to obtain a required stock quantity; the time series prediction model is used for representing a model for predicting the stock quantity based on the time series of the related parameters in the second data.
2. The stock quantity prediction method according to claim 1, wherein the time series prediction model is trained by:
obtaining a plurality of parameter samples;
performing correlation analysis on the plurality of parameter samples, and performing first rejection processing on the plurality of parameter samples according to the result of the correlation analysis to obtain a first characteristic sample;
Carrying out parameter relevance analysis on the first characteristic sample to obtain a sample coefficient; performing second rejection processing on the plurality of parameter samples according to the sample coefficients to obtain second characteristic samples;
and inputting the second characteristic sample into a time sequence prediction model to obtain a prediction result, calculating a loss value by adopting a loss function according to the prediction result and a real result, and updating parameters of the time sequence prediction model according to the loss value to obtain the trained time sequence prediction model.
3. The inventory level prediction method according to claim 2, wherein the first feature sample includes a training sample, a validation sample, and a test sample, the method further comprising:
inputting the training sample into a first prediction model to obtain a first result; the first prediction model is used for representing a time sequence prediction model of an open loop state;
inputting the training sample and the first result into the first prediction model, and performing closed-loop training on the first prediction model to obtain a second prediction model;
training the second prediction model through the verification sample and the real result corresponding to the verification sample to obtain a third prediction model;
And adjusting the parameters of the third prediction model through the test sample and the real result corresponding to the test sample until the prediction result reaches a set threshold value to obtain a time sequence prediction model.
4. The inventory level prediction method according to claim 3, characterized in that the time series prediction model includes an input layer, the method further comprising the steps of:
determining the number of input nodes in the input layer according to the second characteristic sample;
determining the number of output nodes in the input layer according to the first result; wherein the number of output nodes is less than the number of input nodes.
5. The stock quantity prediction method according to claim 1, wherein the step of performing correlation analysis on the plurality of parameter information and performing first culling processing on the plurality of parameter information based on a result of the correlation analysis comprises:
acquiring a first correlation coefficient between the first parameter information and the second parameter information; the first parameter information and the second parameter information are any two parameter information in the plurality of parameter information;
if the first correlation coefficient is a preset coefficient, eliminating the first parameter information; or if the first correlation coefficient is a preset coefficient, eliminating the second parameter information.
6. The stock quantity prediction method according to claim 1, wherein the parameter correlation analysis is performed on the first data to obtain a correlation coefficient; and performing a second elimination process on the plurality of parameter information according to the association coefficient to obtain second data, including:
establishing an original data matrix according to the first data; the original data matrix is used for representing data based on time sequence change;
based on the data of the first time period, carrying out initial change on the original data matrix to obtain a first data matrix;
performing absolute difference processing on the first data matrix to obtain a first sequence;
carrying out association coefficient solving on the first sequence to obtain association coefficients;
based on the association coefficient, arranging the parameter information and carrying out final bit elimination processing; or, eliminating the parameter information corresponding to the association coefficient smaller than a preset threshold value.
7. The stock quantity prediction method according to claim 1, characterized in that the method further comprises:
classifying the plurality of parameter information to obtain third parameter information, fourth parameter information and fifth parameter information; factors influencing the classification include enterprise aspects, market aspects and consumption aspects;
Performing correlation analysis and association analysis on the third parameter information, and performing rejection processing on the third parameter information according to a preset rejection proportion;
performing correlation analysis and association analysis on the fourth parameter information, and performing rejection processing on the fourth parameter information according to a preset rejection proportion;
performing correlation analysis and association analysis on the fifth parameter information, and performing rejection processing on the fifth parameter information according to a preset rejection proportion.
8. An inventory prediction system, comprising:
a first module for acquiring a plurality of parameter information related to the inventory quantity;
the second module is used for carrying out correlation analysis on the plurality of parameter information, and carrying out first elimination processing on the plurality of parameter information according to the result of the correlation analysis to obtain first data; the first elimination is used for eliminating repeated parameters in a plurality of parameter information;
the third module is used for carrying out parameter relevance analysis on the first data to obtain a relevance coefficient; performing second elimination processing on the plurality of parameter information according to the association coefficient to obtain second data; the association coefficient is used for representing the influence degree of each parameter information on the stock quantity;
A fourth module, configured to input the second data into a time sequence prediction model, to obtain a required inventory; the time series prediction model is used for representing a model for predicting the stock quantity based on the time series of the related parameters in the second data.
9. An inventory quantity prediction device, comprising:
at least one processor;
at least one memory for storing at least one program;
the at least one program, when executed by the at least one processor, causes the at least one processor to implement the inventory level prediction method of any one of claims 1 to 7.
10. A computer-readable storage medium in which a processor-executable program is stored, characterized in that the processor-executable program is for realizing the stock quantity prediction method according to any one of claims 1 to 7 when being executed by a processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310722937.9A CN116777345A (en) | 2023-06-16 | 2023-06-16 | Stock quantity prediction method, system, device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310722937.9A CN116777345A (en) | 2023-06-16 | 2023-06-16 | Stock quantity prediction method, system, device and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116777345A true CN116777345A (en) | 2023-09-19 |
Family
ID=88010928
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310722937.9A Pending CN116777345A (en) | 2023-06-16 | 2023-06-16 | Stock quantity prediction method, system, device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116777345A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118643939A (en) * | 2024-06-19 | 2024-09-13 | 烟台中科新智软件技术有限公司 | A warehouse optimization method, device and medium based on code recognition |
-
2023
- 2023-06-16 CN CN202310722937.9A patent/CN116777345A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118643939A (en) * | 2024-06-19 | 2024-09-13 | 烟台中科新智软件技术有限公司 | A warehouse optimization method, device and medium based on code recognition |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111352965B (en) | Sequence mining model training methods, sequence data processing methods and equipment | |
| CN112633962B (en) | Service recommendation method and device, computer equipment and storage medium | |
| CN111932269B (en) | Equipment information processing method and device | |
| CN111667308A (en) | Advertisement recommendation prediction system and method | |
| CN112766402A (en) | Algorithm selection method and device and electronic equipment | |
| CN116048912B (en) | Cloud server configuration anomaly identification method based on weak supervision learning | |
| KR20240081703A (en) | Method, program and apparatus for demand prediction based on artificial intelligence | |
| CN117667673A (en) | Processing method and device of test cases, storage medium and electronic equipment | |
| CN116777345A (en) | Stock quantity prediction method, system, device and storage medium | |
| CN117077813A (en) | A training method and training system for machine learning models | |
| CN119670742B (en) | Data analysis method, apparatus, computer device, readable storage medium, and program product | |
| Sun et al. | Short-Term Stock Price Forecasting Based on an SVD-LSTM Model. | |
| Vaghela et al. | Boost a weak learner to a strong learner using ensemble system approach | |
| Escovedo et al. | Neuroevolutionary learning in nonstationary environments | |
| CN117312912A (en) | Method, device and computer equipment for generating business data classification prediction model | |
| CN117573961A (en) | Information recommendation method, device, electronic equipment, storage medium and program product | |
| CN115169692A (en) | Time series prediction method and related device | |
| CN115641198A (en) | User operation method, device, electronic equipment and storage medium | |
| CN115147207A (en) | Sample evaluation method, device, storage medium and electronic device | |
| CN116933049A (en) | Feature selection method, device, electronic equipment and storage medium | |
| CN114238726A (en) | User classification method, device, equipment and storage medium | |
| CN111768306A (en) | Risk identification method and system based on intelligent data analysis | |
| CN114117902B (en) | New user load prediction model construction method, device, equipment and storage medium | |
| CN119783156B (en) | Noise generation method, device, terminal and medium based on differential privacy | |
| Xu et al. | Engineering Financial Performance Evaluation of Wireless Network Based on Intelligent Neural Network Model |
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 |