Business data evaluation method and device
By calculating the similarity between the business data of each historical period and the business data to be evaluated, selecting the target business data and generating the corresponding business evaluation model, the problem of low accuracy of the business evaluation model in the existing technology is solved, and more efficient business data evaluation is achieved.
Patent Information
- Application Number
- CN202311512456.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-11-14
- Publication Date
- 2025-05-16
AI Technical Summary
In the prior art, the business data in the most recent historical period cannot effectively represent the business data to be evaluated, resulting in low accuracy of the business evaluation model, which in turn affects the evaluation effect of the business data.
By calculating the similarity between the business data of each historical period and the business data to be evaluated, the target business data is selected, and the business evaluation model trained in the target business data is evaluated.
It improves the accuracy of the business evaluation model and improves the evaluation effect of business data.
Smart Images
Figure CN120011819A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of big data technology, and in particular to a method and device for evaluating business data. Background Art
[0002] In business projects, set business strategies are often implemented to enhance the value of the project, and the business data of the business project is evaluated to evaluate the execution effect of the business strategy. The current business data evaluation scheme is to use the business data of the recent historical period to train the model, obtain the business evaluation model, and then evaluate the business data through the business evaluation model.
[0003] In the process of implementing the present invention, the inventors found that there are at least the following problems in the prior art:
[0004] The business data in the recent historical period cannot represent the business data to be evaluated. The accuracy of the business evaluation model obtained by using the business data in the recent historical period is low, and thus the evaluation effect of the business data is poor. Summary of the invention
[0005] In view of this, an embodiment of the present invention provides a business data evaluation method and device, which calculates the similarity between the business data of each historical period and the business data to be evaluated, selects the target business data, and uses the business evaluation model trained based on the target business data to evaluate the business data to improve the accuracy of the business evaluation model and enhance the evaluation effect of the business data.
[0006] To achieve the above objective, according to one aspect of an embodiment of the present invention, a method for evaluating business data is provided.
[0007] A method for evaluating business data comprises: in response to a request for evaluating business data, obtaining business data to be evaluated and business data of no less than one historical period; for the business data of each historical period, determining the similarity between the business data of the historical period and the business data to be evaluated, and obtaining the similarity corresponding to the business data of the historical period; according to the similarity corresponding to the business data of each historical period, taking the business data of the historical period with the highest similarity as target business data; performing model training based on the target business data, generating a target business evaluation model, so as to evaluate the business data to be evaluated through the target business evaluation model.
[0008] Optionally, the business data of the historical period includes at least one historical feature data; determining the similarity between the business data of the historical period and the business data to be evaluated includes: for each historical feature data, obtaining the feature data to be evaluated corresponding to the historical feature data from the business data to be evaluated, and calculating the similarity between the historical feature data and the feature data to be evaluated by calculating the information divergence, as the similarity corresponding to the historical feature data; based on the similarity corresponding to each historical feature data, determining the similarity between the business data of the historical period and the business data to be evaluated.
[0009] Optionally, each historical feature data has feature importance; obtaining the similarity of the business data of the historical period based on the similarity corresponding to each historical feature data includes: according to the feature importance, performing weighted averaging on the similarity corresponding to each historical feature data to obtain the similarity of the business data of the historical period.
[0010] Optionally, the feature importance is generated in the following manner: obtaining historical business data samples within a preset time period, the historical business data samples including business indicator samples and at least one feature data sample; using the feature data samples as input, and performing model training with the business indicator samples as training targets to generate a feature importance model; and generating the feature importance corresponding to each feature data based on the model parameters of the feature importance model.
[0011] Optionally, the target historical business data includes business indicators and at least one historical feature data; the model training based on the target business data to generate a business evaluation model includes: taking the historical feature data in the target business data as input, and performing model training with the business indicator as a training target to generate the target business evaluation model.
[0012] Optionally, the target business data includes business indicators and at least one historical feature data; after taking the business data of the historical period with the highest similarity as the target business data, it also includes: obtaining at least one pre-generated business evaluation model; for each business evaluation model, inputting the historical feature data in the target business data into the evaluation model, and obtaining the accuracy of the business evaluation model according to the evaluation results and the business indicators; and taking the business evaluation model with the highest accuracy as the target business evaluation model.
[0013] Optionally, the business data to be evaluated has a time period to be evaluated, and the length of the historical time period is the same as the length of the time period to be evaluated.
[0014] According to another aspect of an embodiment of the present invention, a device for evaluating business data is provided.
[0015] A business data evaluation device comprises: a business data acquisition module, which is used to respond to a business data evaluation request and obtain business data to be evaluated and business data of no less than one historical period; a similarity determination module, which is used to determine the similarity between the business data of each historical period and the business data to be evaluated, and obtain the similarity corresponding to the business data of the historical period; a target business data determination module, which is used to take the business data of the historical period with the highest similarity as target business data according to the similarity corresponding to the business data of each historical period; and a business data evaluation module, which is used to perform model training based on the target business data and generate a target business evaluation model, so as to evaluate the business data to be evaluated through the target business evaluation model.
[0016] Optionally, the business data of the historical period includes no less than one historical feature data; the similarity determination module is also used to: for each of the historical feature data, obtain the feature data to be evaluated corresponding to the historical feature data from the business data to be evaluated, and calculate the similarity between the historical feature data and the feature data to be evaluated by calculating the information divergence, as the similarity corresponding to the historical feature data; based on the similarity corresponding to each historical feature data, determine the similarity between the business data of the historical period and the business data to be evaluated.
[0017] Optionally, each historical feature data has feature importance; the similarity determination module is further used to: perform weighted averaging of the similarities corresponding to each historical feature data according to the feature importance, to obtain the similarity of the business data in the historical period.
[0018] Optionally, it also includes a feature importance generation module, which is used to: obtain historical business data samples within a preset time period, the historical business data samples include business indicator samples and no less than one feature data sample; use the feature data samples as input, and use the business indicator samples as training targets to perform model training to generate a feature importance model; and generate the feature importance corresponding to each feature data according to the model parameters of the feature importance model.
[0019] Optionally, the target historical business data includes business indicators and at least one historical feature data; the business data evaluation module is also used to: use the historical feature data in the target business data as input, use the business indicators as training targets to perform model training, and generate the target business evaluation model.
[0020] Optionally, the target business data includes business indicators and at least one historical feature data; the business data evaluation module is also used to: obtain at least one pre-generated business evaluation model; for each business evaluation model, input the historical feature data in the target business data into the evaluation model, and obtain the accuracy of the business evaluation model based on the evaluation results and the business indicators; and use the business evaluation model with the highest accuracy as the target business evaluation model.
[0021] Optionally, the business data to be evaluated has a time period to be evaluated, and the length of the historical time period is the same as the length of the time period to be evaluated.
[0022] According to yet another aspect of the embodiments of the present invention, an electronic device is provided.
[0023] An electronic device comprises: one or more processors; a memory for storing one or more programs, and when the one or more programs are executed by the one or more processors, the one or more processors implement the business data evaluation method provided by an embodiment of the present invention.
[0024] According to yet another aspect of an embodiment of the present invention, a computer readable medium is provided.
[0025] A computer-readable medium stores a computer program, which, when executed by a processor, implements a method for evaluating business data provided by an embodiment of the present invention.
[0026] An embodiment of the above invention has the following advantages or beneficial effects: by responding to an evaluation request for business data, obtaining business data to be evaluated and business data of no less than one historical period; for the business data of each historical period, determining the similarity between the business data of the historical period and the business data to be evaluated, and obtaining the similarity corresponding to the business data of the historical period; according to the similarity corresponding to the business data of each historical period, taking the business data of the historical period with the highest similarity as the target business data; performing model training based on the target business data to generate a target business evaluation model, and using the target business evaluation model to evaluate the business data to be evaluated. The technical solution is to select the target business data by calculating the similarity between the business data of each historical period and the business data to be evaluated, and use the business evaluation model trained based on the target business data to evaluate the business data to be evaluated, which can improve the accuracy of the business evaluation model and improve the evaluation effect of the business data.
[0027] The further effects of the above-mentioned non-conventional optional manner will be described below in conjunction with the specific implementation manner. BRIEF DESCRIPTION OF THE DRAWINGS
[0028] The accompanying drawings are used to better understand the present invention and do not constitute an improper limitation of the present invention.
[0029] Figure 1 is a schematic diagram of the main steps of a method for evaluating business data according to an embodiment of the present invention;
[0030] Figure 2 is a flow chart of a method for evaluating business data according to an embodiment of the present invention;
[0031] Figure 3 is a schematic diagram of main modules of a device for evaluating business data according to an embodiment of the present invention;
[0032] Figure 4 is an exemplary system architecture diagram to which embodiments of the present invention may be applied;
[0033] Figure 5 It is a schematic diagram of the structure of a computer system of a terminal device or a server suitable for implementing an embodiment of the present invention. DETAILED DESCRIPTION
[0034] The following is a description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for clarity and conciseness, the description of well-known functions and structures is omitted in the following description.
[0035] It should be noted that the collection, collection, updating, analysis, processing, use, transmission, storage and other aspects of user personal information involved in the technical solution of the present invention are in compliance with the provisions of relevant laws and regulations, are used for legitimate purposes, and do not violate public order and good morals. Necessary measures are taken for user personal information to prevent illegal access to user personal information data and maintain the security of user personal information, network security and national security.
[0036] Figure 1 4 is a schematic diagram of main steps of a method for evaluating business data according to an embodiment of the present invention.
[0037] like Figure 1 As shown, a method for evaluating business data according to an embodiment of the present invention mainly includes the following steps S101 to S104.
[0038] Step S101: In response to a request for business data evaluation, business data to be evaluated and business data of no less than one historical period are obtained.
[0039] Specifically, the business data to be evaluated may have a period to be evaluated, and the length of the historical period is the same as the length of the period to be evaluated. For example, if the business data to be evaluated is the business data of the business project from January 1, 2023 to January 30, 2023, and the period length of the business data to be evaluated is 30 days, then a sliding time window with a length of 30 days can be set, and no less than one historical period is obtained through the sliding time window to obtain the business data of the business project in each historical period. The historical period can be January 1, 2022 to January 30, 2022, January 2, 2022 to January 31, 2022, January 3, 2022 to February 1, 2022, ..., December 2, 2022 to December 31, 2022, and so on.
[0040] Among them, the business data for the historical period may include business indicators and no less than one historical characteristic data. For example, the business indicator may be the total revenue of the business project in the corresponding historical period, and the characteristic data may be the average daily revenue, number of active users, praise rate, repurchase rate, etc. in the last 7 days of the historical period.
[0041] Step S102: for the business data of each historical period, determine the similarity between the business data of the historical period and the business data to be evaluated, and obtain the similarity corresponding to the business data of the historical period.
[0042] In one embodiment, determining the similarity between business data of a historical period and business data to be evaluated may include: for each historical feature data, obtaining feature data to be evaluated corresponding to the historical feature data from the business data to be evaluated, and calculating the similarity between the historical feature data and the feature data to be evaluated by calculating information divergence as the similarity corresponding to the historical feature data; based on the similarity corresponding to each historical feature data, determining the similarity between the business data of the historical period and the business data to be evaluated.
[0043] Specifically, according to the historical feature data included in the business data of the historical period, the feature data to be evaluated of the same feature dimension is obtained from the business data to be evaluated. For example, the business data to be evaluated is the business data from January 1, 2023 to January 30, 2023, and the business data of the historical period may be the business data from January 1, 2022 to January 30, 2022. According to the active user volume data from January 1, 2022 to January 30, 2022 (i.e., the historical feature data), the active user volume data from January 1, 2023 to January 30, 2023 (i.e., the feature data to be evaluated) in the active user volume dimension can be obtained.
[0044] The similarity between the historical feature data and the feature data to be evaluated is calculated by calculating the information divergence. The specific calculation method can be: divide the feature values corresponding to the feature data into a preset number of intervals (such as 10), respectively count the distribution frequencies of the two feature data in each interval, approximate the distribution frequencies as their probability distribution, and use the information divergence to calculate the similarity between the two feature data. It should be noted that when the distribution frequency is 0, 0 needs to be replaced with 0.01, otherwise the calculation will be infinite. For example, the characteristic value is divided into three intervals. The active user data from January 1, 2022 to January 30, 2022 includes 100 active user values, of which 30 values are for users ranging from 0 to 1000, 30 values are for users ranging from 1000 to 10000, and 40 values are for users ranging from 10000 to 100000. Then p(x) is 0.3, 0.3, and 0.4 respectively. The active user data from January 1, 2023 to January 30, 2023 includes 100 active user values. Then 20 values are for users ranging from 0 to 1000, 20 values are for users ranging from 1000 to 10000, and 60 values are for users ranging from 10000 to 100000. Then q(x) is 0.2, 0.2, and 0.6 respectively.
[0045] The information divergence may be calculated as KL divergence. KL divergence may represent an asymmetric measure of the difference between two probability distributions. The KL divergence may be calculated as follows:
[0046]
[0047] Among them, p(x) is the characteristic distribution of historical characteristic data, and q(x) is the characteristic distribution of characteristic data to be evaluated.
[0048] In one embodiment, each historical feature data has feature importance; the feature importance can be generated in the following manner: obtaining historical business data samples within a preset time period, the historical business data samples including business indicator samples and no less than one feature data sample; taking the feature data samples as input and the business indicator samples as training targets to perform model training to generate a feature importance model; generating the feature importance corresponding to each feature data according to the model parameters of the feature importance model.
[0049] Specifically, the historical business data sample can be the business data from January 1, 2022 to December 31, 2022, the business indicator sample can be the total revenue from January 1, 2022 to December 31, 2022, and the feature dimension of the feature data sample can be the same as the feature dimension of the data to be evaluated. Based on the xgboost model (an additive model based on the boosting enhancement strategy), the feature data sample is used as input, and the business indicator sample is used as the training target for model training to generate a feature importance model. According to the feature importance model, the model parameters of the feature importance model are obtained. The model parameters can be gain gains. Gain gains identify the relative contribution of the corresponding feature to the model calculated by taking the contribution of each feature to each tree in the model, that is, the importance of the feature in the model.
[0050] In one embodiment, obtaining the similarity of the business data of the historical period based on the similarity corresponding to each historical feature data may include: performing weighted averaging on the similarity corresponding to each historical feature data according to feature importance to obtain the similarity of the business data of the historical period.
[0051] For example, the feature data may be the average daily revenue, number of active users, favorable comment rate, and repurchase ratio in the last seven days of a historical period, and the corresponding feature importances may be 0.2, 0.5, 0.2, and 0.1, respectively. The corresponding similarities of the average daily revenue, number of active users, favorable comment rate, and repurchase ratio in the last seven days of the business data of a certain historical period are 0.82, 0.68, 0.36, and 0.72, respectively. Then, the similarity between the business data of the historical period and the data to be evaluated is: 0.2×0.82+0.5×0.68+0.2×0.36+0.1×0.72=0.648, that is, the similarity of the business data of the historical period is 0.648.
[0052] Step S103: Based on the similarity of the business data of each historical period, the business data of the historical period with the highest similarity is used as the target business data. In this way, based on the similarity between the business data of each historical period and the business data to be evaluated, the business data of a historical period with the information distribution closest to the business data to be evaluated can be determined as the target business data, so that the model training can be performed based on the similarity of data in different periods, thereby improving the accuracy and interpretability of the model prediction.
[0053] Step S104: Model training is performed based on the target business data to generate a target business evaluation model, so as to evaluate the business data to be evaluated through the target business evaluation model.
[0054] In one embodiment, performing model training based on target business data to generate a business evaluation model may include: taking historical feature data in the target business data as input, performing model training with business indicators as training targets, and generating a target business evaluation model.
[0055] In another embodiment, after taking the business data of the historical period with the highest similarity as the target business data, it can also include: obtaining no less than one pre-generated business evaluation model; for each business evaluation model, inputting the historical feature data in the target business data into the evaluation model, and obtaining the accuracy of the business evaluation model according to the evaluation results and business indicators; and taking the business evaluation model with the highest accuracy as the target business evaluation model.
[0056] Specifically, the target business data can be used as test data to test the evaluation accuracy of each business evaluation model, and the business evaluation model with the highest accuracy is selected from each business evaluation model as the target business evaluation model to evaluate the business data to be evaluated. The target business data of the embodiment of the present invention has a high similarity with the business data to be evaluated, thereby ensuring the evaluation accuracy of the target business evaluation model on the business data to be evaluated.
[0057] In one embodiment, the business data to be evaluated is evaluated by the target business evaluation model to obtain the predicted business indicators of the business data to be evaluated. The actual business indicators in the time period of the business data to be evaluated are obtained, and the predicted business indicators are compared with the actual business indicators to obtain the impact of the business strategy executed in the time period of the business data to be evaluated on the business indicators, thereby adjusting the execution strategy of the business project.
[0058] Figure 2 The figure is a flow chart of a method for evaluating business data according to an embodiment of the present invention.
[0059] like Figure 2 As shown, in one embodiment, the business data to be evaluated is obtained, and the business data of multiple historical periods are obtained through a time sliding window. For the business data of each historical period, the similarity between each historical feature data and the corresponding feature data to be evaluated is calculated by calculating the information divergence, and then the similarity corresponding to each historical feature data is weighted averaged according to the feature importance obtained in advance to obtain the similarity of the business data of the historical period, wherein the model training can be performed according to the historical business data samples to obtain the feature importance model, and the feature importance is determined according to the model parameters of the feature importance model. According to the similarity corresponding to the business data of each historical period, the business data of the historical period with the highest similarity is used as the target business data. Model training is performed based on the target business data to generate a target business evaluation model, so as to evaluate the business data to be evaluated through the target business evaluation model.
[0060] The embodiment of the present invention calculates the similarity between the business data of each historical period and the business data to be evaluated, selects the target business data, and uses the business evaluation model obtained based on the target business data for evaluation, thereby improving the accuracy of the business evaluation model and enhancing the evaluation effect of the business data.
[0061] Figure 3 4 is a schematic diagram of main modules of a device for evaluating business data according to an embodiment of the present invention.
[0062] like Figure 3 As shown, a business data evaluation device 300 according to an embodiment of the present invention mainly includes: a business data acquisition module 301 , a similarity determination module 302 , a target business data determination module 303 , and a business data evaluation module 304 .
[0063] The business data acquisition module 301 is used to respond to a business data evaluation request and acquire the business data to be evaluated and business data of no less than one historical period.
[0064] The similarity determination module 302 is used to determine the similarity between the business data in each historical period and the business data to be evaluated, and obtain the similarity corresponding to the business data in the historical period.
[0065] The target business data determination module 303 is used to take the business data of the historical period with the highest similarity as the target business data according to the similarity corresponding to the business data of each historical period.
[0066] The business data evaluation module 304 is used to perform model training based on the target business data and generate a target business evaluation model so as to evaluate the business data to be evaluated through the target business evaluation model.
[0067] In one embodiment, the business data of the historical period may include no less than one historical feature data; the similarity determination module 302 is specifically used to: for each historical feature data, obtain the feature data to be evaluated corresponding to the historical feature data from the business data to be evaluated, and calculate the similarity between the historical feature data and the feature data to be evaluated by calculating the information divergence, as the similarity corresponding to the historical feature data; based on the similarity corresponding to each historical feature data, determine the similarity between the business data of the historical period and the business data to be evaluated.
[0068] In one embodiment, each historical feature data may have feature importance; the similarity determination module 302 is specifically used to: perform weighted averaging of the similarities corresponding to each historical feature data according to the feature importance, to obtain the similarity of the business data in the historical period.
[0069] In one embodiment, it also includes a feature importance generation module (not shown in the figure), which is used to: obtain historical business data samples within a preset time period, the historical business data samples include business indicator samples and no less than one feature data sample; use the feature data samples as input, and use the business indicator samples as training targets to perform model training to generate a feature importance model; generate the feature importance corresponding to each feature data according to the model parameters of the feature importance model.
[0070] In one embodiment, the target historical business data may include business indicators and no less than one historical feature data; the business data evaluation module 304 is specifically used to: use the historical feature data in the target business data as input, use the business indicators as training targets to perform model training, and generate a target business evaluation model.
[0071] In one embodiment, the target business data may include business indicators and no less than one historical feature data; the business data evaluation module 304 is specifically used to: obtain no less than one pre-generated business evaluation model; for each business evaluation model, input the historical feature data in the target business data into the evaluation model, and obtain the accuracy of the business evaluation model based on the evaluation results and business indicators; and use the business evaluation model with the highest accuracy as the target business evaluation model.
[0072] In one embodiment, the to-be-evaluated business data has a to-be-evaluated period, and the length of the historical period is the same as the length of the to-be-evaluated period.
[0073] In addition, the specific implementation content of the business data evaluation device in the embodiment of the present invention has been described in detail in the above business data evaluation method, so the repeated content will not be described again here.
[0074] Figure 4 An exemplary system architecture 400 is shown to which the business data evaluation method or business data evaluation device according to the embodiment of the present invention can be applied.
[0075] like Figure 4 As shown, system architecture 400 may include terminal devices 401, 402, 403, network 404 and server 405. Network 404 is used to provide a medium for communication links between terminal devices 401, 402, 403 and server 405. Network 404 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.
[0076] Users can use terminal devices 401, 402, 403 to interact with server 405 through network 404 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 401, 402, 403, such as business data evaluation applications, model training applications, training data generation applications, instant messaging tools, email clients, social platform software, etc. (only as examples).
[0077] The terminal devices 401 , 402 , and 403 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, and desktop computers.
[0078] Server 405 may be a server that provides various services, such as a background management server that provides support for evaluation websites of business data browsed by users using terminal devices 401, 402, and 403 (for example only). The background management server may respond to the evaluation request for business data and other data received, obtain the business data to be evaluated, and the business data of no less than one historical period; for the business data of each historical period, determine the similarity between the business data of the historical period and the business data to be evaluated, and obtain the similarity corresponding to the business data of the historical period; according to the similarity corresponding to the business data of each historical period, take the business data of the historical period with the highest similarity as the target business data; perform model training based on the target business data, generate a target business evaluation model, and perform evaluation and other processing on the business data to be evaluated through the target business evaluation model, and feed back the processing result (such as the evaluation result of the business data - for example only) to the terminal device.
[0079] It should be noted that the business data evaluation method provided in the embodiment of the present invention is generally executed by the server 405 , and accordingly, the business data evaluation device is generally disposed in the server 405 .
[0080] It should be understood that Figure 4 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to the implementation requirements.
[0081] Reference below Figure 5 , which shows a schematic diagram of the structure of a computer system 500 of a terminal device or a server suitable for implementing an embodiment of the present invention. Figure 5 The terminal device or server shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.
[0082] like Figure 5As shown, the computer system 500 includes a central processing unit (CPU) 501, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 502 or a program loaded from a storage part 508 into a random access memory (RAM) 503. In the RAM 503, various programs and data required for the operation of the system 500 are also stored. The CPU 501, the ROM 502, and the RAM 503 are connected to each other via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.
[0083] The following components are connected to the I / O interface 505: an input section 506 including a keyboard, a mouse, etc.; an output section 507 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 508 including a hard disk, etc.; and a communication section 509 including a network interface card such as a LAN card, a modem, etc. The communication section 509 performs communication processing via a network such as the Internet. A drive 510 is also connected to the I / O interface 505 as needed. A removable medium 511, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 510 as needed, so that a computer program read therefrom is installed into the storage section 508 as needed.
[0084] In particular, according to the embodiments disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 509, and / or installed from the removable medium 511. When the computer program is executed by the central processing unit (CPU) 501, the above-mentioned functions defined in the system of the present invention are executed.
[0085] It should be noted that the computer-readable medium shown in the present invention may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-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 of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer 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. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.
[0086] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.
[0087] The modules involved in the embodiments of the present invention may be implemented in software or hardware. The modules described may also be set in a processor. For example, they may be described as: a processor includes a business data acquisition module, a similarity determination module, a target business data determination module, and a business data evaluation module. The names of these modules do not, in some cases, constitute limitations on the modules themselves. For example, the business data acquisition module may also be described as "a module for responding to a request for business data evaluation, acquiring business data to be evaluated, and business data of no less than one historical period".
[0088] As another aspect, the present invention also provides a computer-readable medium, which may be included in the device described in the above embodiment; or it may exist independently without being assembled into the device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by a device, the device includes: in response to an evaluation request for business data, obtaining business data to be evaluated, and business data of no less than one historical period; for the business data of each historical period, determining the similarity between the business data of the historical period and the business data to be evaluated, and obtaining the similarity corresponding to the business data of the historical period; according to the similarity corresponding to the business data of each historical period, taking the business data of the historical period with the highest similarity as the target business data; performing model training based on the target business data, generating a target business evaluation model, so as to evaluate the business data to be evaluated through the target business evaluation model.
[0089] According to the technical solution of the embodiment of the present invention, in response to the evaluation request of the business data, the business data to be evaluated and the business data of no less than one historical period are obtained; for the business data of each historical period, the similarity between the business data of the historical period and the business data to be evaluated is determined, and the similarity corresponding to the business data of the historical period is obtained; according to the similarity corresponding to the business data of each historical period, the business data of the historical period with the highest similarity is used as the target business data; model training is performed based on the target business data to generate a target business evaluation model, so as to evaluate the business data to be evaluated through the target business evaluation model. By calculating the similarity between the business data of each historical period and the business data to be evaluated, selecting the target business data, and using the business evaluation model trained based on the target business data to evaluate the business data to be evaluated, the accuracy of the business evaluation model can be improved, and the evaluation effect of the business data can be improved.
[0090] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may occur depending on design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.
Claims
1. A method for evaluating business data, characterized in that: include: In response to a request for business data evaluation, obtain the business data to be evaluated and business data for no less than one historical period; For the business data of each historical period, determine the similarity between the business data of the historical period and the business data to be evaluated, and obtain the similarity corresponding to the business data of the historical period; According to the similarity corresponding to the business data of each historical period, the business data of the historical period with the highest similarity is used as the target business data; Model training is performed based on the target business data to generate a target business evaluation model, so as to evaluate the business data to be evaluated through the target business evaluation model.
2. The method according to claim 1, characterized in that The business data of the historical period includes no less than one historical characteristic data; The determining the similarity between the business data in the historical period and the business data to be evaluated includes: For each of the historical feature data, obtain the feature data to be evaluated corresponding to the historical feature data from the business data to be evaluated, and calculate the similarity between the historical feature data and the feature data to be evaluated by calculating the information divergence, so as to serve as the similarity corresponding to the historical feature data; Based on the similarity corresponding to each historical feature data, the similarity between the business data in the historical period and the business data to be evaluated is determined.
3. The method according to claim 2, characterized in that Each historical feature data has feature importance; The obtaining the similarity of the business data of the historical period based on the similarity corresponding to each historical feature data includes: According to the feature importance, a weighted average is performed on the similarity corresponding to each historical feature data to obtain the similarity of the business data in the historical period.
4. The method according to claim 3, characterized in that The feature importance is generated as follows: Acquire historical business data samples within a preset time period, wherein the historical business data samples include a business indicator sample and at least one feature data sample; Taking the feature data sample as input and the business indicator sample as a training target for model training, a feature importance model is generated; The feature importance corresponding to each feature data is generated according to the model parameters of the feature importance model.
5. The method according to claim 1, characterized in that The target historical business data includes business indicators and no less than one historical feature data; The performing model training based on the target business data to generate a business evaluation model includes: The historical feature data in the target business data is used as input, and the business indicators are used as training targets for model training to generate the target business evaluation model.
6. The method according to claim 1, characterized in that The target business data includes business indicators and no less than one piece of historical characteristic data; After taking the business data of the historical period with the highest similarity as the target business data, the method further includes: Obtaining no less than one pre-generated business valuation model; For each of the business evaluation models, historical feature data in the target business data is input into the evaluation model, and the accuracy of the business evaluation model is obtained according to the evaluation result and the business indicator; The business evaluation model with the highest accuracy is used as the target business evaluation model.
7. The method according to claim 1, characterized in that The to-be-evaluated business data has an to-be-evaluated period, and the length of the historical period is the same as the length of the to-be-evaluated period.
8. A device for evaluating business data, characterized in that: include: A business data acquisition module, used to respond to a business data evaluation request and acquire the business data to be evaluated and business data of not less than one historical period; A similarity determination module, for determining, for each historical period of business data, the similarity between the business data of the historical period and the business data to be evaluated, and obtaining the similarity corresponding to the business data of the historical period; A target business data determination module, configured to determine the business data of the historical period with the highest similarity as the target business data according to the similarity corresponding to the business data of each historical period; The business data evaluation module is used to perform model training based on the target business data to generate a target business evaluation model so as to evaluate the business data to be evaluated through the target business evaluation model.
9. An electronic device, characterized in that: include: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 7.
10. A computer readable medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.