Project input degree prediction method and device, computer equipment and storage medium

Abstract

The invention relates to a project input degree prediction method and device, computer equipment and a storage medium. The method comprises the following steps: acquiring historical input data of a project, and generating sample data according to the historical input data, so that the sample data comprises training data and verification data; training a regression model through training data in the sample data to generate a plurality of intermediate regression models; performing cross verification on the corresponding intermediate regression model through verification data in the sample data to obtain a verification result; determining a final regression model from the plurality of intermediate regression models according to the verification result; and predicting the current input degree of the project through the final regression model to obtain a predicted input degree. According to the scheme, the accuracy of project investment degree prediction can be improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a method, device, computer equipment and storage medium for predicting project investment. Background Art

[0002] After undertaking a project, an enterprise needs to predict the project's man-hours, personnel and other input information to facilitate financial planning. However, in the existing project management process, it is not possible to predict the project's future man-hours, personnel and other input information. Summary of the invention

[0003] Based on this, it is necessary to provide a method, device, computer equipment and storage medium for predicting project investment that can improve the accuracy of project investment prediction in response to the above technical problems.

[0004] On the one hand, a method for predicting project investment is provided, the method comprising:

[0005] Acquire historical input data of the project, and generate sample data based on the historical input data, so that the sample data includes training data and verification data;

[0006] Training a regression model using the training data in the sample data to generate a plurality of intermediate regression models;

[0007] Cross-checking the corresponding intermediate regression model using the verification data in the sample data to obtain a verification result;

[0008] Determining a final regression model from the multiple intermediate regression models according to the verification result;

[0009] The current investment degree of the project is predicted by the final regression model to obtain the predicted investment degree.

[0010] In another aspect, a device for predicting project investment is provided, the device comprising:

[0011] A data acquisition module, used to acquire historical input data of a project, and generate sample data based on the historical input data, so that the sample data includes training data and verification data;

[0012] A model generation module, used for training a regression model using training data in the sample data to generate a plurality of intermediate regression models;

[0013] A verification module, used to cross-check the corresponding intermediate regression model through the verification data in the sample data to obtain a verification result;

[0014] A model determination module, used to determine a final regression model from the multiple intermediate regression models according to the verification result;

[0015] The prediction module is used to predict the current investment degree of the project through the final regression model to obtain the predicted investment degree.

[0016] In some embodiments, the historical investment data includes: a plurality of preset periods and the actual investment degree corresponding to each preset period, and the data acquisition module includes:

[0017] A period setting submodule, used for sequentially selecting a preset period from a plurality of preset periods and setting it as a target preset period;

[0018] A data setting submodule, used to set the target preset period and the actual investment degree corresponding to the target preset period as verification data in a set of sample data, and set the remaining preset periods in a plurality of preset periods and the actual investment degrees corresponding to the remaining preset periods as training data in a set of sample data;

[0019] The sample data submodule is used to return to the step of sequentially selecting a preset cycle from a plurality of preset cycles and setting it as a target preset cycle, until each preset cycle is set as the target preset cycle, thereby obtaining a plurality of groups of sample data.

[0020] In some embodiments, the verification module includes:

[0021] An estimation submodule, used to sequentially input a plurality of preset periods into a corresponding intermediate regression model to obtain a plurality of estimated input degrees;

[0022] The verification submodule is used to sequentially obtain the error between each estimated input degree and the corresponding actual input degree to obtain a verification result.

[0023] In some embodiments, the apparatus further comprises:

[0024] The investment acquisition module is used to obtain the current actual investment of the project;

[0025] An error determination module is used to determine whether the error between the current actual investment degree and the predicted investment degree is within a preset error range;

[0026] The sample data generation module is used to generate a new set of sample data according to the current actual investment degree in response to not being within the preset error range.

[0027] In some embodiments, the apparatus further comprises:

[0028] The development stage acquisition module is used to obtain the development stage corresponding to the current actual investment level;

[0029] A range determination module is used to determine a preset error range according to the development stage.

[0030] In some embodiments, the apparatus further comprises:

[0031] The evaluation result acquisition module is used to evaluate the risk factors at different development stages of the project and obtain the evaluation results;

[0032] The error range determination module is used to determine the preset error ranges corresponding to different development stages based on the evaluation results.

[0033] In some embodiments, the apparatus further comprises:

[0034] The historical investment data acquisition module is used to obtain all historical investment data of the project;

[0035] The screening module is used to screen all historical investment data according to the project change information to obtain historical investment data.

[0036] On the other hand, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, and the above steps are implemented when the processor executes the computer program.

[0037] On the other hand, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the above steps are implemented.

[0038] The above-mentioned project investment prediction method, device, computer equipment and storage medium generate sample data through the historical investment data of the project, then generate multiple intermediate regression models based on the sample data and cross-check the multiple intermediate regression models, determine the final regression model based on the verification results, and finally use the final regression model to predict the investment, thereby improving the accuracy of the project investment prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 An application environment diagram of the project investment prediction method provided by an embodiment of the present invention;

[0040] Figure 2 A schematic diagram of a flow chart of a method for predicting project investment provided by an embodiment of the present invention;

[0041] Figure 3 A schematic diagram of a scenario of a method for predicting project investment provided by an embodiment of the present invention;

[0042] Figure 4 A schematic diagram of a model training method for predicting project investment provided by an embodiment of the present invention;

[0043] Figure 5Another schematic diagram of model training for the project investment prediction method provided by an embodiment of the present invention;

[0044] Figure 6 A structural block diagram of a device for predicting project investment provided by an embodiment of the present invention;

[0045] Figure 7 An internal structure diagram of a computer device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0047] The project investment prediction method provided in this application can be applied to Figure 1 In the application environment shown, the terminal 102 communicates with the server 104 via the network.

[0048] The terminal 102 sends the historical investment data of the project to the server 104. After receiving the historical investment data, the server 104 generates sample data based on the historical investment data, so that the sample data includes training data and verification data. Then, the retrospective model is trained based on the training data in the sample data to generate multiple intermediate regression models. Then, the corresponding intermediate regression model is cross-checked with the verification data in the sample data to obtain the verification result. Then, the final regression model is determined from the multiple intermediate regression models based on the verification result. Finally, the current investment degree of the project is predicted by the final regression model to obtain the predicted investment degree.

[0049] The terminal 102 may be, but is not limited to, various personal computers, laptop computers, smart phones, tablet computers, and portable wearable devices, and the server 104 may be implemented as an independent server or a server cluster consisting of multiple servers.

[0050] In one embodiment, Figure 2 As shown in the figure, a method for predicting project investment is provided. Figure 1 The server in the example is used to illustrate the following steps:

[0051] Step S101, obtaining historical investment data of the project, and generating sample data according to the historical investment data, so that the sample data includes training data and verification data.

[0052] The historical investment data of a project refers to the investment data corresponding to the completed part of the project. The investment data may include data such as employees and working hours invested in the project development cycle. In one embodiment, all the historical investment data of the project may be obtained first, and then all the historical investment data may be filtered according to the change information of the project to obtain the historical investment data. The sample data includes training data and verification data. The training data is used to train the preset regression model, and the verification data is used to verify the preset regression model. In one embodiment, the historical investment data may be divided according to certain rules to obtain multiple groups of sample data to expand the data volume for model training. Figure 3 As shown in FIG. 1 , the historical input data is divided into 11 groups of sample data. The training data in each group of sample data can be used to train the regression model, and the verification data in each group of sample data can also be used to verify the regression model.

[0053] Among them, the project change information includes information when the project changes in external conditions such as business requirements, work environment, team size, employee skill level and work efficiency, work content, and work hours. The historical input data that meets the preset change conditions can be eliminated to obtain the filtered historical input data.

[0054] In the actual project development process, the project investment required for the same project at different development stages is often different. For example, the early development stage and mid-term development stage of the project require a large number of employees and a lot of man-hours to develop the project. The late development stage of the project often enters the final stage where the project team conducts acceptance and successful delivery of the project. At this time, the number of employees and man-hours required are greatly reduced. Therefore, for projects with long development cycles and a large amount of historical investment data, all historical investment data can be screened in combination with the current development stage of the project, and the historical investment data of the current development stage of the project can be obtained. The historical investment data of other stages are eliminated, which not only reduces the amount of data calculation in the process of predicting the investment degree of the project in this application, but also improves the accuracy of the prediction.

[0055] In one embodiment, the historical investment data includes: a plurality of preset periods and an actual investment degree corresponding to each preset period, and the steps of obtaining the historical investment data of the project and generating sample data according to the historical investment data so that the sample data includes training data and verification data include:

[0056] (A1) A preset period is selected in sequence from a plurality of preset periods and set as a target preset period.

[0057] (A2) setting a target preset period and an actual level of involvement corresponding to the target preset period as verification data in a set of sample data, and setting remaining preset periods in the plurality of preset periods and the actual level of involvement corresponding to the remaining preset periods as training data in a set of sample data.

[0058] (A3) Return to the step of sequentially selecting a preset cycle from a plurality of preset cycles and setting it as a target preset cycle, until each preset cycle is set as a target preset cycle, thereby obtaining sample data.

[0059] like Figure 3 As shown, the preset period in the historical investment data is January to November, and the corresponding actual investment degrees from January to November are 1.1, 1.14, 1.05, 2.8375, 3.46, 9.6, 14.8875, 9.49, 9.425, 8.355 and 11.05 respectively. First, November is set as the target preset period, and the target preset period November and the corresponding actual investment degree 11.05 constitute the verification data in the first group of sample data, and the remaining preset periods 1-10 and the corresponding actual investment degrees are used as training data in the first group of sample data. Then, October is set as the target preset period, and the target preset period October and the corresponding actual investment degree 8.355 are used as verification data in the second group of sample data, and the remaining preset periods 1-9 and 11 and the corresponding actual investment degrees are used as training data in the second group of sample data. This is repeated until the last preset period of 1 month is set as the target preset period, and the actual investment degree 1.1 corresponding to the target preset period of 1 month is used as the training data in the 11th group of sample data. The remaining preset periods 2-11 and the corresponding actual investment degrees are used as the training data for the 11th group of sample data. In summary, we can get the following: Figure 3 The 11 sets of sample data are shown.

[0060] Step S102: training a regression model using training data in the sample data to generate a plurality of intermediate regression models.

[0061] The regression model can be a multinomial regression equation. The regression model is a multinomial regression equation Y=β0+β1*X ^1 +β2*X ^2 +β3*X ^3 +...+βn*X ^n Take β0 as an example for description. Among them, Y is the estimated investment degree, X is the preset period, β0, β1, β2, β3, ... βn are the coefficients of the model, and n is the highest order of the polynomial.

[0062] like Figure 4As shown, by using the training data corresponding to the working hours from January to October in the first group of sample data, curve fitting is performed to determine the coefficients of the regression model, and five polynomial regression equations of different orders can be obtained as intermediate regression models. When the order is 6, the following equation 1.1 is obtained: y = 0.0013x 6 -0.031x 5 +0.206x 4 -0.1008x 3 -2.6577x 2 +6.8972x-3.295. When the order is 5, we get the following equation 1.2: y = 0.0135x 5 -0.3647x 4 +3.4798x 3 -13.988x 2 +23.501x-11.71. When the order is 4, we get the following equation 1.3: y = 0.0068x 4 -0.2577x 3 +2.7296x 2 -8.3438x+7.6075. When the order is 3, the following equation 1.4 is obtained: y = -0.1076x 3 +1.6311x 2 -5.3416x+5.2657. When the order is 2, we get the following equation 1.5: y = -0.1443x 2 +2.8466x-3.9661.

[0063] like Figure 5 As shown, by determining the coefficients of the regression model and performing curve fitting on the training data of working hours from January to September and November in the second group of sample data, 5 polynomial regression equations of different orders can be obtained. When the order is 6, the following equation 2.1 is obtained: y = -3E-05x 6 +0.0117x 5 -0.3087x 4 +2.9377x 3 -11.732x 2 +19.529x-9.4532. When the order is 5, we get the following equation 2.2:

[0064] y=0.0107x 5 -0.2945x 4 +2.8448x 3 -11.428x 2 +19.072x-9.2175. When the order is 4, we get the following equation 2.3: y = 0.018x 4 -0.4801x 3+4.2006x 2 -11.974x+10.199. When the order is 3, we get the following equation 2.4: y = -0.0511x 3 +0.8201x 2 -2.1191x+2.1208. When the order is 2, we get the following equation 2.5: y = -0.1031x 2 +2.5133x-3.4858.

[0065] This process is repeated until the training data in the 11th set of sample data also generates 5 polynomial regression equations of different orders. Since the training data of each set of sample data can generate 5 polynomial regression equations of different orders, a total of 11*5=55 polynomial regression equations can be obtained for the 11 sets of sample data, that is, 55 intermediate regression models can be obtained.

[0066] Step S103, cross-checking the corresponding intermediate regression model using the verification data in the sample data to obtain a verification result.

[0067] In one embodiment, a plurality of preset cycles may be sequentially input into a corresponding intermediate regression model to obtain a plurality of estimated input degrees, and then the error degree between each estimated input degree and the corresponding actual input degree may be sequentially obtained to obtain a verification result.

[0068] Each preset period corresponds to a group of estimated input levels, and each group of estimated input levels includes a plurality of estimated input levels. Figure 3 As shown, the preset period includes January to November. First, x=11 is input into the polynomial regression equations 1.1-1.6, and the first set of estimated input degrees Y are obtained in sequence: -56.678, 20.4826, 2.6674, 0.6556 and 9.8862. Then x=10 is input into the polynomial regression equations 2.1-2.6, and the second set of estimated input degrees Y are obtained in sequence: 3.3368, 8.5025, 10.419, 11.8398 and 11.3372. And so on, until x=1 is input into the polynomial regression equations 11.1-11.6, and all the estimated input degrees Y are obtained.

[0069] The error between the two can be determined by taking the absolute value of the difference between each estimated investment and the corresponding actual investment, or by squaring the difference between each estimated investment and the corresponding actual investment, etc., and the specific method for calculating the error is not specifically limited here. Since the actual investment corresponding to the first group of predicted investment Y is 11.05, the error between the first group of estimated investment and the corresponding actual investment can be obtained by using the absolute value of the difference method: 67.728, 9.4326, 8.3826, 10.3944 and 1.1638. The actual investment corresponding to the second group of predicted investment Y is 8.355, and the error between the first group of estimated investment and the corresponding actual investment can be obtained by using the absolute value of the difference method: 5.0182, 0.1475, 2.064, 3.4848 and 2.9822. By analogy, 55 error degrees of 11 groups of estimated investment and the corresponding actual investment can be obtained to obtain the verification results.

[0070] Step S104: determining a final regression model from a plurality of intermediate regression models according to the verification result.

[0071] The smaller the error, the higher the accuracy of the investment estimation using the intermediate regression model. Therefore, the error with the smallest value is selected from multiple error degrees, and then the intermediate regression model corresponding to the smallest error is set as the final regression model for subsequent investment prediction. For example, among the 55 error degrees mentioned above, the multinomial regression equation y=-0.1237x generated by the training data of the 10th group of sample data 2 The error between the estimated input degree and the actual input degree predicted by +2.6509x-3.6308 is the smallest, so the polynomial regression equation can be set as the final regression model later.

[0072] Step S105, predicting the current investment degree of the project through the final regression model to obtain the predicted investment degree.

[0073] You can get the current cycle of the project and input the current cycle into the final regression model to get the predicted investment. For example, when predicting the investment in December, substitute x=12 into the final regression model y=-0.1237x 2 +2.6509x-3.6308, we can get the predicted investment level for December as 10.3672.

[0074] In some embodiments, after the step of predicting the current investment degree of the project by using the final regression model to obtain the predicted investment degree, the method further includes:

[0075] (B1) Obtain the current actual investment level of the project.

[0076] (B2) Determine whether the error between the current actual investment degree and the predicted investment degree is within a preset error range.

[0077] (B3) In response to the error not being within the preset error range, a new set of sample data is generated according to the current actual investment level.

[0078] Compare the current actual investment and the predicted investment. When the error between the two is within the preset error range, it means that the prediction accuracy of the final regression model is high, and the final regression model can continue to be used for subsequent investment prediction. When the error between the two is not within the preset error range, it means that the prediction accuracy of the final scale needs to be further improved. Therefore, a new set of sample data can be generated for the current actual investment and the current cycle, and the final regression model can be regenerated in a cross-checking manner.

[0079] The same project encounters different risk factors at different development stages. The more risk factors a project encounters at the development stage, the worse the stability of the project at this development stage, and the larger the error of the predicted investment relative to the actual investment. The fewer risk factors a project encounters at the development stage, the higher the stability of the project at this development stage, and the smaller the error of the predicted investment relative to the actual investment. Among them, risk factors include demand changes, project function changes, and sudden force majeure. Therefore, the risk factors of different development stages of the project can be evaluated in advance to obtain evaluation results. Then, according to the evaluation results, the preset error ranges corresponding to different development stages are determined. When the evaluation result shows that the risk of the development stage is high, the preset error range corresponding to the development stage can be set to a larger range. When the evaluation result shows that the risk of the development stage is low, the preset error range corresponding to the development stage can be set to a smaller range. When the evaluation result shows that the risk of the development stage is medium, the preset error range corresponding to the development stage can be set to a medium range.

[0080] In one embodiment, the development stage of the project can be divided into the early stage, the middle stage and the late stage. Due to the early stage of project development, there are many internal and external uncertain factors, such as changes in the initial Party A's requirements, and better options for functions proposed by relevant departments of the company, which will cause a large deviation in the prediction of the investment degree in the early stage of the project. Therefore, an assessment result with a higher risk in the early stage can be obtained, so the preset error range is set in a larger range, such as between 0% and 15%. All investment prediction data with an error degree within this preset error range are acceptable. Entering the mid-term stage of project development, the requirements of all related parties have been basically determined, the project has developed steadily, and basically no major changes in requirements will occur. However, there will be occasional sudden force majeure factors, such as the introduction of national standards, which require the cooperation of various departments of the company to allocate functions and requirements. This special situation will also cause certain deviations and influences on the prediction of project investment. At this time, an assessment result with medium risk in the mid-term stage can be obtained, so the preset error range is set in a medium range, such as between 0% and 10%. When the project is in the late stage of the development cycle, the project enters the late closing stage, and the project team conducts acceptance and delivers project results. At this stage, the project will proceed normally to completion, and the impact on the forecast of project investment should be minimal. At this time, the assessment results with lower risks in the later stages can be obtained, so the preset error range is set to a smaller range, such as between 0%-5%.

[0081] After presetting the association between different development stages and different preset error ranges, in the actual investment prediction process, the development stage corresponding to the current actual investment can be obtained, and then the preset error range can be determined according to the development stage. For example, when the development stage corresponding to the current actual investment is the late stage, the preset error range obtained is between 0% and 5%.

[0082] The above-mentioned project investment prediction method generates multiple groups of sample data through the historical investment data of the project, and then generates multiple intermediate regression models based on the multiple groups of sample data and cross-checks the multiple intermediate regression models, determines the final regression model according to the verification results, and finally uses the final regression model to predict the investment, thereby improving the accuracy of the project investment prediction.

[0083] It should be understood that although Figure 2 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 2At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0084] In one embodiment, Figure 6 As shown, a device 20 for predicting project investment is provided, comprising: a data acquisition module 201, a model generation module 202, a verification module 203, a model determination module 204 and a prediction module 205, wherein: the data acquisition module 201 is used to obtain the historical investment data of the project, and generate sample data based on the historical investment data, so that the sample data includes training data and verification data. The model generation module 202 is used to train the regression model through the training data in the sample data to generate multiple intermediate regression models. The verification module 203 is used to cross-check the corresponding intermediate regression model through the verification data in the sample data to obtain the verification result. The model determination module 204 is used to determine the final regression model from the multiple intermediate regression models according to the verification result. The prediction module 205 is used to predict the current investment of the project through the final regression model to obtain the predicted investment.

[0085] In some embodiments, the historical investment data includes: multiple preset periods and the actual investment degree corresponding to each preset period, and the data acquisition module 201 includes: a period setting submodule, a data setting submodule and a sample data submodule. Among them, the period setting submodule is used to select a preset period from multiple preset periods in turn and set it as a target preset period. The data setting submodule is used to set the target preset period and the actual investment degree corresponding to the target preset period as verification data in a group of sample data, and set the remaining preset periods in the multiple preset periods and the actual investment degree corresponding to the remaining preset periods as training data in a group of sample data. The sample data submodule is used to return to the step of selecting a preset period from multiple preset periods in turn and setting it as a target preset period, until each preset period is set as a target preset period, to obtain multiple groups of sample data.

[0086] In some embodiments, the verification module includes: an estimation submodule and a verification submodule. The estimation submodule is used to sequentially input a plurality of preset cycles into a corresponding intermediate regression model to obtain a plurality of estimated input degrees. The verification submodule is used to sequentially obtain the error between each estimated input degree and the corresponding actual input degree to obtain a verification result.

[0087] In some embodiments, the device 20 further includes: an engagement acquisition module, an error determination module, and a sample data generation module. The engagement acquisition module is used to acquire the current actual engagement of the project. The error determination module is used to determine whether the error between the current actual engagement and the predicted engagement is within a preset error range. The sample data generation module is used to generate a new set of sample data according to the current actual engagement in response to not being within the preset error range.

[0088] In some embodiments, the device 20 further includes: a development stage acquisition module and a range determination module. The development stage acquisition module is used to acquire the development stage corresponding to the current actual investment level. The range determination module is used to determine a preset error range according to the development stage.

[0089] In some embodiments, the device 20 further includes: an evaluation result acquisition module and an error range determination module. The evaluation result acquisition module is used to evaluate the risk factors of different development stages of the project to obtain evaluation results. The error range determination module is used to determine the preset error ranges corresponding to different development stages according to the evaluation results.

[0090] In some embodiments, the device 20 further includes: a historical investment data acquisition module and a screening module. The historical investment data acquisition module is used to acquire all historical investment data of the project. The screening module is used to screen all historical investment data according to the change information of the project to obtain historical investment data.

[0091] The above-mentioned project investment prediction device generates sample data through the historical investment data of the project, and then generates multiple intermediate regression models based on the sample data and cross-checks the multiple intermediate regression models, determines the final regression model based on the verification results, and finally uses the final regression model to predict the investment, thereby improving the accuracy of the project investment prediction.

[0092] For the specific definition of the device for predicting project investment, please refer to the definition of the method for predicting project investment mentioned above, which will not be repeated here. Each module in the above-mentioned device for predicting project investment can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0093] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 7As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for predicting project investment is implemented.

[0094] Those skilled in the art will understand that Figure 7 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0095] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program:

[0096] Acquire historical investment data of the project, and generate sample data based on the historical investment data, so that the sample data includes training data and verification data; train a regression model through the training data in the sample data to generate multiple intermediate regression models; cross-check the corresponding intermediate regression models through the verification data in the sample data to obtain verification results; determine a final regression model from the multiple intermediate regression models based on the verification results; predict the current investment degree of the project through the final regression model to obtain a predicted investment degree.

[0097] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0098] Select one preset period from the multiple preset periods in sequence and set it as the target preset period. Set the target preset period and the actual investment degree corresponding to the target preset period as verification data in a set of sample data, and set the remaining preset periods in the multiple preset periods and the actual investment degrees corresponding to the remaining preset periods as training data in a set of sample data. Return to the step of selecting one preset period from the multiple preset periods in sequence and setting it as the target preset period, until each preset period is set as the target preset period, to obtain multiple sets of sample data.

[0099] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0100] Multiple preset periods are sequentially input into the corresponding intermediate regression model to obtain multiple estimated input degrees. The error between each estimated input degree and the corresponding actual input degree is sequentially obtained to obtain a verification result.

[0101] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0102] The current actual investment of the project is obtained. It is determined whether the error between the current actual investment and the predicted investment is within a preset error range. In response to not being within the preset error range, a new set of sample data is generated according to the current actual investment.

[0103] In one embodiment, when the processor executes the computer program, the following steps are further implemented: obtaining the development stage corresponding to the current actual investment level, and determining a preset error range according to the development stage.

[0104] In one embodiment, when the processor executes the computer program, the following steps are further implemented: risk factors of different development stages of the project are evaluated to obtain evaluation results, and preset error ranges corresponding to different development stages are determined according to the evaluation results.

[0105] In one embodiment, when the processor executes the computer program, the following steps are further implemented: obtaining all historical investment data of the project. According to the change information of the project, all historical investment data are screened to obtain historical investment data.

[0106] The processor generates sample data through the historical investment data of the project, generates multiple intermediate regression models based on the sample data and cross-checks the multiple intermediate regression models, determines the final regression model based on the verification results, and finally uses the final regression model to predict the investment degree, thereby improving the accuracy of the project investment degree prediction.

[0107] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0108] Acquire historical investment data of the project, and generate sample data based on the historical investment data, so that the sample data includes training data and verification data; train a regression model through the training data in the sample data to generate multiple intermediate regression models; cross-check the corresponding intermediate regression models through the verification data in the sample data to obtain verification results; determine a final regression model from the multiple intermediate regression models based on the verification results; predict the current investment degree of the project through the final regression model to obtain a predicted investment degree.

[0109] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0110] Select one preset period from the multiple preset periods in sequence and set it as the target preset period. Set the target preset period and the actual investment degree corresponding to the target preset period as verification data in a set of sample data, and set the remaining preset periods in the multiple preset periods and the actual investment degrees corresponding to the remaining preset periods as training data in a set of sample data. Return to the step of selecting one preset period from the multiple preset periods in sequence and setting it as the target preset period, until each preset period is set as the target preset period, to obtain multiple sets of sample data.

[0111] In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: inputting multiple preset cycles into the corresponding intermediate regression model in sequence to obtain multiple estimated investment degrees, and obtaining the error between each estimated investment degree and the corresponding actual investment degree in sequence to obtain a verification result.

[0112] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0113] The current actual investment of the project is obtained. It is determined whether the error between the current actual investment and the predicted investment is within a preset error range. In response to not being within the preset error range, a new set of sample data is generated according to the current actual investment.

[0114] In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: obtaining the development stage corresponding to the current actual investment level, and determining a preset error range according to the development stage.

[0115] In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: risk factors of different development stages of the project are evaluated to obtain evaluation results, and preset error ranges corresponding to different development stages are determined according to the evaluation results.

[0116] In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: obtaining all historical investment data of the project, and filtering all historical investment data according to the change information of the project to obtain the historical investment data.

[0117] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0118] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0119] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A method for predicting project investment, characterized in that: include: Acquire historical input data of the project, and generate sample data based on the historical input data, so that the sample data includes training data and verification data; Training a regression model using the training data in the sample data to generate a plurality of intermediate regression models; Cross-checking the corresponding intermediate regression model using the verification data in the sample data to obtain a verification result; Determining a final regression model from the multiple intermediate regression models according to the verification result; The current investment degree of the project is predicted by the final regression model to obtain the predicted investment degree.

2. The method for predicting project investment according to claim 1, characterized in that: The historical investment data includes: a plurality of preset periods and the actual investment degree corresponding to each preset period. The steps of obtaining the historical investment data of the project and generating sample data according to the historical investment data so that the sample data includes training data and verification data include: Selecting one preset period from the plurality of preset periods in sequence and setting it as a target preset period; The target preset period and the actual investment degree corresponding to the target preset period are set as verification data in a set of sample data, and the remaining preset periods in the plurality of preset periods and the actual investment degrees corresponding to the remaining preset periods are set as training data in a set of sample data; Return to the step of sequentially selecting a preset period from the multiple preset periods and setting it as the target preset period, until each preset period is set as the target preset period, to obtain multiple groups of sample data.

3. The method for predicting project investment according to claim 2, characterized in that: The step of cross-checking the corresponding intermediate regression model using the verification data in the sample data to obtain the verification result includes: Inputting the plurality of preset periods into the corresponding intermediate regression model in sequence to obtain a plurality of estimated investment degrees; The error between each estimated investment degree and the corresponding actual investment degree is obtained in turn to obtain the verification result.

4. The method for predicting project investment according to claim 1, characterized in that: After the step of predicting the current investment degree of the project by using the final regression model to obtain the predicted investment degree, the method further includes: Get the current actual investment of the project; Determine whether the error between the current actual investment degree and the predicted investment degree is within a preset error range; In response to not being within the preset error range, new sample data is generated according to the current actual engagement level.

5. The method for predicting project investment according to claim 1, characterized in that: Before the step of determining whether the error between the current actual investment degree and the predicted investment degree is within a preset error range, the step further includes: Obtain the development stage corresponding to the current actual investment level; The preset error range is determined according to the development stage.

6. The method for predicting project investment according to claim 5, characterized in that: Before the step of determining the preset error range according to the development stage, the method further includes: Evaluate the risk factors of the project at different development stages and obtain evaluation results; According to the evaluation results, preset error ranges corresponding to different development stages are determined.

7. The method for predicting project investment according to claim 1, characterized in that: Before the step of obtaining the historical input data of the project and generating sample data according to the historical input data so that the sample data includes training data and verification data, the step further includes: Obtain all historical investment data for the project; According to the change information of the project, all the historical investment data are screened to obtain the historical investment data.

8. A device for predicting project investment, characterized in that: The device comprises: A data acquisition module, used to acquire historical input data of a project, and generate sample data based on the historical input data, so that the sample data includes training data and verification data; A model generation module, used for training a regression model using training data in the sample data to generate a plurality of intermediate regression models; A verification module, used to cross-check the corresponding intermediate regression model through the verification data in the sample data to obtain a verification result; A model determination module, used to determine a final regression model from the multiple intermediate regression models according to the verification result; The prediction module is used to predict the current investment degree of the project through the final regression model to obtain the predicted investment degree.

9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images