JP7183319B2 - Information processing device, prediction method and prediction program - Google Patents

Description

The present invention relates to an information processing apparatus and the like.

After the occurrence of a business negotiation (project) is recognized, a contract is made through various internal procedures and deliberation processes. During this period from the recognition of the deal to the conclusion of the contract, an estimate of the profitability of the project is made.

When estimating the profitability of a project, the project manager (PM), who is responsible for promoting the project when the business negotiation is accepted, confirms materials such as RFP and interviews with the customer, I make my own judgments.
In various in-house procedures and deliberation processes, an experienced expert who has served as a project manager many times in the past serves as an auditor, and from a third-party perspective, audits the estimate that the project manager independently created.

JP 2004-287849 A JP-A-2006-18532

Due to the recent economic situation, not only large-scale projects but also small- and medium-sized projects are increasing, and it is required to appropriately predict the profitability of each project.

However, due to the limited number of experts, it may be difficult for experts to do all the work related to project estimation. In addition, there is a possibility that an unprofitable project may be accepted by mistake due to a failed estimate.

In one aspect, an object of the present invention is to provide an information processing device, a prediction method, and a prediction program capable of detecting an unprofitable project.

In the first plan, the information processing device has a learning section and a prediction section. The learning part is a characteristic value related to the risk of the project implemented in the past, which includes the scale of the hardware provided in the project and the content of the service provided in the project, and the relationship between the actual value of the cost rate related to the project. Generate a model by performing machine learning based on The prediction unit inputs the feature amount of a certain project into the model and predicts the cost rate of the certain project.

Anyone can detect unprofitable projects.

FIG. 1 is a diagram showing a system according to this embodiment. FIG. 2 is a diagram showing an example of basic information of a project. FIG. 3 is a diagram illustrating an example of feature amount information of a project. FIG. 4 is a diagram showing an example of screen information of a prediction result. FIG. 5 is a functional block diagram showing the configuration of the information processing apparatus according to this embodiment. FIG. 6 is a diagram showing an example of the data structure of the history database. FIG. 7 is a diagram showing an example of a bubble chart. FIG. 8 is a flow chart showing the processing procedure of the information processing apparatus according to this embodiment. FIG. 9 is a diagram illustrating an example of a hardware configuration of a computer that implements functions similar to those of the information processing apparatus of the embodiment.

Hereinafter, embodiments of an information processing apparatus, a prediction method, and a prediction program disclosed in the present application will be described in detail based on the drawings. In addition, this invention is not limited by this Example.

FIG. 1 is a diagram showing a system according to this embodiment. As shown in FIG. 1, this system 1 has an information processing device 100 and terminal devices 10a, 10b, and 10c. The information processing device 100 and the terminal devices 10a to 10c are interconnected via a network 20. FIG. Although terminal devices 10a to 10c are shown in the example shown in FIG. 1, this system 1 may have other terminal devices. In the following description, the terminal devices 10a to 10c are collectively referred to as the terminal device 10. FIG.

The terminal device 10 is a terminal device used by a user (all employees) and corresponds to a PC (Personal Computer), a notebook PC, a tablet terminal, a smart phone, or the like. For example, when the user operates the terminal device 10 to input information about a project (hereinafter referred to as project information), the terminal device 10 transmits a project information prediction request to the information processing device 100 . Project information is attached to the forecast request.

The project information includes basic information about the project and information about the feature amount of the project (hereinafter referred to as feature amount information). FIG. 2 is a diagram showing an example of basic information of a project. As shown in FIG. 2, the basic information includes project name, customer name, identification number, planned cost rate, person in charge, sales amount, and the like. The project name is the name of the project. The customer name is the name of the customer. The identification number is a number that identifies the project.

The planned cost rate is the cost rate of the project predicted by the person in charge. The person in charge is the person in charge of the project. The sales amount is the sales amount of the project. The basic information may further include information other than the project name, customer name, project number, expected cost rate, person in charge, and expected sales amount shown in FIG.

The feature amount information of a project is information related to the risks of projects implemented in the past, and includes the scale of hardware used in the project and the content of services provided in the project.

FIG. 3 is a diagram illustrating an example of feature amount information of a project. In the example shown in FIG. 3, the feature amount information associates feature amount items with feature amount input values. The items include facility construction, server/storage construction, NW (New Work) construction, DB (Data Base) construction, cloud construction, integration with applications (application software), migration work, number of server devices, and number of terminal devices. , number of bases, number of departments, technical skills, product selection method, etc.

Facility construction indicates various works such as electrical work, facility construction, DC (Data Center) construction, equipment relocation, etc. When performing work related to facility construction, the input value is "1" and the work is not performed. In this case, the input value is "0".

Server/storage construction indicates the design, setting, operation verification, etc. of a server or storage. When performing work related to server/storage construction, the input value is "1", and when no work is performed, the input value becomes "0".

NW construction indicates network design, setting, operation verification, etc. When work related to network construction is performed, the input value is "1", and when work is not performed, the input value is "0". .

DB construction indicates the design, setting, operation verification, etc. of a database including PaaS (Platform as a Service). , the input value becomes "0".

Cloud construction indicates cloud design, settings, operation verification, etc. including IaaS (Infrastructure as a Service). When working on cloud construction, the input value is "1", and when not working , the input value becomes "0".

Consistency with the application indicates operation verification, etc. including the application, and the input value is "1" when work related to consistency with the application is performed, and the input value is "0" when no work is performed. becomes.

The migration work indicates data migration or application migration, and the input value is "1" when the migration work is performed, and the input value is "0" when the migration work is not performed.

The number of server devices indicates the number of server devices used in the project. The number of server devices is set as the input value.

The number of terminal devices indicates the number of terminal devices used in the project. The number of terminal devices is set as the input value.

The number of bases indicates the number of bases used by the worker when executing the project. The number of bases (the number of bases) is set as the input value.

The number of departments indicates the number of departments to which workers belong when executing a project. The number of departments (the number of departments) is set as the input value.

Technical skill indicates the technical skill of the workers who carry out the project. The input value includes selection item candidates for a plurality of technical skills, and one of the selection items is selected.

The product selection method indicates the product selection method for the project. The input value includes selection item candidates for a plurality of product selection methods, and one of the selection items is selected.

The items of the feature amount information explained in FIG. , the number of departments, technical skills, and product selection methods have been described, but other items may also be included.

Returning to the description of FIG. The information processing apparatus 100 generates a prediction model by executing machine learning using the history of past project information. In this embodiment, the prediction model is described as a decision tree, but it is not limited to this, and may be an NN (Neural Network) or the like. For example, by inputting feature information into a prediction model, a predicted value of the cost rate is calculated.

When the information processing apparatus 100 receives a prediction request for project information from the terminal device 10, the information processing apparatus 100 predicts the cost rate of the project by inputting the feature amount information included in the project information into the prediction model. The cost rate predicted using the prediction model is referred to as "predicted cost rate." The information processing apparatus 100 generates screen information of the prediction result based on the prediction result, and causes the terminal device 10 to display the screen information.

FIG. 4 is a diagram showing an example of screen information of a prediction result. As shown in FIG. 4, this screen information 30 includes display areas 30a and 30b. Items related to basic information of the project information are displayed in the display area 30a. For example, the display area 30a includes project name, customer name, sales department, identification number, sales amount, and the like.

The display area 30b includes a person in charge, a planned cost rate, and a cost rate prediction category. The person in charge corresponds to the person in charge included in the basic information of the project information. The planned cost rate corresponds to the planned cost rate included in the basic information of the project information. The cost rate prediction category is information indicating the degree of divergence between the planned cost rate and the predicted cost rate.

For example, when the value obtained by subtracting the planned cost rate from the predicted cost rate is equal to or greater than the first threshold, the product is classified as "Classification A." If the difference value obtained by subtracting the planned cost rate from the predicted cost rate is less than the first threshold and equal to or greater than the second threshold (first threshold > second threshold), the product is classified as "Classification B." If the difference value obtained by subtracting the planned cost rate from the predicted cost rate is less than the second threshold and equal to or greater than the third threshold (second threshold > third threshold), the product is classified as "Class C."

If the difference value obtained by subtracting the planned cost rate from the predicted cost rate is less than the third threshold and equal to or greater than the fourth threshold (third threshold>fourth threshold), the product is classified as “Class D”. If the difference value obtained by subtracting the planned cost rate from the predicted cost rate is less than the fourth threshold and equal to or greater than the fifth threshold (fourth threshold>fifth threshold), it is classified as "Classification E." If the difference value obtained by subtracting the planned cost rate from the predicted cost rate is less than the fifth threshold, it is classified as "Classification F".

For example, the first and second thresholds are positive thresholds, and the third, fourth, and fifth thresholds are negative thresholds. Then, categories A and B indicate that the projected cost rate is higher than the expected cost rate assumed by the user, and that the project tends to be unprofitable. On the other hand, categories C to F indicate that the deviation between the planned cost rate assumed by the user and the predicted cost rate is small, or the predicted cost rate is lower than the planned cost rate, indicating that the project is profitable.

Although illustration is omitted, in FIG. 4, a display area for displaying information in which the item (explanatory variable) of the feature information that influenced the prediction result of the prediction model (decision tree) is associated with the degree of influence is displayed. It may further contain: In this display area, items having a large degree of influence are displayed in order.

As described above, the information processing apparatus 100 according to this embodiment generates a prediction model by machine learning based on past history. The information processing apparatus 100 inputs the feature amount information received from the terminal device 10 to the prediction model, calculates the predicted value of the cost rate (predicted cost rate), and displays screen information 30 of the prediction result on the terminal device 10. Let The user can detect an unprofitable project by referring to the cost rate prediction category that indicates the degree of divergence between the planned cost rate predicted by the user and the predicted cost rate shown in the screen information 30 .

Next, an example of the configuration of the information processing apparatus 100 described with reference to FIG. 1 will be described. FIG. 5 is a functional block diagram showing the configuration of the information processing apparatus according to this embodiment. As shown in FIG. 5, the information processing apparatus 100 has a communication section 110, an input section 120, a display section 130, a storage section 140, and a control section 150. FIG.

The communication unit 110 transmits and receives information to and from the terminal device 10 or an external device (not shown) via the network 20 . For example, communication unit 110 is realized by a NIC (Network Interface Card) or the like.

The input unit 120 is an input device for inputting various information. The input unit 120 corresponds to a keyboard, mouse, touch panel, or the like.

The display unit 130 is a display device that displays information output from the control unit 150 . The display unit 130 corresponds to a liquid crystal display, an organic EL (Electro Luminescence) display, a touch panel, or the like.

The storage unit 140 has a history database 141 and a prediction model 142 . The storage unit 140 is realized by, for example, a semiconductor memory device such as a RAM (Random Access Memory) or flash memory, or a storage device such as a hard disk or an optical disk.

The history database 141 is a database that holds past project information. FIG. 6 is a diagram showing an example of the data structure of the history database. As shown in FIG. 6, this history database 141 holds project information and cost rates (actual values) in association with item numbers. The project information includes the basic information described with reference to FIG. 2 and the feature amount information described with reference to FIG. The cost rate (actual value) indicates the actual cost rate when receiving an order for the corresponding project. The history database 141 corresponds to teacher data used when the prediction model 142 is machine-learned.

The prediction model 142 is a model generated by machine learning based on the history database 141 . For example, the prediction model 142 corresponds to a model such as a decision tree, a random forest, etc., and when feature amount information is input, a cost rate (actual value) corresponding to the feature amount information is calculated.

Returning to the description of FIG. The control unit 150 has an acquisition unit 151 , a learning unit 152 , a prediction unit 153 and a generation unit 154 . The control unit 150 is implemented by, for example, a CPU (Central Processing Unit) or an MPU (Micro Processing Unit). Also, the control unit 150 may be implemented by an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array).

The acquisition unit 151 acquires various types of information regarding the history of project information. For example, the acquisition unit 151 acquires a set of past project information and cost rate (actual value) from the terminal device 10 or another external device, etc., and acquires the past project information and cost rate (actual value). are associated with each other and registered in the history database 141 .

The acquisition unit 151 repeatedly executes the above process every time it acquires a set of past project information and cost rate (actual value) from the terminal device 10 or another external device.

The learning unit 152 performs machine learning based on the relationship between the feature amount information held in the history database 141 and the cost rate (actual value) to generate the prediction model 142 . The learning unit 152 records information on the generated prediction model 142 in the storage unit 140 . Any conventional technique may be used for the machine learning performed by the learning unit 152 .

The learning unit 152 may relearn the prediction model 142 when a predetermined period of time has passed since the previous machine learning was performed and new information is accumulated in the history database.

When receiving a prediction request for project information from the terminal device 10 , the prediction unit 153 inputs the feature amount information of the project information to the prediction model 142 to calculate the predicted cost rate. The prediction unit 153 outputs the predicted cost rate and the received project information to the generation unit 154 . The prediction unit 153 also specifies conditional branching (explanatory variables) included in the prediction model 142 and outputs the degree of influence of each explanatory variable when calculating the predicted cost rate to the generation unit 154 . It can be said that the larger the influence of the explanatory variable, the larger the influence on the calculation of the forecast cost rate.

Here, when the project information is received from each of the plurality of terminal devices 10, the prediction unit 153 inputs the feature amount information of each project information to the prediction model 142, and calculates each predicted cost rate. The prediction unit 153 outputs the predicted cost rate corresponding to each piece of project information to the generation unit 154 .

The generation unit 154 acquires the project information, the predicted cost rate, and the degree of influence of each explanatory variable from the prediction unit 153, and generates screen information. The screen information generated by the generator 154 corresponds to the screen information shown in FIG. For example, the generation unit 154 generates information for the display area 30a of the screen information 30 based on basic information included in the project information.

The generation unit 154 generates information for the display area 30 b of the screen information 30 based on the planned cost rate included in the project information and the predicted cost rate predicted by the prediction unit 153 . For example, the generation unit 154 calculates a difference value by subtracting the “predicted cost rate” from the “predicted cost rate”, and compares the difference value with the first to fifth thresholds to obtain the cost prediction categories A to Identify any of F. As described with reference to FIG. 4, the generation unit 154 sets the planned cost rate, the cost rate prediction category, and the person in charge of the basic information in the display area 30b.

Based on the relationship between the item (explanatory variable) and the degree of influence acquired from the prediction unit 153, the generation unit 154 generates an item (explanatory variable) of feature information that has influenced the prediction result of the prediction model (decision tree) and Generate information in which the degree of influence is associated with the degree of influence. For example, the generating unit 154 sorts the items in descending order of influence and selects the top N items. The generation unit 154 visually sets the relationship between the selected top N items and the degree of influence in the screen information 30 .

Here, when the generation unit 154 acquires predicted cost rates for a plurality of pieces of project information, it identifies each cost prediction category based on the scheduled cost rate and the predicted cost rate for each piece of project information. The generation unit 154 sets the set of the planned cost rate, cost forecast category, and person in charge for each piece of project information in the display area 30b. By referring to the display area 30b, the user can compare the planned cost rate and the cost forecast category for each person in charge.

In addition to the screen information 30 described above, the generation unit 154 may generate screen information for a bubble chart, which will be described below. FIG. 7 is a diagram showing an example of a bubble chart. In FIG. 7, the horizontal axis corresponds to the planned cost rate. The vertical axis corresponds to the cost rate forecast category. The generation unit 154 sets an ellipse corresponding to the project information in an area corresponding to the planned cost rate and the cost rate prediction classification.

For example, if the planned cost rate corresponding to the first project information is "86%" and the cost rate prediction category is "A", the ellipse corresponding to the first project information is ellipse O1. Assuming that the planned cost rate corresponding to the second project information is "70%" and the cost rate prediction category is "D", the ellipse corresponding to the second project information is ellipse O2.

Assuming that the planned cost rate corresponding to the third project information is "83%" and the cost rate prediction category is "E", the ellipse corresponding to the first project information is ellipse O3. Assuming that the planned cost rate corresponding to the fourth project information is "86%" and the cost rate prediction category is "F", the ellipse corresponding to the fourth project information is ellipse O4.

The generating unit 154 increases the size of the ellipse as the amount of sales set in the basic information of the project information increases. The generation unit 154 sets a color corresponding to the combination of the planned cost rate and the cost rate prediction category to an ellipse. In the example shown in FIG. 7, ellipse O1 is displayed in a first color (eg, red), ellipse O2 is displayed in a second color (eg, blue), and ellipses O3 and O4 are displayed in a third color (eg, blue). , yellow). The relationship between the set of planned cost rate and cost rate forecast category and the color is set in advance by the administrator. By displaying with , it is possible to easily grasp the project information that becomes unprofitable.

The generation unit 154 transmits the screen information 30 described with reference to FIG. 4 and the screen information of the bubble chart described with reference to FIG. 7 to the terminal device 10 for display. The generation unit 154 may output the screen information 30 described with reference to FIG. 4 and the screen information of the bubble chart described with reference to FIG. 7 to the display unit 130 for display.

Next, an example of the processing procedure of the information processing apparatus 100 according to this embodiment will be described. FIG. 8 is a flow chart showing the processing procedure of the information processing apparatus according to this embodiment. As shown in FIG. 8, the learning unit 152 of the information processing device 100 performs machine learning based on the history database 141 to generate the prediction model 142 (step S101).

The prediction unit 153 of the information processing device 100 acquires a prediction request for project information (step S102). The prediction unit 153 inputs the feature amount information to the prediction model 142 to calculate the predicted cost rate (step S103).

The generation unit 154 of the information processing device 100 identifies the cost rate prediction classification based on the difference value obtained by subtracting the planned cost rate from the predicted cost rate (step S104). The generation unit 154 generates screen information of the prediction result (step S105).

The generation unit 154 generates screen information of the bubble chart based on the planned cost rate and the cost rate prediction category (step S106). The generation unit 154 causes the terminal device 10 to display the prediction result and the screen information of the bubble chart (step S107).

Next, the effects of the information processing apparatus 100 according to this embodiment will be described. The information processing apparatus 100 generates the prediction model 142 by executing machine learning based on the relationship between the feature amount information regarding past projects and the actual value of the cost rate regarding such past projects. When the information processing apparatus 100 receives a prediction request, the information processing apparatus 100 predicts the predicted cost rate by inputting the feature amount information related to the project of the prediction request into the prediction model 142 . This makes it possible to detect unprofitable projects.

The information processing apparatus 100 uses facility construction, server/storage construction, NW construction, DB construction, cloud construction, matching with applications, migration work, the number of server devices, the number of terminal devices, bases, etc. as feature amount information of project information. Use the number, number of departments, technical skills, product selection method, etc. As a result, it is possible to set the feature amount related to the risk of the project information and improve the accuracy of the forecast cost rate.

The information processing device 100 generates a bubble chart based on the relationship between the planned cost rate and the predicted cost rate category, and causes the terminal device 10 to display the bubble chart. This makes it possible to visually grasp unprofitable projects easily.

When the information processing apparatus 100 acquires predicted cost rates for a plurality of pieces of project information, the information processing apparatus 100 identifies each cost prediction category based on the planned cost rate and the predicted cost rate for each piece of project information. The information processing device 100 causes the screen information to display the set of the planned cost rate, the cost forecast category, and the person in charge for each piece of project information. By referring to such screen information, the user can compare the planned cost rate and the cost forecast category for each person in charge.

The information processing apparatus 100 relearns the prediction model 142 when new information is accumulated in the history database 141 after a predetermined period of time has elapsed since the previous machine learning was performed. This allows the predictive model 142 to be upgraded with the latest information.

By the way, in the present embodiment, the case where the information processing apparatus 100 uses the history database 141 to perform machine learning and generate the prediction model 142 has been described, but the present invention is not limited to this. A predicted cost rate may be predicted by obtaining a learned prediction model. That is, the information processing device 100 does not necessarily have to generate the prediction model 142 .

The information processing apparatus 100 performs machine learning of the prediction model 142 using the feature amount information before receiving the project order, but is not limited to this. The predictive model 142 may be generated by performing learning.

The information processing apparatus 100 generates screen information for each piece of project information. good.

When the information processing apparatus 100 receives from the user a selection of an item to be excluded from among the plurality of items (explanatory variables) having a large degree of influence displayed on the screen information 30, the information processing apparatus 100 corresponds to the selected item. The predicted cost rate may be calculated again by excluding the information to be used from the feature quantity information. For example, when the information processing apparatus 100 receives an instruction from the user operating the terminal device 10 to exclude the item "number of server devices", the information processing apparatus 100 excludes the record of the item "number of server devices". The calculated feature amount information is input to the prediction model 142 to calculate the predicted cost rate.

Next, an example of the hardware configuration of a computer that implements the same functions as the information processing apparatus 100 shown in the above embodiment will be described. FIG. 9 is a diagram illustrating an example of a hardware configuration of a computer that implements functions similar to those of the information processing apparatus of the embodiment.

As shown in FIG. 9, a computer 200 has a CPU 201 that executes various arithmetic processes, an input device 202 that receives data input from a user, and a display 203 . The computer 200 also has a communication device 204 and an interface device 205 for exchanging data with an external device or the like via a wired or wireless network. The computer 200 also has a RAM 206 that temporarily stores various information, and a hard disk device 207 . Each device 201 - 207 is then connected to a bus 208 .

The hard disk device 207 has an acquisition program 207a, a learning program 207b, a prediction program 207c, and a generation program 207d. Also, the CPU 201 reads out each of the programs 207 a to 207 d and develops them in the RAM 206 .

Acquisition program 207a functions as acquisition process 206a. Learning program 207b functions as learning process 206b. The prediction program 207c functions as a prediction process 206c. The generation program 207d functions as a generation process 206d.

The processing of the acquisition process 206 a corresponds to the processing of the acquisition unit 151 . The processing of the learning process 206 b corresponds to the processing of the learning unit 152 . The processing of the prediction process 206 c corresponds to the processing of the prediction section 153 . Processing of the generation process 206 d corresponds to processing of the generation unit 154 .

Note that the programs 207a to 207d do not necessarily have to be stored in the hard disk device 207 from the beginning. For example, each program is stored in a “portable physical medium” such as a flexible disk (FD), CD-ROM, DVD, magneto-optical disk, IC card, etc. inserted into the computer 200 . Then, the computer 200 may read and execute each of the programs 207a-207d.

10a, 10b, 10c terminal device 20 network 100 information processing device 110 communication unit 120 input unit 130 display unit 140 storage unit 141 history database 142 prediction model 150 control unit 151 acquisition unit 152 learning unit 153 prediction unit 154 generation unit

Claims (8)

Characteristic quantities related to risks of projects implemented in the past, including the scale of hardware used in said projects and the details of services provided in said projects, and the relationship between said characteristic quantities and actual values of cost rates related to said projects. A learning unit that generates a model by executing machine learning based on
An information processing apparatus, comprising: a prediction unit that inputs a feature amount of a certain project into the model and predicts a cost rate of the certain project. The scale of the hardware includes the number of server devices, the number of terminal devices, and the number of bases, and the content of the service includes information on whether to build facilities, information on whether to build servers and storage, 2. The information processing apparatus according to claim 1, further comprising information as to whether or not to construct a network, information as to whether or not to construct a cloud, and information as to whether or not to perform data migration. calculating a difference between the cost rate of the project predicted by the prediction unit and the planned cost rate of the project preset by the user, and creating a graph showing the relationship between the planned cost rate and the difference 3. The information processing apparatus according to claim 1, further comprising a generating unit for generating. The prediction unit obtains a plurality of feature amounts set by a plurality of users with respect to the same project, inputs the plurality of feature amounts into the model, predicts a plurality of cost rates, and predicts a plurality of cost rates. 4. The information processing apparatus according to claim 3, further generating information in which the user and the cost rate corresponding to the feature set by the user are associated with each other. an acquiring unit for accumulating the feature amount of the project and the actual cost ratio of the project in a storage unit each time the actual value of the cost ratio of the project is acquired; 5. The information processing apparatus according to any one of claims 1 to 4, wherein the machine learning of the model is executed again using the information accumulated in the storage unit. Characteristic quantities related to risks of projects implemented in the past, including the scale of hardware used in said projects and the details of services provided in said projects, and the relationship between said characteristic quantities and actual values of cost rates related to said projects. A storage unit that stores a model generated by executing machine learning based on
An information processing apparatus, comprising: a prediction unit that inputs a feature amount of a certain project into the model and predicts a cost rate of the certain project. A computer-implemented prediction method comprising:
Characteristic quantities related to risks of projects implemented in the past, including the scale of hardware used in said projects and the details of services provided in said projects, and the relationship between said characteristic quantities and actual values of cost rates related to said projects. Generate a model by performing machine learning based on
A forecasting method comprising: inputting a characteristic amount of a certain project into the model, and executing a process of forecasting a cost rate of the certain project. to the computer,
Characteristic quantities related to risks of projects implemented in the past, including the scale of hardware used in said projects and the details of services provided in said projects, and the relationship between said characteristic quantities and actual values of cost rates related to said projects. Generate a model by performing machine learning based on
A forecasting program, characterized by inputting a feature quantity of a certain project into the model and causing it to execute a process of forecasting a cost rate of the certain project.

Images