Data relationship determination method and device, storage medium and electronic equipment
Through data processing based on feature dimensions and pre-trained model fitting, the difficulty of price elasticity calculation of B-end customers is solved, which improves efficiency and reduces costs.
Patent Information
- Application Number
- CN202310664445.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-06-06
- Publication Date
- 2024-12-06
AI Technical Summary
It is difficult to effectively calculate the price elasticity of B-end customers in the prior art, especially due to the small number of customers and the long settlement cycle, it is difficult to draw conclusions in a timely manner, and the cost is high and there are many influencing factors, making it difficult to attribute changes in business volume.
By acquiring feature data based on the feature dimensions associated with the target user, using the pre-trained single-quantity prediction model to generate prediction data, and fit the historical and prediction data, determining the data traffic relationship, and expanding the traffic data to improve computing efficiency.
The efficiency of price elasticity calculation of B-end customers is improved, and the problem of difficult relationships due to small business volume data is avoided, which reduces the risk of customer churn and the overall cost.
Smart Images

Figure CN119090523A_ABST
Abstract
Description
Background Art
[0002] Among the current methods for calculating the price elasticity of customers, the elasticity calculation method for C-end customers is relatively mature, and usually adopts the price experiment method. Since the number of C-end customers is large, different types of customers can be selected to offer different prices, and the conversion rates under different prices can be counted, and then the price elasticity of C-end customers can be calculated. There is currently no scientific and effective calculation method for the price elasticity of B-end customers.
[0003] For B-end customers, the number of customers is small, the settlement cycle is long, and there are many influencing factors. If a price experiment method similar to that for C-end customers is applied, the risk of customer churn is high and the overall cost is too high. Due to the long experimental cycle, it is difficult to draw timely and effective conclusions. There are many influencing factors, and it is difficult to attribute changes in business volume and explain changes in business volume affected by prices. C-end experimental methods are difficult to apply.
[0004] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention
[0005] The purpose of the present disclosure is to provide a method, device, storage medium and electronic device for determining a data relationship, thereby overcoming the problem of difficulty in storing an extremely long character string at least to a certain extent.
[0006] According to one aspect of the present disclosure, a method for determining a data relationship is provided, comprising:
[0007] According to the characteristic dimensions associated with the target user, characteristic data of the target user under different characteristic dimensions are obtained, and a first business volume corresponding to the target user under historical data is obtained;
[0008] Using a pre-trained order volume prediction model, the feature data under the different feature dimensions are processed to generate a second business volume corresponding to the target user under the predicted data; the pre-trained order volume prediction model is obtained by training the initial order volume prediction model;
[0009] The historical data and the corresponding first traffic volume as well as the predicted data and the corresponding second traffic volume are fitted to determine a data traffic relationship corresponding to the target user.
[0010] Optionally, the acquiring feature data of the target user in different feature dimensions includes:
[0011] Acquire first feature data of the target user under a preset feature dimension;
[0012] Acquire second characteristic data corresponding to a user having a preset relationship with the target user.
[0013] Optionally, the fitting the historical data and the corresponding first traffic volume and the predicted data and the corresponding second traffic volume to determine the data traffic volume relationship corresponding to the target user includes:
[0014] Using a preset fitting method, fitting the historical data and the corresponding first business volume and the predicted data and the corresponding second business volume is performed to obtain a target fitting function of the target user;
[0015] The data traffic relationship corresponding to the target user is determined according to the parameters in the target fitting function.
[0016] Optionally, when the target user is a new user, obtaining feature data of the target user in different feature dimensions includes:
[0017] Selecting other users with similar feature dimensions;
[0018] The available features of the other users in the different feature dimensions are used as feature data of the target user.
[0019] Optionally, the obtaining the first service volume corresponding to the target user under historical data includes:
[0020] According to the available traffic volumes corresponding to the other users under the available data, a first traffic volume corresponding to the target user under the historical data is determined.
[0021] Optionally, the method further includes:
[0022] Obtaining sample feature data of sample users under different feature dimensions and sample traffic corresponding to the sample data, as well as the relationship between the real data traffic corresponding to the sample users;
[0023] The sample feature data, sample data and the corresponding sample traffic volume and the real data traffic volume relationship are used as a training sample pair;
[0024] The initial single volume prediction model is iteratively trained using training sample pairs to obtain the pre-trained single volume prediction model whose sample business volume output under different sample prediction data is consistent with the real data business volume relationship.
[0025] Optionally, the sample data includes sample historical data and sample prediction data, the sample traffic includes a first sample traffic and a second sample traffic, and the acquiring of the sample feature data of the sample user under different feature dimensions and the sample traffic corresponding to the sample data, and the relationship between the real data traffic corresponding to the sample user, includes:
[0026] Determine sample feature data of the sample user under the different feature dimensions, and determine a first sample business volume corresponding to the sample user under the sample historical data;
[0027] According to a preset business volume prediction method, obtaining a second sample business volume corresponding to the sample user under the sample prediction data;
[0028] The sample historical data and the corresponding first sample traffic volume as well as the sample predicted data and the corresponding second sample traffic volume are fitted to obtain the real data traffic relationship corresponding to the sample user.
[0029] According to one aspect of the present disclosure, a data relationship determination device is provided, characterized by comprising:
[0030] A first acquisition module, configured to acquire feature data of the target user under different feature dimensions according to feature dimensions associated with the target user, and to acquire a first business volume corresponding to the target user under historical data;
[0031] A generating module, used to process the feature data under the different feature dimensions by using a pre-trained order volume prediction model, and generate a second business volume corresponding to the target user under the predicted data; the pre-trained order volume prediction model is obtained by training the initial order volume prediction model;
[0032] A fitting module is used to fit the historical data and the corresponding first business volume and the predicted data and the corresponding second business volume to determine the data business volume relationship corresponding to the target user.
[0033] Optionally, the first acquisition module is further used to:
[0034] Acquire first feature data of the target user under a preset feature dimension;
[0035] Acquire second characteristic data corresponding to a user having a preset relationship with the target user.
[0036] Optionally, the fitting module is further used for:
[0037] Using a preset fitting method, fitting the historical data and the corresponding first business volume and the predicted data and the corresponding second business volume is performed to obtain a target fitting function of the target user;
[0038] The data traffic relationship corresponding to the target user is determined according to the parameters in the target fitting function.
[0039] Optionally, when the target user is a new user, the first acquisition module is further used to:
[0040] Selecting other users with similar feature dimensions;
[0041] The available features of the other users in the different feature dimensions are used as feature data of the target user.
[0042] Optionally, the first acquisition module is further used to:
[0043] According to the available traffic volumes corresponding to the other users under the available data, a first traffic volume corresponding to the target user under the historical data is determined.
[0044] Optionally, the device further comprises:
[0045] A second acquisition module is used to acquire sample feature data of sample users under different feature dimensions and sample traffic corresponding to the sample data, as well as a relationship between the real data traffic corresponding to the sample users;
[0046] A determination module, configured to use the sample feature data, the sample data and the corresponding sample traffic volume as a training sample pair with the real data traffic volume relationship;
[0047] The training module is used to iteratively train the initial single volume prediction model using training samples to obtain the pre-trained single volume prediction model whose sample business volume output under different sample prediction data conforms to the real data business volume relationship.
[0048] Optionally, the sample data includes sample historical data and sample prediction data, the sample traffic includes a first sample traffic and a second sample traffic, and the second acquisition module is further used to:
[0049] Determine sample feature data of the sample user under the different feature dimensions, and determine a first sample business volume corresponding to the sample user under the sample historical data;
[0050] According to a preset business volume prediction method, obtaining a second sample business volume corresponding to the sample user under the sample prediction data;
[0051] The sample historical data and the corresponding first sample traffic volume as well as the sample predicted data and the corresponding second sample traffic volume are fitted to obtain the real data traffic relationship corresponding to the sample user.
[0052] According to one aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the method for determining a data relationship described in any one of the above is implemented.
[0053] According to one aspect of the present disclosure, there is provided an electronic device, including:
[0054] Processor; and
[0055] A memory, configured to store executable instructions of the processor;
[0056] The processor is configured to execute any one of the above-mentioned methods for determining a data relationship by executing the executable instructions.
[0057] In summary, the data relationship determination method provided by the embodiment of the present disclosure can obtain the feature data of the target user under different feature dimensions according to the feature dimensions associated with the target user, and obtain the first business volume corresponding to the target user under historical data, and use the pre-trained single volume prediction model to process the feature data under different feature dimensions to generate the second business volume corresponding to the target user under the predicted data. The pre-trained single volume prediction model is obtained by training the initial single volume prediction model, and the historical data and the corresponding first business volume and the predicted data and the corresponding second business volume are fitted to determine the data-business volume relationship corresponding to the target user. In this way, by increasing the business volume corresponding to the target user under the predicted data through the single volume prediction model, the business volume data corresponding to the target user can be expanded, avoiding the problem of difficulty in determining the data-business volume relationship of the target user due to the small amount of business volume data corresponding to the target user, thereby improving the efficiency of determining the user data-business volume relationship.
[0058] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS
[0059] The accompanying drawings herein are incorporated into the specification and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the specification are used to explain the principles of the present disclosure. Obviously, the accompanying drawings described below are only some embodiments of the present disclosure, and for ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without creative work.
[0060] Figure 1is a flowchart of a method for determining a data relationship provided by an embodiment of the present disclosure;
[0061] Figure 2 is a flow chart of acquiring feature data provided by an embodiment of the present disclosure;
[0062] Figure 3 is a flow chart of determining a data traffic relationship provided by an embodiment of the present disclosure;
[0063] Figure 4 is a flow chart of obtaining a pre-trained single quantity prediction model provided by an embodiment of the present disclosure;
[0064] Figure 5 is a block diagram of a data relationship determination device provided by an embodiment of the present disclosure;
[0065] Figure 6 An electronic device provided by an embodiment of the present disclosure is used to implement the above-mentioned data relationship determination method. DETAILED DESCRIPTION
[0066] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as being limited to the examples set forth herein; on the contrary, these embodiments are provided so that the present disclosure will be more comprehensive and complete, and the concepts of the example embodiments are fully conveyed to those skilled in the art. The described features, structures, or characteristics may be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure may be practiced while omitting one or more of the specific details, or other methods, components, devices, steps, etc. may be adopted. In other cases, known technical solutions are not shown or described in detail to avoid obscuring various aspects of the present disclosure.
[0067] In addition, the accompanying drawings are only schematic illustrations of the present disclosure and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor devices and / or microcontroller devices.
[0068] Figure 1 is a flowchart of a method for determining a data relationship provided by an embodiment of the present disclosure. Figure 1 As shown, the method may include:
[0069] Step S101: acquiring feature data of the target user in different feature dimensions according to feature dimensions associated with the target user, and acquiring a first business volume corresponding to the target user in historical data.
[0070] In the embodiment of the present disclosure, the characteristic dimension associated with the target user may be a characteristic dimension related to the target user's business processing. For example, when the target user's business processing is to send goods, the characteristic dimension associated with the target user may be a characteristic dimension related to the delivery order quantity, delivery service, delivery time, etc. According to the characteristic dimension associated with the target user, the characteristic data of the target user under different characteristic dimensions are obtained, which may be to use the data corresponding to the target user under the associated characteristic dimension as the characteristic data of the target user. The first business volume corresponding to the target user under the historical data is obtained, which may be the first business volume executed by the target user under the historical data in the historical time, wherein the historical data may be the data accepted by the target user when processing the business, for example, the historical data may be the price purchased by the target user before, the status of the previously processed data, etc. The first business volume may be the purchase volume of the target user under the historical data, or the number of executions of the target user under the historical data, etc. The number of historical data may be multiple, and multiple corresponding first business volumes may be obtained. The target user may refer to a B-end (Business) customer, that is, the target user may be a large-scale customer such as an enterprise customer.
[0071] Step S102: Use a pre-trained order volume prediction model to process the feature data under the different feature dimensions to generate a second business volume corresponding to the target user under the predicted data; the pre-trained order volume prediction model is obtained by training the initial order volume prediction model.
[0072] In the disclosed embodiment, the pre-trained single quantity prediction model can be obtained by pre-training the initial single quantity prediction model using the sample feature data, sample data and sample business volume of the sample user. Through continuous iterative training, the single quantity prediction model can learn to predict the corresponding business volume under different sample data in accordance with the real data business volume relationship based on the input feature data. Among them, the single quantity prediction model can be a convolutional neural network (CNN), a gradient boosting decision tree (GBDT) network, or other machine learning networks. For example, the single quantity prediction model can be a gradient boosting model (Extreme Gradient Boosting, XGBoost).
[0073] In the embodiment of the present disclosure, a pre-trained order volume prediction model is used to process feature data under different feature dimensions to generate a second business volume corresponding to the target user under the predicted data. The feature data under different feature dimensions can be input into the pre-trained order volume prediction model to predict the second business volume corresponding to the target user under the predicted data. The predicted data can be manually pre-set or randomly generated, wherein the number of predicted data can be multiple, and each predicted data can correspond to a second business volume.
[0074] Step S103: Fit the historical data and the corresponding first traffic volume and the predicted data and the corresponding second traffic volume to determine a data traffic relationship corresponding to the target user.
[0075] In the disclosed embodiment, the historical data and the first business volume may be taken as an array, the predicted data and the second business volume may be taken as an array, the corresponding fitting function may be obtained by fitting the above multiple arrays, and the corresponding data-business volume relationship may be determined according to the fitting function, wherein the fitting may be linear fitting or nonlinear fitting. The fitting function may be a functional relationship between data and business volume, and the parameters in the fitting function may be used as the data-business volume relationship.
[0076] In summary, the data relationship determination method provided by the embodiment of the present disclosure can obtain the feature data of the target user under different feature dimensions according to the feature dimensions associated with the target user, and obtain the first business volume corresponding to the target user under historical data, and use the pre-trained single volume prediction model to process the feature data under different feature dimensions to generate the second business volume corresponding to the target user under the predicted data. The pre-trained single volume prediction model is obtained by training the initial single volume prediction model, and the historical data and the corresponding first business volume and the predicted data and the corresponding second business volume are fitted to determine the data-business volume relationship corresponding to the target user. In this way, by increasing the business volume corresponding to the target user under the predicted data through the single volume prediction model, the business volume data corresponding to the target user can be expanded, avoiding the problem of difficulty in determining the data-business volume relationship of the target user due to the small amount of business volume data corresponding to the target user, thereby improving the efficiency of determining the user data-business volume relationship.
[0077] Optionally, in the embodiment of the present disclosure, the above operation of obtaining the characteristic data of the target user in different characteristic dimensions is as follows: Figure 2 As shown, it may specifically include:
[0078] Step S1011: Acquire first feature data of the target user under a preset feature dimension.
[0079] In the embodiment of the present disclosure, the preset feature dimension may be a feature dimension pre-set according to actual conditions, and the preset feature dimension may be a dimension determined based on the user's historical operating habits, the service conditions provided by the processing business, the data content of competitors in the same industry, etc. The first feature data of the target user under the preset feature dimension is obtained by obtaining the first feature data of the target user under the preset feature dimension for each preset feature dimension, and a plurality of first feature data are obtained, and one preset feature dimension may correspond to a plurality of first feature data.
[0080] For example, the preset feature dimension can be historical operation habits. Since the customer's delivery behavior often has a certain inertia, it can help predict the user's order volume next month. The first feature data corresponding to the preset feature dimension can include the customer's historical delivery volume, delivery route, consignment, delivery weight, etc. The preset feature dimension can be contract data. Since the customer contract data is the customer's commitment behavior, it is related to the customer's own credit evaluation. The first feature data corresponding to the preset feature dimension can include the customer's contract price, promised order volume, industry attributes, etc. The preset feature dimension can be service quality. Since the platform's service quality of the product will have a certain impact on whether the customer uses it, the first feature data corresponding to the preset feature dimension can include express delivery time, collection time, complaint rate, etc. The preset feature dimension can be a time feature. Since customer delivery is seasonal, it is necessary to consider the impact of the peak and off-peak seasons, holidays, and promotion dates of the customer's corresponding industry. The first feature data corresponding to the preset feature dimension can be the shipment volume of the corresponding season.
[0081] Step S1012: Acquire second characteristic data corresponding to a user who has a preset relationship with the target user.
[0082] In the embodiment of the present disclosure, the preset relationship may be a competitive relationship, and obtaining the second characteristic data corresponding to the user having the preset relationship with the target user may be to first determine the user having the competitive relationship with the target user, and then obtain the characteristic data corresponding to the user having the competitive relationship as the second characteristic data of the target user. For example, user B having the preset relationship with target user A needs to refer to the data of the competitive user because the customer will select the supplier based on merit, and accordingly, the second characteristic data may include the competitive price, service quality, etc.
[0083] Optionally, in the embodiment of the present disclosure, the operation of fitting the historical data and the corresponding first traffic volume and the predicted data and the corresponding second traffic volume to determine the data traffic relationship corresponding to the target user is as follows: Figure 3 As shown, it may specifically include:
[0084] Step S1031: Use a preset fitting method to fit the historical data and the corresponding first business volume, and the predicted data and the corresponding second business volume to obtain a target fitting function of the target user.
[0085] In the embodiment of the present disclosure, the preset fitting method can be a predetermined linear fitting method, and the historical data and the first business volume can be a pair of arrays, and the predicted data and the second business volume can be a pair of arrays. A fitting relationship is established according to the above arrays to obtain a target fitting function. Among them, since it is necessary to ensure that the data-business volume relationship is a linear relationship, the data and the business volume can be subjected to a natural logarithm operation. For example, the target fitting function can be expressed as follows:
[0086] ln(traffic volume) = k·ln(data) + b
[0087] Among them, k can be a data traffic relationship, and b can be a constant determined according to the fitting relationship, which is not limited in the present disclosure.
[0088] Step S1032: Determine the data traffic relationship corresponding to the target user according to the parameters in the target fitting function.
[0089] In the embodiment of the present disclosure, since the data traffic relationship can be used to characterize the degree to which the target user's traffic changes due to data changes, the parameters of the target fitting function for the data can be used as the data traffic relationship corresponding to the target user. For example, the parameter k in the above target fitting function can be used as the data traffic relationship.
[0090] Optionally, in an implementation manner, when the target user is a new user, the operation of obtaining the feature data of the target user in different feature dimensions in the embodiment of the present disclosure may specifically include:
[0091] Selecting other users who are similar to the feature dimensions; and using the available features of the other users under the different feature dimensions as feature data of the target user.
[0092] In the disclosed embodiment, since new users often lack historical data, other users with similar feature dimensions to the target user may be selected. The other users may be users in the same industry as the target user.
[0093] In the embodiment of the present disclosure, using the available features of other users in different feature dimensions as the feature data of the target user may be performed by first determining the available features of other users in different feature dimensions and using the available features as the feature data of the target user.
[0094] Optionally, in the embodiment of the present disclosure, the operation of obtaining the first service volume corresponding to the target user under the historical data may specifically include:
[0095] According to the available traffic volumes corresponding to the other users under the available data, a first traffic volume corresponding to the target user under the historical data is determined.
[0096] In an embodiment of the present disclosure, the available business volume corresponding to other users under the available data may be determined first, and the average of the available business volumes of multiple other users may be taken as the first business volume corresponding to the target user under the historical data, wherein the available business volume of other users may be the average of the historical order volumes of other users, or the average order volume of customers in the same industry, or the average order volume of customers in the same industry and region, or the average order volume in the same industry and region during the same period, and the present disclosure does not impose any limitation on this.
[0097] It should be noted that, when collating data, it is found that historical data is often distributed in a scattered manner. For example, the price ranges of express products in different industries are not the same. The prices of small commodities are lower, while the prices of the furniture industry are higher. The full price range is larger, and most users cannot cover the full price range. Therefore, historical data and predicted data can be represented according to a preset ratio value, so that the data representation is more concentrated, which is convenient for subsequent data prediction and processing. Since the business volume is also distributed in a scattered manner, most customers are small and medium-sized customers, and the order volume is relatively concentrated. A small number of customers are extremely large, resulting in very scattered data, making it difficult to estimate outlier data. Therefore, the first business volume and the second business volume can be represented according to the preset logarithmic value, so that the business volume data is more concentrated, so as to facilitate subsequent data fitting and other processing.
[0098] Optionally, in another implementation, such as Figure 4 As shown, the data relationship determination method in the embodiment of the present disclosure may also include:
[0099] Step S201: Obtain sample feature data of sample users in different feature dimensions and sample traffic corresponding to the sample data, as well as the relationship between the real data traffic corresponding to the sample users.
[0100] In the disclosed embodiment, sample feature data corresponding to sample users under different feature dimensions may be obtained based on preset feature dimensions, and then the sample business volume corresponding to the sample users under the sample data may be obtained, and the real data business volume relationship corresponding to the sample users may be determined based on the sample business volume corresponding to the sample data.
[0101] Step S202: The sample feature data, sample data, corresponding sample traffic volume and the real data traffic volume relationship are taken as a training sample pair.
[0102] For example, the sample feature data may include the number of shipments, fulfillment time, and service quality. The sample business volume corresponding to the sample data may be the transaction volume corresponding to the price. The real data business volume relationship may be X1. The sample feature data, shipment number, fulfillment time, and service quality may be used as input quantities, the transaction volume corresponding to the price may be used as the learning quantity of the order quantity prediction model, and the real data business volume relationship may be used as the real label, so that the order quantity prediction model can learn to predict the sample business volume that conforms to the real data business volume relationship based on different sample data.
[0103] Step S203: iteratively train the initial order volume prediction model using training samples to obtain the pre-trained order volume prediction model whose sample business volume output under different sample prediction data is consistent with the real data business volume relationship.
[0104] In the disclosed embodiments, the sample feature data in the training sample pairs may be input into the initial single quantity prediction model for calculation and processing, and the relationship between the sample prediction data and the sample business volume may be learned, so that the sample business volume that is consistent with the real data business volume relationship may be output under different sample prediction data. If the output sample business volume does not conform to the real data business volume relationship, the training parameters in the initial single quantity prediction model are adjusted, and the sample feature data is re-input into the single quantity prediction model for iterative training until the output sample business volume conforms to the real data business volume relationship. The single quantity prediction model whose output sample business volume conforms to the real data business volume relationship is used as the pre-trained single quantity prediction model.
[0105] Optionally, in the embodiment of the present disclosure, the sample data may include sample historical data and sample prediction data, the sample traffic may include a first sample traffic and a second sample traffic, and the operation of obtaining the sample feature data of the sample user under different feature dimensions and the sample traffic corresponding to the sample data, and the relationship between the real data traffic corresponding to the sample user may specifically include:
[0106] First, the sample feature data of the sample user under the different feature dimensions is determined, and the first sample business volume corresponding to the sample user under the sample historical data is determined.
[0107] In the embodiment of the present disclosure, sample feature data corresponding to the sample user under the preset feature dimension may be determined according to the preset feature dimension, and the first sample business volume corresponding to the sample user under the sample historical data may be obtained from the historical records of the sample user.
[0108] Secondly, according to a preset business volume prediction method, a second sample business volume corresponding to the sample user under the sample prediction data is obtained.
[0109] In the embodiment of the present disclosure, the preset business volume prediction method can be a predicted business volume set manually, or a business volume predicted by a single volume prediction model, which is not limited in the present disclosure. According to the preset business volume prediction method, obtaining the second sample business volume corresponding to the sample user under the sample prediction data can be the sample prediction data of the sample user, and obtaining the predicted second sample business volume by using the preset business volume prediction method.
[0110] Finally, the sample historical data and the corresponding first sample traffic volume as well as the sample predicted data and the corresponding second sample traffic volume are fitted to obtain the real data traffic relationship corresponding to the sample user.
[0111] In the embodiment of the present disclosure, the sample historical data and the first sample business volume can be taken as an array, the sample prediction data and the second sample business volume can be taken as an array, and the corresponding fitting function can be obtained by fitting the above multiple arrays, and then the real data business volume relationship corresponding to the sample user can be determined based on the fitting function.
[0112] Figure 5 is a block diagram of a data relationship determination device provided by an embodiment of the present disclosure, such as Figure 5 As shown, the device 30 may include:
[0113] A first acquisition module 301 is used to acquire feature data of the target user in different feature dimensions according to feature dimensions associated with the target user, and to acquire a first business volume corresponding to the target user in historical data;
[0114] A generating module 302 is used to process the feature data under the different feature dimensions by using a pre-trained order volume prediction model to generate a second business volume corresponding to the target user under the prediction data; the pre-trained order volume prediction model is obtained by training the initial order volume prediction model;
[0115] The fitting module 303 is used to fit the historical data and the corresponding first traffic volume and the predicted data and the corresponding second traffic volume to determine the data traffic relationship corresponding to the target user.
[0116] In summary, the data relationship determination device provided by the embodiment of the present disclosure can obtain the feature data of the target user under different feature dimensions according to the feature dimensions associated with the target user, and obtain the first business volume corresponding to the target user under historical data, and use the pre-trained single volume prediction model to process the feature data under different feature dimensions to generate the second business volume corresponding to the target user under the predicted data. The pre-trained single volume prediction model is obtained by training the initial single volume prediction model, and the historical data and the corresponding first business volume and the predicted data and the corresponding second business volume are fitted to determine the data-business volume relationship corresponding to the target user. In this way, by increasing the business volume corresponding to the target user under the predicted data through the single volume prediction model, the business volume data corresponding to the target user can be expanded, avoiding the problem of difficulty in determining the data-business volume relationship of the target user due to the small amount of business volume data corresponding to the target user, thereby improving the efficiency of determining the user data-business volume relationship.
[0117] Optionally, the first acquisition module 301 is further used to:
[0118] Acquire first feature data of the target user under a preset feature dimension;
[0119] Acquire second characteristic data corresponding to a user having a preset relationship with the target user.
[0120] Optionally, the fitting module 303 is further used for:
[0121] Using a preset fitting method, fitting the historical data and the corresponding first business volume and the predicted data and the corresponding second business volume is performed to obtain a target fitting function of the target user;
[0122] The data traffic relationship corresponding to the target user is determined according to the parameters in the target fitting function.
[0123] Optionally, when the target user is a new user, the first acquisition module 301 is further used to:
[0124] Selecting other users with similar feature dimensions;
[0125] The available features of the other users in the different feature dimensions are used as feature data of the target user.
[0126] Optionally, the first acquisition module 301 is further used to:
[0127] According to the available traffic volumes corresponding to the other users under the available data, a first traffic volume corresponding to the target user under the historical data is determined.
[0128] Optionally, the device 30 further includes:
[0129] A second acquisition module is used to acquire sample feature data of sample users under different feature dimensions and sample traffic corresponding to the sample data, as well as a relationship between the real data traffic corresponding to the sample users;
[0130] A determination module, configured to use the sample feature data, the sample data and the corresponding sample traffic volume as a training sample pair with the real data traffic volume relationship;
[0131] The training module is used to iteratively train the initial single volume prediction model using training samples to obtain the pre-trained single volume prediction model whose sample business volume output under different sample prediction data conforms to the real data business volume relationship.
[0132] Optionally, the sample data includes sample historical data and sample prediction data, the sample traffic includes a first sample traffic and a second sample traffic, and the second acquisition module is further used to:
[0133] Determine sample feature data of the sample user under the different feature dimensions, and determine a first sample business volume corresponding to the sample user under the sample historical data;
[0134] According to a preset business volume prediction method, obtaining a second sample business volume corresponding to the sample user under the sample prediction data;
[0135] The sample historical data and the corresponding first sample traffic volume as well as the sample predicted data and the corresponding second sample traffic volume are fitted to obtain the real data traffic relationship corresponding to the sample user.
[0136] The specific details of each module in the above-mentioned data relationship determination device have been described in detail in the corresponding data relationship determination method, so they will not be repeated here.
[0137] It should be noted that, although several modules or units of the device for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into multiple modules or units to be embodied.
[0138] In addition, although the steps of the method in the present disclosure are described in a specific order in the drawings, this does not require or imply that the steps must be performed in this specific order, or that all the steps shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps, etc.
[0139] In an exemplary embodiment of the present disclosure, an electronic device capable of implementing the above method is also provided.
[0140] Those skilled in the art will appreciate that various aspects of the present disclosure may be implemented as systems, methods or program products. Therefore, various aspects of the present disclosure may be specifically implemented in the following forms, namely: complete hardware implementation, complete software implementation (including firmware, microcode, etc.), or a combination of hardware and software, which may be collectively referred to herein as "circuits", "modules" or "systems".
[0141] Refer to the following Figure 6 The electronic device 400 according to this embodiment of the present disclosure is described. Figure 6 The electronic device 400 shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.
[0142] like Figure 6 As shown, the electronic device 400 is in the form of a general computing device. The components of the electronic device 400 may include, but are not limited to: the at least one processing unit 410, the at least one storage unit 420, a bus 430 connecting different system components (including the storage unit 420 and the processing unit 410), and a display unit 440.
[0143] The storage unit stores program codes, which can be executed by the processing unit 410, so that the processing unit 410 performs the steps according to various exemplary embodiments of the present disclosure described in the above “Exemplary Method” section of this specification. For example, the processing unit 410 can perform the following steps: Figure 1Step S101 shown in the figure, according to the feature dimensions associated with the target user, obtain the feature data of the target user under different feature dimensions, and obtain the first business volume corresponding to the target user under historical data; step S102, use the pre-trained order volume prediction model to process the feature data under the different feature dimensions, and generate the second business volume corresponding to the target user under the predicted data; the pre-trained order volume prediction model is obtained by training the initial order volume prediction model; step S103, fit the historical data and the corresponding first business volume and the predicted data and the corresponding second business volume to determine the data business volume relationship corresponding to the target user.
[0144] The storage unit 420 may include a readable medium in the form of a volatile storage unit, such as a random access storage unit (RAM) 4201 and / or a cache storage unit 4202 , and may further include a read-only storage unit (ROM) 4203 .
[0145] The storage unit 420 may also include a program / utility 4204 having a set (at least one) of program modules 4205, such program modules 4205 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.
[0146] Bus 430 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.
[0147] The electronic device 400 may also communicate with one or more external devices 500 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 400, and / or communicate with any device that enables the electronic device 400 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed via an input / output (I / O) interface 450. Furthermore, the electronic device 400 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via a network adapter 460. As shown, the network adapter 460 communicates with other modules of the electronic device 400 via a bus 430. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 400, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.
[0148] Through the description of the above implementation, it is easy for those skilled in the art to understand that the example implementation described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the implementation of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the implementation of the present disclosure.
[0149] In an exemplary embodiment of the present disclosure, a computer-readable storage medium is also provided, on which a program product capable of implementing the above method of the present specification is stored. In some possible implementations, various aspects of the present disclosure may also be implemented in the form of a program product, which includes a program code, and when the program product is run on a terminal device, the program code is used to enable the terminal device to execute the steps according to various exemplary implementations of the present disclosure described in the above "Exemplary Method" section of the present specification.
[0150] According to the program product for implementing the above method in the embodiment of the present disclosure, it can adopt a portable compact disk read-only memory (CD-ROM) and include program code, and can be run on a terminal device, such as a personal computer. However, the program product of the present disclosure is not limited thereto. In this document, a readable storage medium can be any tangible medium containing or storing a program, which can be used by or in combination with an instruction execution system, an apparatus or a device.
[0151] The program product may use any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.
[0152] Computer readable signal media may include data signals propagated in baseband or as part of a carrier wave, in which readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Readable signal media may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.
[0153] The program code embodied on the readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the foregoing.
[0154] Program code for performing the operations of the present disclosure may be written in any combination of one or more programming languages, including object-oriented programming languages such as Java, C++, etc., and conventional procedural programming languages such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., through the Internet using an Internet service provider).
[0155] In addition, the above-mentioned figures are only schematic illustrations of the processes included in the method according to the exemplary embodiments of the present disclosure, and are not intended to be limiting. It is easy to understand that the processes shown in the above-mentioned figures do not indicate or limit the time sequence of these processes. In addition, it is also easy to understand that these processes can be performed synchronously or asynchronously, for example, in multiple modules.
[0156] Other embodiments of the present disclosure will be readily apparent to those skilled in the art after considering the specification and practicing the inventions invented herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not invented by the present disclosure. The specification and examples are to be considered merely exemplary, and the true scope and spirit of the present disclosure are indicated by the claims.
Claims
1. A method for determining a data relationship, It is characterized in that The method comprises: According to the characteristic dimensions associated with the target user, characteristic data of the target user under different characteristic dimensions are obtained, and a first business volume corresponding to the target user under historical data is obtained; Using a pre-trained order volume prediction model, the feature data under the different feature dimensions are processed to generate a second business volume corresponding to the target user under the predicted data; the pre-trained order volume prediction model is obtained by training the initial order volume prediction model; The historical data and the corresponding first traffic volume as well as the predicted data and the corresponding second traffic volume are fitted to determine a data traffic relationship corresponding to the target user.
2. The method according to claim 1, It is characterized in that The obtaining of feature data of the target user in different feature dimensions includes: Acquire first feature data of the target user under a preset feature dimension; Acquire second characteristic data corresponding to a user having a preset relationship with the target user.
3. The method according to claim 1, It is characterized in that The fitting of the historical data and the corresponding first traffic volume and the predicted data and the corresponding second traffic volume to determine the data traffic volume relationship corresponding to the target user includes: Using a preset fitting method, fitting the historical data and the corresponding first business volume and the predicted data and the corresponding second business volume is performed to obtain a target fitting function of the target user; The data traffic relationship corresponding to the target user is determined according to the parameters in the target fitting function.
4. The method according to claim 1, It is characterized in that When the target user is a new user, obtaining feature data of the target user in different feature dimensions includes: Selecting other users with similar feature dimensions; The available features of the other users in the different feature dimensions are used as feature data of the target user.
5. The method according to claim 4, It is characterized in that The obtaining of the first service volume corresponding to the target user under the historical data includes: According to the available traffic volumes corresponding to the other users under the available data, a first traffic volume corresponding to the target user under the historical data is determined.
6. The method according to claim 1, It is characterized in that The method further comprises: Obtaining sample feature data of sample users under different feature dimensions and sample traffic corresponding to the sample data, as well as the relationship between the real data traffic corresponding to the sample users; The sample feature data, sample data and the corresponding sample traffic volume and the real data traffic volume relationship are used as a training sample pair; The initial single volume prediction model is iteratively trained using training sample pairs to obtain the pre-trained single volume prediction model whose sample business volume output under different sample prediction data is consistent with the real data business volume relationship.
7. The method according to claim 6, It is characterized in that The sample data includes sample historical data and sample prediction data, the sample traffic includes a first sample traffic and a second sample traffic, and the relationship between obtaining the sample feature data of the sample user under different feature dimensions and the sample traffic corresponding to the sample data, and the real data traffic corresponding to the sample user includes: Determine sample feature data of the sample user under the different feature dimensions, and determine a first sample business volume corresponding to the sample user under the sample historical data; According to a preset business volume prediction method, obtaining a second sample business volume corresponding to the sample user under the sample prediction data; The sample historical data and the corresponding first sample traffic volume as well as the sample predicted data and the corresponding second sample traffic volume are fitted to obtain a real data traffic relationship corresponding to the sample user.
8. A data relationship determination device, It is characterized in that The device comprises: A first acquisition module, configured to acquire feature data of the target user under different feature dimensions according to feature dimensions associated with the target user, and to acquire a first business volume corresponding to the target user under historical data; A generating module, used to process the feature data under the different feature dimensions by using a pre-trained order volume prediction model, and generate a second business volume corresponding to the target user under the predicted data; the pre-trained order volume prediction model is obtained by training the initial order volume prediction model; A fitting module is used to fit the historical data and the corresponding first business volume and the predicted data and the corresponding second business volume to determine the data business volume relationship corresponding to the target user.
9. A computer-readable storage medium having a computer program stored thereon, It is characterized in that When the computer program is executed by a processor, the method for determining a data relationship according to any one of claims 1 to 7 is implemented.
10. An electronic device, It is characterized in that include: processor; as well as A memory, configured to store executable instructions of the processor; The processor is configured to execute the data relationship determination method according to any one of claims 1 to 7 by executing the executable instructions.
