CN111882363A - Sales prediction method, system and terminal - Google Patents
Sales prediction method, system and terminal Download PDFInfo
- Publication number
- CN111882363A CN111882363A CN202010782742.XA CN202010782742A CN111882363A CN 111882363 A CN111882363 A CN 111882363A CN 202010782742 A CN202010782742 A CN 202010782742A CN 111882363 A CN111882363 A CN 111882363A
- Authority
- CN
- China
- Prior art keywords
- sales
- data
- predicted
- training
- model
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0202—Market predictions or forecasting for commercial activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
Abstract
The invention relates to the field of sales forecasting, aims to solve the problem of low accuracy of the existing product sales forecasting method, and provides a sales forecasting method, a sales forecasting system and a terminal, wherein the technical scheme is summarized as follows: acquiring historical sales data and external environment variable data of a product to be predicted based on a time sequence; training at least one time sequence analysis model and at least one machine learning model, and respectively obtaining the optimal parameters and corresponding training errors of each time sequence analysis model and each machine learning model by using a grid search method; and carrying out sales forecast on the product to be forecasted based on the time to be forecasted and according to a plurality of sales forecast models with training errors smaller than a preset value and the corresponding optimal parameters to obtain a plurality of sales forecast values, and determining the final sales forecast value according to the plurality of sales forecast values. The method improves the accuracy of sales prediction, and is suitable for the sales prediction of air-conditioning products.
Description
Technical Field
The invention relates to the field of sales forecasting, in particular to a sales forecasting method, a sales forecasting system and a terminal.
Background
The existing product sales prediction usually adopts a technical method of predicting by using a time sequence analysis algorithm or a machine learning algorithm, a general flow is to construct a characteristic project according to historical sales data, final prediction is carried out after training by adopting a fixed specific algorithm or specific algorithms, and because factors influencing the product sales are more, the single sales prediction method based on the time sequence has the problem of low accuracy, and if the fixed algorithm model is adopted for sales prediction, the model cannot be adjusted according to the sales data change, and the defect of low prediction accuracy also exists.
Disclosure of Invention
The invention aims to solve the problem of low accuracy of the existing product sales prediction method, and provides a sales prediction method, a sales prediction system and a terminal.
The technical scheme adopted by the invention for solving the technical problems is as follows: the sales forecasting method comprises the following steps:
step 1, obtaining historical sales data and external environment variable data of a product to be predicted based on a time sequence, wherein the external environment variable data is data influencing the historical sales of the product to be predicted;
step 2, training at least one time sequence analysis model by taking the historical sales data based on the time sequence as training data, training at least one machine learning model by taking the historical sales data based on the time sequence and external environment variable data as training data, and obtaining optimal parameters and corresponding training errors of each time sequence analysis model and each machine learning model by respectively utilizing a grid search method;
and 3, determining a plurality of sales forecasting models with the training errors smaller than a preset value, forecasting sales of products to be forecasted based on time to be forecasted and according to the determined plurality of sales forecasting models and the corresponding optimal parameters thereof to obtain a plurality of sales forecasting values, and determining a final sales forecasting value according to the plurality of sales forecasting values.
Further, in order to facilitate the seller to make corresponding adjustment to the product sales strategy, the method further comprises:
step 4, in a machine learning model for sales prediction, acquiring the importance degree of each feature in external environment variable data according to an import method;
and 5, normalizing the importance degrees of the features, calculating the average value of the importance degrees of the features corresponding to the machine learning models, and sequencing and outputting the features according to the average value of the importance degrees of the features.
To further improve the accuracy of sales volume predictions, the external environment variable data comprises: mission data, auction data, macro-economic data, and/or air temperature data, the macro-economic data comprising: and the residential consumption price index and/or the real estate construction area, wherein each type of data is corresponding characteristics.
In order to further improve the accuracy of the sales volume prediction, the step 1 further includes:
drawing a historical sales volume graph of the product to be predicted according to historical sales volume data of the product to be predicted based on the time sequence, determining a statistical rule according to the historical sales volume graph, performing sales volume prediction on the product to be predicted based on the time to be predicted and according to the statistical rule to obtain a statistical prediction value, and adding the statistical prediction value into external environment variable data as a statistical characteristic.
In order to further improve the accuracy of the sales volume prediction, the step 1 further includes:
and based on the time sequence, eliminating historical sales data and external environment variable data with more missing values and a variance close to zero, and filling the data with less missing values by adopting a mean value or neighbor method.
Furthermore, in order to realize the sales prediction of the product to be predicted through the machine learning model, in step 3, for the machine learning model in the determined multiple sales prediction models, external environment variable data corresponding to the time to be predicted is predicted based on the time to be predicted, and the sales prediction of the product to be predicted is performed according to the time to be predicted and the external environment variable data corresponding to the time to be predicted and based on the machine learning model.
Specifically, the time series analysis model includes: an STL model, an ETS model, an ARIMA model, and/or a SARIMA model, the machine learning model comprising: a GBM model, an XGBOST model, an RF model, a GBDT model, and/or a CATBOOST model.
Further, to achieve the determination of the final sales volume predicted value, the method for determining the final sales volume predicted value according to the plurality of sales volume predicted values includes:
and calculating the average value of the plurality of sales predicted values to obtain the final sales predicted value.
In order to solve the problem of low accuracy rate of the existing product sales prediction method, the invention also provides a sales prediction system, which comprises the following steps: the device comprises an acquisition unit, a training unit and a determination unit;
the acquisition unit is used for acquiring historical sales data and external environment variable data of a product to be predicted based on a time sequence, wherein the external environment variable data is data influencing the historical sales of the product to be predicted;
the training unit is used for training at least one time sequence analysis model by taking the historical sales data based on the time sequence as training data, training at least one machine learning model by taking the historical sales data based on the time sequence and external environment variable data as training data, and obtaining optimal parameters and corresponding training errors of each time sequence analysis model and each machine learning model by respectively utilizing a grid search method;
the determining unit is used for determining a plurality of models with the training errors smaller than a preset value, carrying out sales forecast on products to be forecasted based on the time to be forecasted and according to the determined plurality of models and the corresponding optimal parameters thereof to obtain a plurality of sales forecast values, and determining the final sales forecast value according to the plurality of sales forecast values.
In order to solve the problem of low accuracy rate of the existing product sales prediction method, the invention also provides a sales prediction terminal, wherein the terminal comprises a processor, a memory and a communication bus;
the communication bus is used for realizing connection communication between the processor and the memory;
the processor is configured to execute one or more programs in the memory to implement the steps of the sales prediction method described above.
The invention has the beneficial effects that: according to the sales prediction method, the sales prediction system and the terminal, the sales prediction of the product to be predicted is realized through two models, namely the time sequence analysis model and the machine learning model, during model training, model training is performed not only through historical sales data corresponding to the time sequence but also based on external environment variable data, and finally, the sales prediction of the product to be predicted is performed through selecting a plurality of sales prediction models with smaller prediction errors from the plurality of sales prediction models, so that the accuracy of the sales prediction of the product to be predicted is improved. In addition, the output is sorted according to the importance degree of each characteristic, so that a seller can adjust a corresponding sale strategy according to the importance degree of each characteristic, and multi-dimensional guidance is provided for the sale of products.
Drawings
FIG. 1 is a schematic flow chart illustrating a sales prediction method according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a sales prediction system according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of a sales prediction terminal according to an embodiment of the present invention.
Detailed Description
Embodiments of the present invention will be described in detail below with reference to the accompanying drawings.
The invention aims to solve the problem of low accuracy of the existing product sales prediction method, and provides a sales prediction method, a sales prediction system and a terminal, wherein the main technical concept is as follows: acquiring historical sales data and external environment variable data of a product to be predicted based on a time sequence, wherein the external environment variable data is data influencing the historical sales of the product to be predicted; training at least one time sequence analysis model by taking the historical sales data based on the time sequence as training data, training at least one machine learning model by taking the historical sales data based on the time sequence and external environment variable data as training data, and obtaining optimal parameters and corresponding training errors of each time sequence analysis model and each machine learning model by respectively utilizing a grid search method; determining a plurality of sales forecasting models with the training errors smaller than a preset value, forecasting sales of products to be forecasted based on time to be forecasted and according to the determined sales forecasting models and the optimal parameters corresponding to the sales forecasting models to obtain a plurality of sales forecasting values, and determining a final sales forecasting value according to the sales forecasting values.
Specifically, sales prediction models may exist in the candidate model library, which include two categories: for the time sequence model, training the product to be predicted based on historical sales data corresponding to the time sequence, and obtaining the optimal parameters and corresponding training errors of each time sequence analysis model by using a grid search method in the training process; for the machine learning models, training is carried out on historical sales data corresponding to products to be predicted based on time sequences and external environment variable data influencing the historical sales of the products to be predicted, and in the training process, optimal parameters and corresponding training errors of the machine learning models are obtained by using a grid search method; and after traversing all the sales forecasting models in the candidate model library, determining a preset number of sales forecasting models with the minimum training error to forecast the sales of the product to be forecasted, and finally obtaining a final sales forecasting value to realize the sales forecasting of the product to be forecasted.
Examples
The sales prediction method of the embodiment of the invention, as shown in fig. 1, includes the following steps:
step S1, acquiring historical sales data and external environment variable data of a product to be predicted based on a time sequence, wherein the external environment variable data is data influencing the historical sales of the product to be predicted;
wherein the external environment variable data includes: mission data, contest data, macro-economic data, and/or air temperature data, wherein the mission data is provided internally by the company; the competitive product data is provided by professional companies and comprises historical sales data of products to be tested of competitive manufacturers; the macro economic data and the air temperature data are crawled by crawlers, and the air temperature data comprise monthly average air temperature, monthly maximum air temperature, monthly minimum air temperature and the like. The macro-economic data comprises: and the residential consumption price index and/or the real estate construction area, wherein each type of data is corresponding characteristics. The data is made up of multiple columns, the first being a timestamp, the remaining columns being the corresponding features.
In order to improve the accuracy of predicting the sales volume of the product to be predicted, in this embodiment, the step S1 further includes: drawing a historical sales volume graph of the product to be predicted according to historical sales volume data of the product to be predicted based on the time sequence, determining a statistical rule according to the historical sales volume graph, performing sales volume prediction on the product to be predicted based on the time to be predicted and according to the statistical rule to obtain a statistical prediction value, and adding the statistical prediction value into external environment variable data as a statistical characteristic. Specifically, a historical sales amount line graph can be drawn, some statistical rules, such as the shape of sales curves of each year, are found to have similarity, and the statistically found rules are converted into formulas to predict the sales values, and the results are used as statistical characteristics.
In order to ensure the integrity of the data to further improve the accuracy of the sales volume prediction, the step S1 further includes: and based on the time sequence, eliminating historical sales data and external environment variable data with more missing values and a variance close to zero, and filling the data with less missing values by adopting a mean value or neighbor method.
The results of the historical sales data and the external environment variable data based on the time series obtained according to the steps are shown in the following table:
step S2, training at least one time sequence analysis model by taking the historical sales data based on the time sequence as training data, training at least one machine learning model by taking the historical sales data based on the time sequence and external environment variable data as training data, and obtaining optimal parameters and corresponding training errors of each time sequence analysis model and each machine learning model by respectively utilizing a grid search method;
specifically, a sales prediction model may be placed into the library of candidate models, the sales prediction model including two categories: for the time sequence model, training the product to be predicted based on historical sales data corresponding to the time sequence, and obtaining the optimal parameters and corresponding training errors of each time sequence analysis model by using a grid search method in the training process; for the machine learning models, training is carried out on historical sales data corresponding to products to be predicted based on time sequences and external environment variable data influencing the historical sales of the products to be predicted, and in the training process, optimal parameters and corresponding training errors of the machine learning models are obtained by using a grid search method; and traversing all sales models in the candidate model library to obtain the optimal parameters and corresponding training errors of each sales prediction model.
Wherein the time series analysis model comprises: an STL model, an ETS model, an ARIMA model, and/or a SARIMA model, the machine learning model comprising: a GBM model, an XGBOST model, an RF model, a GBDT model, and/or a CATBOOST model.
And step S3, determining a plurality of sales forecasting models with the training errors smaller than a preset value, carrying out sales forecasting on products to be forecasted based on the time to be forecasted and according to the determined sales forecasting models and the corresponding optimal parameters thereof to obtain a plurality of sales forecasting values, and determining the final sales forecasting value according to the sales forecasting values.
Specifically, for all the sales prediction models in the candidate model library, sorting is carried out according to the corresponding training errors, the top K sales prediction models with the minimum training errors are determined, wherein K is a positive integer, and the sales prediction of the product to be predicted is carried out according to the determined K sales prediction models.
It can be understood that, for the time series analysis model in the K sales prediction models, the time to be predicted and the corresponding optimal parameters are input into the time series analysis model, and then the corresponding sales prediction value can be obtained. For the machine learning model in the K sales prediction models, external environment variable data corresponding to the time to be predicted is predicted based on the time to be predicted, the external environment variable data corresponding to the time to be predicted and the corresponding optimal parameters are input into the machine learning model, and then the corresponding sales prediction value can be obtained.
After the sales predicted values corresponding to each sales prediction model are obtained respectively, the average value of the sales predicted values is calculated to obtain the final sales predicted value, and further the sales prediction of the product to be predicted is achieved.
As a further optimization, the sales prediction method according to this embodiment further includes:
step S4, in a machine learning model for sales prediction, the importance degree of each feature in external environment variable data is obtained according to an import method;
step S5, the importance levels of the features are normalized, the average value of the importance levels of the features corresponding to the machine learning models is calculated, and the features are sorted and output according to the average value of the importance levels of the features.
Specifically, in this embodiment, the machine learning model is implemented based on a sklern machine learning package of python, so that the importance of each feature can be obtained according to the import method, if N (0< N < K) machine learning models exist in the K sales prediction models in step S3, N sets of feature importance level values can be obtained, the importance level values of the N sets of features are normalized and then fused, the final ranking of the feature importance levels can be obtained, and a seller can make corresponding adjustment on the production and sales strategies of the product by analyzing and predicting the obtained important features.
Based on the above technical solution, this embodiment further provides a sales prediction system, as shown in fig. 2, including: the device comprises an acquisition unit, a training unit and a determination unit;
the acquisition unit is used for acquiring historical sales data and external environment variable data of a product to be predicted based on a time sequence, wherein the external environment variable data is data influencing the historical sales of the product to be predicted;
the training unit is used for training at least one time sequence analysis model by taking the historical sales data based on the time sequence as training data, training at least one machine learning model by taking the historical sales data based on the time sequence and external environment variable data as training data, and obtaining optimal parameters and corresponding training errors of each time sequence analysis model and each machine learning model by respectively utilizing a grid search method;
the determining unit is used for determining a plurality of models with the training errors smaller than a preset value, carrying out sales forecast on products to be forecasted based on the time to be forecasted and according to the determined plurality of models and the corresponding optimal parameters thereof to obtain a plurality of sales forecast values, and determining the final sales forecast value according to the plurality of sales forecast values.
Based on the above technical solution, this embodiment further provides a sales prediction terminal, as shown in fig. 3, where the terminal includes a processor, a memory, and a communication bus;
the communication bus is used for realizing connection communication between the processor and the memory;
the processor is configured to execute one or more programs in the memory to implement the steps of the sales prediction method described in the above embodiments.
It can be understood that, because the sales predicting system and the sales predicting terminal according to the embodiments of the present invention are a system and a terminal for implementing the sales predicting method according to the embodiments, the system disclosed in the embodiments is relatively simple in description because it corresponds to the method disclosed in the embodiments, and the related points can be referred to only in the partial description of the method.
Claims (10)
1. The sales prediction method is characterized by comprising the following steps:
step 1, obtaining historical sales data and external environment variable data of a product to be predicted based on a time sequence, wherein the external environment variable data is data influencing the historical sales of the product to be predicted;
step 2, training at least one time sequence analysis model by taking the historical sales data based on the time sequence as training data, training at least one machine learning model by taking the historical sales data based on the time sequence and external environment variable data as training data, and obtaining optimal parameters and corresponding training errors of each time sequence analysis model and each machine learning model by respectively utilizing a grid search method;
and 3, determining a plurality of sales forecasting models with the training errors smaller than a preset value, forecasting sales of products to be forecasted based on time to be forecasted and according to the determined plurality of sales forecasting models and the corresponding optimal parameters thereof to obtain a plurality of sales forecasting values, and determining a final sales forecasting value according to the plurality of sales forecasting values.
2. The sales prediction method of claim 1, further comprising:
step 4, in a machine learning model for sales prediction, acquiring the importance degree of each feature in external environment variable data according to an import method;
and 5, normalizing the importance degrees of the features, calculating the average value of the importance degrees of the features corresponding to the machine learning models, and sequencing and outputting the features according to the average value of the importance degrees of the features.
3. The sales prediction method of claim 1, wherein the external environment variable data comprises: mission data, auction data, macro-economic data, and/or air temperature data, the macro-economic data comprising: and the residential consumption price index and/or the real estate construction area, wherein each type of data is corresponding characteristics.
4. The sales prediction method of claim 1, wherein the step 1 further comprises:
drawing a historical sales volume graph of the product to be predicted according to historical sales volume data of the product to be predicted based on the time sequence, determining a statistical rule according to the historical sales volume graph, performing sales volume prediction on the product to be predicted based on the time to be predicted and according to the statistical rule to obtain a statistical prediction value, and adding the statistical prediction value into external environment variable data as a statistical characteristic.
5. The sales prediction method of claim 1, wherein the step 1 further comprises:
and based on the time sequence, eliminating historical sales data and external environment variable data with more missing values and a variance close to zero, and filling the data with less missing values by adopting a mean value or neighbor method.
6. The sales prediction method of claim 1, wherein in step 3, for the machine learning model of the determined sales prediction models, external environment variable data corresponding to the time to be predicted is predicted based on the time to be predicted, and sales prediction is performed on the product to be predicted according to the time to be predicted and the external environment variable data corresponding to the time to be predicted and based on the machine learning model.
7. The sales prediction method of claim 1, wherein the time series analysis model comprises: an STL model, an ETS model, an ARIMA model, and/or a SARIMA model, the machine learning model comprising: a GBM model, an XGBOST model, an RF model, a GBDT model, and/or a CATBOOST model.
8. The method of predicting sales of claim 1, wherein the method of determining a final sales forecast based on a plurality of sales forecasts comprises:
and calculating the average value of the plurality of sales predicted values to obtain the final sales predicted value.
9. A sales prediction system, comprising: the device comprises an acquisition unit, a training unit and a determination unit;
the acquisition unit is used for acquiring historical sales data and external environment variable data of a product to be predicted based on a time sequence, wherein the external environment variable data is data influencing the historical sales of the product to be predicted;
the training unit is used for training at least one time sequence analysis model by taking the historical sales data based on the time sequence as training data, training at least one machine learning model by taking the historical sales data based on the time sequence and external environment variable data as training data, and obtaining optimal parameters and corresponding training errors of each time sequence analysis model and each machine learning model by respectively utilizing a grid search method;
the determining unit is used for determining a plurality of models with the training errors smaller than a preset value, carrying out sales forecast on products to be forecasted based on the time to be forecasted and according to the determined plurality of models and the corresponding optimal parameters thereof to obtain a plurality of sales forecast values, and determining the final sales forecast value according to the plurality of sales forecast values.
10. A sales prediction terminal, wherein the terminal comprises a processor, a memory, and a communication bus;
the communication bus is used for realizing connection communication between the processor and the memory;
the processor is configured to execute one or more programs in the memory to implement the steps of the sales prediction method of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010782742.XA CN111882363A (en) | 2020-08-06 | 2020-08-06 | Sales prediction method, system and terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010782742.XA CN111882363A (en) | 2020-08-06 | 2020-08-06 | Sales prediction method, system and terminal |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111882363A true CN111882363A (en) | 2020-11-03 |
Family
ID=73210884
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010782742.XA Pending CN111882363A (en) | 2020-08-06 | 2020-08-06 | Sales prediction method, system and terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111882363A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112613719A (en) * | 2020-12-17 | 2021-04-06 | 南京光普信息技术有限公司 | Intelligent ordering system and method for cooked food store |
| CN113743643A (en) * | 2021-02-05 | 2021-12-03 | 北京京东振世信息技术有限公司 | Method, device, equipment and medium for determining commodity data prediction accuracy |
| CN113743721A (en) * | 2021-07-29 | 2021-12-03 | 深圳市东信时代信息技术有限公司 | Marketing strategy generation method and device, computer equipment and storage medium |
-
2020
- 2020-08-06 CN CN202010782742.XA patent/CN111882363A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112613719A (en) * | 2020-12-17 | 2021-04-06 | 南京光普信息技术有限公司 | Intelligent ordering system and method for cooked food store |
| CN113743643A (en) * | 2021-02-05 | 2021-12-03 | 北京京东振世信息技术有限公司 | Method, device, equipment and medium for determining commodity data prediction accuracy |
| CN113743643B (en) * | 2021-02-05 | 2023-11-03 | 北京京东振世信息技术有限公司 | Method, device, equipment and medium for determining commodity data prediction accuracy |
| CN113743721A (en) * | 2021-07-29 | 2021-12-03 | 深圳市东信时代信息技术有限公司 | Marketing strategy generation method and device, computer equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108364085B (en) | Takeout delivery time prediction method and device | |
| CN111882363A (en) | Sales prediction method, system and terminal | |
| CN102890803B (en) | The defining method of the abnormal process of exchange of electronic goods and device thereof | |
| CN111563706A (en) | Multivariable logistics freight volume prediction method based on LSTM network | |
| CN108629436B (en) | Method and electronic equipment for estimating warehouse goods picking capacity | |
| CN110689190A (en) | Power grid load prediction method and device and related equipment | |
| CN112712203B (en) | Day-highest load prediction method and system for power distribution network | |
| CN111680841B (en) | Short-term load prediction method, system and terminal equipment based on principal component analysis | |
| CN114169434A (en) | Load prediction method | |
| JPH04372046A (en) | Method and device for predicting demand amount | |
| CN112508254B (en) | Method for determining investment prediction data of transformer substation engineering project | |
| CN116826745B (en) | Layered and partitioned short-term load prediction method and system in power system background | |
| CN113962504A (en) | Power distribution network planning scheme cost-benefit ratio calculation method | |
| CN112330030A (en) | System and method for predicting requirements of expansion materials | |
| CN116384251A (en) | New energy generated power combination prediction method and system considering risk avoidance | |
| CN113962638B (en) | Intelligent fruit and vegetable quality detection method and system | |
| Kanwal et al. | An attribute weight estimation using particle swarm optimization and machine learning approaches for customer churn prediction | |
| CN107291722B (en) | Descriptor classification method and device | |
| CN114202174A (en) | Electricity price risk grade early warning method and device and storage medium | |
| CN115310648A (en) | Medium-and-long-term wind power combination prediction method based on multi-meteorological variable model identification | |
| CN112668784A (en) | Regional macro economy prediction model and method based on big data | |
| CN114169763A (en) | Measuring instrument demand prediction method, system, computing device and storage medium | |
| CN113919610A (en) | ARIMA model construction method and evaluation method for low-voltage transformer area line loss prediction | |
| JP2020115276A (en) | Information processing method, information processing device, program, and generation method of learned model | |
| CN117010929B (en) | Agricultural product public opinion information construction method |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20201103 |
|
| RJ01 | Rejection of invention patent application after publication |