Information processing device, information processing method, and information processing program

Abstract

Accurately predict which projects are at risk. [Solution] The information processing device 100 has an acquisition unit 122 that acquires one or more pieces of information selected from a questionnaire on the project status and information on working hours, and information regarding the income and expenditure of the project, a prediction unit 123 that uses a learning model to predict information regarding the development deficit of the project using the one or more pieces of information selected from a questionnaire on the project status and information on working hours, and information regarding the income and expenditure of the project acquired by the acquisition unit 122, as input, and an output unit 124 that outputs the information regarding the development deficit of the project predicted by the prediction unit 123.

Description

[Technical field]

[0001] The present invention relates to an information processing device, an information processing method, and an information processing program. [Background technology]

[0002] Conventionally, there are techniques for quantitatively and efficiently determining the risks in a project and facilitating the consideration of subsequent countermeasures (for example, Patent Literature 1). For example, there is a technique for calculating the failure probability of a project and calculating the failure factor as the contribution rate of explanatory variables that contributed to the calculation of the failure probability of the project. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent Publication No. 2022-032115 Summary of the Invention [Problem to be solved by the invention]

[0004] The conventional techniques have problems in that they are unable to accurately predict projects that are at risk. For example, they may not be able to detect early signs of a development deficit. In addition, they may not use information that is suitable for predicting projects that are at risk.

[0005] The present invention has been made in consideration of the above, and aims to provide an information processing device, an information processing method, and an information processing program that enable risk countermeasures using information on similar past cases. [Means for solving the problem]

[0006] In order to solve the above-mentioned problems and achieve the objective, the information processing device of the present invention is characterized by having an acquisition unit that acquires one or more of a questionnaire on the project status and information on working hours, and information on the income and expenditure of the project, a prediction unit that uses a learning model to predict information on the development deficit of the project using as input one or more of the questionnaire on the project status and information on working hours, and the information on the income and expenditure of the project acquired by the acquisition unit, and an output unit that outputs the information on the development deficit of the project predicted by the prediction unit. Effect of the Invention

[0007] According to the present invention, it is possible to predict with high accuracy which projects are at risk. [Brief description of the drawings]

[0008] [Figure 1] FIG. 1 is a diagram showing an overview of a process performed by an information processing device according to an embodiment. [Diagram 2] FIG. 2 is a diagram illustrating an example of a functional configuration of the information processing device according to the embodiment. [Diagram 3] FIG. 3 is a diagram illustrating an example of learning data according to the embodiment. [Figure 4] FIG. 4 is a diagram illustrating an example of learning data according to the embodiment. [Diagram 5] FIG. 5 is a diagram illustrating an example of a learning process according to the embodiment. [Figure 6] FIG. 6 is a diagram illustrating an example of a prediction process according to the embodiment. [Figure 7] FIG. 7 is a diagram illustrating an example of the output process according to the embodiment. [Figure 8] FIG. 8 is a diagram illustrating an example of the output process according to the embodiment. [Figure 9] FIG. 9 is a flowchart showing an example of a flow of processing by the information processing device according to the embodiment. [Figure 10]FIG. 10 is a diagram showing an overview of the process performed by the information processing device according to the embodiment. [Figure 11] FIG. 11 is a diagram illustrating an example of a functional configuration of the information processing device according to the embodiment. [Figure 12] FIG. 12 is a diagram illustrating an example of a calculation process according to the embodiment. [Figure 13] FIG. 13 is a diagram illustrating an example of a calculation process according to the embodiment. [Figure 14] FIG. 14 is a flowchart showing an example of a flow of processing by the information processing device according to the embodiment. [Figure 15] FIG. 15 is a diagram illustrating an example of a hardware configuration of an information processing device according to an embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] Hereinafter, with reference to the drawings, an embodiment of an information processing device, an information processing method, and an information processing program according to the present application will be described in detail. Note that the present invention is not limited to the embodiment. In addition, in the description of the drawings, the same parts are denoted by the same reference numerals, and duplicated explanations will be omitted.

[0010] [First embodiment] 1. Overview First, an overview of the processing performed by the information processing device 100 will be described with reference to Fig. 1. Fig. 1 is a diagram for explaining the overview of the processing performed by the information processing device 100.

[0011] Conventionally, there have been technologies that quantitatively and efficiently determine project risks and make it easy to consider countermeasures, but there are cases where it is not possible to accurately predict projects that have risks. For example, conventional technologies may not be able to detect early signs of development losses. Also, for example, there are cases where information suitable for predicting projects that have risks is not used.

[0012] Therefore, the information processing device 100 according to the embodiment takes as input one or more pieces of information, such as a questionnaire about the project status and information about working hours, and information about the income and expenditures of the project, and uses a learning model to predict and output information about the development deficit of the project.

[0013] For example, as shown in FIG. 1, the information processing device 100 according to the embodiment acquires periodic surveys on project status, information on developer working hours, and project income and expenditure information, and inputs these information into a learning model to predict the probability of a deficit, and outputs information on the probability of a development deficit for each project.

[0014] In this way, the information processing device 100 of this embodiment inputs one or more pieces of information, such as a questionnaire about the project status and information about working hours, and information about the income and expenditures of the project, into a learning model to predict the probability of a development deficit, thereby achieving the effect of being able to accurately predict projects that involve risk.

[0015] 2. Configuration of Information Processing Device Next, an example of the configuration of the information processing device 100 will be described with reference to Fig. 2. Fig. 2 is a diagram showing an example of the functional configuration of the information processing device 100. As shown in Fig. 2, the information processing device 100 has a communication unit 110, a control unit 120, and a storage unit 130. Note that each of these units may be held in a distributed manner by a plurality of devices. The processing of each of these units will be described below.

[0016] (Communication unit 110) The communication unit 110 is realized by a NIC (Network Interface Card) or the like, and enables communication between the control unit 120 and an external device via an electric communication line such as a LAN (Local Area Network) or the Internet. For example, the communication unit 110 transmits and receives information to and from the terminal device 200.

[0017] (Storage unit 130) The storage unit 130 is realized by a semiconductor memory element such as a random access memory (RAM) or a flash memory, or a storage device such as a hard disk or an optical disk. As shown in FIG. 2, the storage unit 130 includes a learning data storage unit 131, a project information storage unit 132, and a prediction information storage unit 133.

[0018] The learning data storage unit 131 stores information used by the learning unit 121 described below for learning a predetermined learning model. For example, the learning data storage unit 131 stores a questionnaire about the project status, comments about the project status, working hours, information about the income and expenditure of the project, and information about a development deficit of the actual project. Note that a development deficit refers to a state in which gross profit on sales is equal to or less than a predetermined threshold. More specifically, a development deficit refers to a state in which gross profit on sales is equal to or less than -1 million yen.

[0019] (Learning data storage unit 131) The information stored in the learning data storage unit 131 will be described with reference to FIG. 3 and FIG. 4. FIG. 3 is an example of a questionnaire on the project status. The questionnaire on the project status is a response result to a question such as whether the personnel participating in the project feel any problems with the project. For example, as shown in FIG. 3, the questionnaire on the project status includes an instruction such as "Do you feel any of the following problems with the project? Please select one that applies." and an item such as "1. There is a shortage of personnel and skills required for the project, which causes poor quality and delays in progress (lack of system)." The personnel participating in the project evaluate each item in stages such as "1. I think so, 2. I think so to some extent, 3. I don't think so to some extent, 4. I don't think so." Note that the above description is merely an example, and the contents of the questionnaire on the project status are not limited to this.

[0020] Comments on the project status are written responses to questions such as whether the people participating in the project feel there are any problems with the project. For example, as shown in Figure 3, comments on the project status are responses to questions such as, "Please write down any concerns you have about moving forward with the project. If there is nothing to write, please write 'None in particular.'" Specific examples include, "We are unable to confirm whether the project is operating normally overall," and "The work content for month XX is unclear."

[0021] 4 is a table diagram showing an example of learning data according to the embodiment. As shown in FIG. 4, the information stored in the learning data storage unit 131 is stored in a list format having items such as "No.", "Variable name," and "Calculation method / content." For example, the questionnaire on the project status in the example of FIG. 4 is information on "No.: 16," "Variable name: Medical checkup," and "Calculation method / content: PJ average of the proportion of people who answered "I think so" or "I think so to some extent" to PJ medical checkup Q1-Q10."

[0022] In the example of FIG. 4, the working hours are information such as "No: 20", "Variable name: Overtime hours_Department average difference", and "Calculation method / content: Difference between average overtime hours by department and three-month moving average overtime hours of the project". In addition, for example, information related to the income and expenditure of the project includes the project's sales profit, the project's customer name, the month the project is executed, and the project's subcontracting costs. In the example of FIG. 4, information related to the income and expenditure of the project is information such as "No: 9", "Variable name: Plan variance_Subcontracting costs vs. labor costs", and "Calculation method / content: Plan variance_Cost of sales_Total subcontracting costs ÷ Plan variance_Cost of sales_Labor costs".

[0023] Furthermore, for example, information regarding the development deficit of the actual project is an actual measurement value of information regarding the income and expenditure of the project, and includes the sales profit of the project, the customer name of the project, the month of execution of the project, and the outsourcing cost of the project. In the example of Fig. 4, information regarding the development deficit of the actual project is information such as "No: 9", "Variable name: Plan variance_Outsourcing cost vs. Labor cost", and "Calculation method / content: Plan variance_Cost of sales_Total outsourcing cost ÷ Plan variance_Cost of sales_Labor cost". Note that the items shown in Fig. 4 are merely examples, and the items stored in the learning data storage unit 131 are not limited to these.

[0024] (Project information storage unit 132) The project information storage unit 132 stores a questionnaire about the project status, comments about the project status, working hours, information about the income and expenditure of the project, and information about the development deficit predicted by the prediction unit 123 described below.

[0025] (Control unit 120) The control unit 120 is realized using a CPU (Central Processing Unit), an NP (Network Processor), an FPGA (Field Programmable Gate Array), etc., and executes a processing program stored in a memory. As shown in Fig. 2, the control unit 120 has a learning unit 121, an acquisition unit 122, a prediction unit 123, and an output unit 124. Each unit of the control unit 120 will be described below.

[0026] (Learning Section 121) The learning unit 121 learns a learning model using at least one of a questionnaire about the project status and information about working hours, information about the income and expenditure of the project, and information about the development deficit of the actual project as correct answer data. For example, the learning unit 121 learns the relationship between at least one of a questionnaire about the project status and information about working hours, information about the income and expenditure of the project, and information about the development deficit of the actual project, and learns a learning model that uses at least one of a questionnaire about the project status and information about working hours, and information about the income and expenditure of the project as inputs and outputs information about the development deficit of the project.

[0027] In this embodiment, the learning unit 121 is described on the assumption that learning is performed using a LightGBM (Light Gradient Boosting Machine), but is not limited to this as long as it is a learning model that can predict information related to the development deficit of a project. In addition, the learning unit 121 can learn a supervised learning model, an unsupervised learning model, a reinforcement learning model, etc.

[0028] Here, an example of the learning process according to the embodiment will be described with reference to Fig. 5. Fig. 5 is a diagram showing an example of the learning process according to the embodiment. In the example shown in Fig. 5, learning was performed by dividing all 21,669 pieces of data into 80% as learning data and 20% as test data. In addition, in order to prevent variation in the conditions of the learning data and the test data, random division was performed multiple times, and a division method that minimized the difference between the standardized main items was adopted.

[0029] Then, the training data was divided into groups using k-fold cross validation, and the accuracy of the model was calculated from the average prediction value for the test data of the model created in each group. Note that in the example of Figure 5, the test data is the data used for accuracy evaluation, the training data is the data used when training each fold, and the validation data is the data used for parameter adjustment when training each fold.

[0030] Furthermore, the learning unit 121 can process data when learning the model. For example, data with a small deficit or surplus is excluded from the learning data. More specifically, if the definition of a development deficit is "gross profit is -1 million yen or less" and the objective variable is "there will be at least one month of development deficit in the next three months", data in which the gross profit for all three months is within the range of -1 million yen < gross profit < 0, or data in which the gross profit for two months is within the range of -1 million yen < gross profit < 0 and does not include a month of development deficit, or data in which the gross profit for one month is within the range of -1 million yen < gross profit < 0 and does not include a month of development deficit is excluded from the learning data.

[0031] Also, for example, the imbalance is corrected for the learning data. More specifically, the majority data is randomly removed to match the minority data so that the number of data with the objective variable of 1 and 0 is approximately equal. Also, for example, for the learning data and test data, data with all income / expense items of 0, such as "sales_total sales, cost of sales_total cost of sales, cost of sales_labor costs, cost of sales_total outsourcing costs, gross profit," are removed for plan variance, current month forecast, and initial plan.

[0032] (Acquisition part 122) The acquiring unit 122 acquires at least one of a questionnaire about the project status and information about working hours, and information about the income and expenditure of the project. For example, the acquiring unit 122 acquires questionnaire information about the project status, "health check", information about working hours, "overtime hours", and information about the income and expenditure of the project, "sales profit of the project, customer name of the project, month of execution of the project, outsourcing cost of the project".

[0033] The acquiring unit 122 periodically collects a questionnaire on the project status, information on working hours, and information on the income and expenditure of the project. For example, the acquiring unit 122 collects a questionnaire on the project status, information on working hours, and information on the income and expenditure of the project every three months. Note that three months is an example, and the acquiring unit 122 can collect information at any period depending on the purpose.

[0034] (Prediction section 123) The prediction unit 123 receives as input one or more pieces of information on the project status, the questionnaire and the working hours information acquired by the acquisition unit 122, and information on the income and expenditures of the project, and predicts information on the development deficit of the project using a learning model.

[0035] For example, the prediction unit 123 predicts information regarding the development deficit of the project by inputting one or more pieces of information regarding the project status and information about working hours acquired by the acquisition unit 122, and information regarding the income and expenditures of the project, into a learning model that outputs information regarding the development deficit of the project.

[0036] The prediction unit 123 predicts the probability of a project running in the red (unprofitability coefficient) using a learning model with a questionnaire about the project status, information about working hours, and information about the project's income and expenditure as input. For example, the prediction unit 123 predicts the probability of a project running in the red by inputting the questionnaire about the project status, information about working hours, and information about the project's income and expenditure as input to a learning model that outputs the probability of a project running in the red. For example, the prediction unit 123 predicts the probability of a development deficit occurring in the target project within a predetermined period. More specifically, the prediction unit 123 predicts the probability of a development deficit occurring in the target project within three months.

[0037] The prediction unit 123 predicts information regarding the deficit using a learning model, using as input one or more of the project's revenue and expenditure information, namely, the project's sales profit, the project's customer name, the project's execution month, and the project's outsourcing costs.

[0038] For example, the prediction unit 123 predicts information regarding deficits using a learning model with input of one or more of the questionnaire information regarding the project status acquired by the acquisition unit 122, "health checkup", and working hour information "overtime hours", and one or more of the project's sales profit "gross profit", the project's customer name "customer name", the project execution month "number of months", and the project's outsourcing cost "outsourcing cost" as information regarding the project's income and expenditure.

[0039] Here, an example of the prediction process according to the embodiment will be described with reference to Fig. 6. Fig. 6 is a diagram showing an example of the prediction process according to the embodiment. Note that the prediction method shown in Fig. 6 is merely a conceptual description of regression analysis based on a learning model, and is not limited to an actual prediction method.

[0040] 6, the prediction unit 123 performs a regression analysis based on the learning model (LightGBM) learned by the learning unit 121, and calculates the probability of a deficit. For example, the prediction unit 123 determines whether the "current month forecast_gross profit" input in the first layer satisfies the condition "XXXXXX yen or less."

[0041] If the above-mentioned condition is satisfied, the prediction unit 123 then judges whether or not the "Plan variance_total sales_gross profit margin" input in the next hierarchical level satisfies the condition of "XX% or less". In this way, the prediction unit 123 repeats the judgment in multiple hierarchical levels to calculate the probability of a deficit (for example, 98%). At this time, the prediction unit 123 may assign an unprofitable label "1" indicating that a deficit is predicted to occur to a project with a deficit probability of 50% or more, and an unprofitable label "0" indicating that a deficit is predicted not to occur to a project with a deficit probability of less than 50%.

[0042] (Output unit 124) The output unit 124 outputs information related to the development deficit of the project predicted by the prediction unit 123. For example, the output unit 124 outputs, for each project, an unprofitability coefficient (deficit occurrence probability) indicating the probability that the project will become unprofitable, and an unprofitability label indicating whether the project will become unprofitable, as information related to the development deficit of the project predicted by the prediction unit 123.

[0043] Here, an example of the output process according to the embodiment will be described with reference to Fig. 7. Fig. 7 is a diagram showing an example of the output process according to the embodiment. As shown in Fig. 7, the output unit 124 outputs information on the business division "*****", department "*****", project aggregation ID "*****", project name "*****", customer name "*****", unprofitable coefficient "0.998", and unprofitable label "1" to the user's terminal device 200 or the like.

[0044] At this time, the output unit 124 can perform output processing by ranking based on any item. For example, the output unit 124 outputs the deficit occurrence probability in the order of the project with the highest deficit occurrence probability predicted by the prediction unit 123. For example, the output unit 124 outputs the business division, department, PJ aggregation ID, PF name, customer name, unprofitability coefficient, and unprofitability label in the order of the project with the highest unprofitability coefficient predicted by the prediction unit 123.

[0045] The output unit 124 outputs the results of the survey on the project status collected by the acquisition unit 122, a list of the working hours of the people participating in the project, and information on the income and expenditure of the project, as well as information on the deficit predicted by the prediction unit 123. For example, the output unit 124 outputs, as the results of the survey on the project status collected by the acquisition unit 122, the negative response rate for each question, the average overtime hours of the people participating in the project and the number of people whose overtime hours exceed 60 hours, and information on the period profit and loss of the project, together with the unprofitability coefficient predicted by the prediction unit 123.

[0046] Here, an example of the output process according to the embodiment will be described with reference to Fig. 8. Fig. 8 is a diagram showing an example of the output process according to the embodiment. For example, as shown in Fig. 8, the output unit 124 outputs to the user's terminal device 200 or the like information on the transition of the unprofitability coefficient of the specific project predicted by the prediction unit 123 (Fig. 8(1)), the transition of the period profit and loss of the specific project (Fig. 8(2)), the transition of the average overtime hours of the personnel participating in the specific project and the number of people whose overtime hours exceed 60 hours (Fig. 8(3)), and the transition of the negative response rate for each question as a questionnaire (PJ health check) on the project status of the specific project (Fig. 8(4)).

[0047] (Terminal device 200) The terminal device 200 is an information processing terminal that receives information output from the information processing device 100 and displays it to a user, etc. Note that the terminal device 200 according to the present embodiment may be, for example, a smartphone, a PDA (Personal Digital Assistant), a tablet terminal, a notebook PC (Personal Computer), etc., and is not limited thereto.

[0048] [3. Flowchart] Next, an example of a process flow by the information processing device 100 will be described with reference to Fig. 9. Fig. 9 is a flowchart showing an example of a process flow by the information processing device 100 according to the embodiment. Note that the following steps can be executed in a different order, and some processes may be omitted.

[0049] First, the acquisition unit 122 acquires at least one of a questionnaire about the project status and information about working hours, and information about the income and expenditure of the project (step S101). For example, the acquisition unit 122 acquires questionnaire information about the project status, "health check," information about working hours, "overtime hours," and information about the income and expenditure of the project, "sales profit of the project, customer name of the project, month of execution of the project, outsourcing cost of the project."

[0050] Next, the prediction unit 123 inputs one or more of the questionnaire about the project status and the working hour information acquired by the acquisition unit 122, and the information about the income and expenditure of the project, and predicts information about the development deficit of the project using a learning model (step S102).

[0051] For example, the prediction unit 123 predicts information regarding the development deficit of the project by inputting one or more pieces of information regarding the project status and information about working hours acquired by the acquisition unit 122, and information regarding the income and expenditures of the project, into a learning model that outputs information regarding the development deficit of the project.

[0052] Next, the prediction unit 123 judges whether or not information on development deficits has been predicted for all projects (step S103). If the prediction unit 123 judges that information on development deficits has not been predicted for all projects (step S103: No), the process of step S102 is performed again.

[0053] On the other hand, if it is determined that the prediction unit 123 has predicted information on the development deficit for all projects (step S103: Yes), the output unit 124 outputs information on the development deficit of the projects predicted by the prediction unit 123 (step S104). For example, the output unit 124 outputs the unprofitability coefficient and unprofitability label of the projects for each project to the user's terminal device 200 as information on the development deficit of the projects predicted by the prediction unit 123.

[0054] Next, the output unit 124 determines whether or not a project selection from the user has been accepted (step S105). If the output unit 124 determines that a project selection from the user has not been accepted (step S105: No), the process of step S105 is performed again.

[0055] On the other hand, if the output unit 124 determines that it has accepted a project selection from the user (step S105: Yes), the output unit 124 outputs, for the project whose selection has been accepted, the results of a survey on the project status, a list of the working hours of those participating in the project, and information regarding the income and expenditure of the project, as well as information regarding the deficit predicted by the prediction unit 123 (step S106).

[0056] For example, for a project for which a selection has been accepted, the output unit 124 outputs to the user's terminal device 200 the negative response rate for each question as a result of a questionnaire on the project status, the average overtime hours of the personnel participating in the project and the number of people whose overtime hours exceeded 60 hours, and information on the project's periodic profit and loss, together with the unprofitability coefficient predicted by the prediction unit 123.

[0057] 4. Effects The information processing device 100 according to the embodiment has an acquisition unit 122 that acquires one or more pieces of information, including a questionnaire about the project status and information about working hours, and information about the income and expenditure of the project, a prediction unit 123 that predicts information about the development deficit of the project using a learning model, using as input one or more pieces of information, including a questionnaire about the project status and information about working hours, and information about the income and expenditure of the project acquired by the acquisition unit 122, and an output unit 124 that outputs information about the development deficit of the project predicted by the prediction unit 123.

[0058] As a result, the information processing device 100 can input information suitable for predicting the occurrence of a development deficit, such as a questionnaire about the project status, working hours, and information about the project's income and expenditures, into a learning model to predict information about a project's development deficit, thereby detecting signs of a deficit at an early stage and accurately predicting projects that are at risk.

[0059] That is, the information processing device 100 makes a prediction using, as an input, factors that have a high degree of influence on the occurrence of a development deficit, thereby making it possible to predict with high accuracy the probability of the occurrence of a development deficit for each project.

[0060] The information processing device 100 according to the embodiment further includes a learning unit 121 that learns a learning model using one or more pieces of information, including a questionnaire about the project status and information about working hours, information about the income and expenditures of the project, and information about the development deficit of the actual project, as correct answer data.

[0061] As a result, the information processing device 100 can accurately predict projects that are at risk by training a learning model using one or more of the questionnaire on the project status and information on working hours, information on the income and expenditure of the project, and information on the development deficit of the actual project as correct answer data.

[0062] The prediction unit 123 of the information processing device 100 according to the embodiment uses a learning model with a questionnaire regarding the project status, information on working hours, and information regarding the income and expenditures of the project as input to predict the probability of a deficit occurring for the project, and the output unit 124 outputs the deficit occurrence probabilities of the projects in order of the highest deficit occurrence probabilities predicted by the prediction unit 123.

[0063] As a result, the information processing device 100 predicts the probability of a project incurring a deficit based on a questionnaire about the project status, information about working hours, and information about the project's income and expenditures, and outputs the results in order of the highest probability of incurring a deficit, making it easier to understand projects that involve risk and reducing the burden on the user.

[0064] The prediction unit 123 of the information processing device 100 according to the embodiment predicts information regarding a deficit using a learning model, using as input one or more of the following as information regarding the income and expenditure of the project: the sales profit of the project, the customer name of the project, the month of execution of the project, and the outsourcing cost of the project.

[0065] As a result, the information processing device 100 can input information suitable for predicting the occurrence of development deficits, such as a questionnaire regarding the project status, working hours, and information regarding the income and expenditures of the project (sales profit of the project, customer name of the project, month of execution of the project, outsourcing costs of the project), into the learning model to predict information regarding the development deficit of the project, thereby enabling it to accurately predict projects that are at risk.

[0066] The acquisition unit 122 of the information processing device 100 according to the embodiment periodically collects a questionnaire on the project status, information on working hours, and information on the income and expenditure of the project, and the output unit 124 outputs the results of the questionnaire on the project status collected by the acquisition unit 122, a list of working hours of those participating in the project, and information on the income and expenditure of the project, as well as information on the deficit predicted by the prediction unit 123.

[0067] In this way, the information processing device 100 can prevent information used for prediction from becoming obsolete by periodically collecting information suitable for predicting the occurrence of a development deficit, and can predict projects with high accuracy for which there is risk. In addition, the information processing device 100 can present detailed project information together with the probability of development deficit occurrence, thereby providing a user who has seen the prediction result with information for determining whether a development deficit is likely to actually occur, thereby reducing the burden on the user.

[0068] [Second embodiment] In the first embodiment, an example was described in which at least one of a questionnaire about the project status and information about working hours, and information about the income and expenditure of the project are input to the learning model to predict information about the development deficit of the project, but in the second embodiment described below, an example is described in which the result of clustering information about comments about the project status is further used as an input to the learning model to predict information about the development deficit of the project. Note that explanations of contents common to the first embodiment will be omitted as appropriate.

[0069] 1. Overview First, an overview of the processing performed by the information processing device 100 will be described with reference to Fig. 10. Fig. 10 is a diagram for explaining an overview of the processing performed by the information processing device 100. As shown in Fig. 10, the information processing device 100 according to the embodiment predicts the probability of occurrence of a development deficit by using a questionnaire on the project status, information on the working hours of the developer, and income / expense information on the project, as well as information on comments on the project status.

[0070] For example, the information processing device 100 vectorizes information on comments on the project status, and classifies the comments into predetermined groups (clusters) (in the example of FIG. 10, BP (Business Partner) dissatisfaction, management dissatisfaction, etc.) by clustering the comments, and calculates the probability that the comment belongs to each group. Then, the information processing device 100 predicts the probability of development deficit occurrence using the project status questionnaire, working hour information, project balance information, and the clustering result of the comments on the project status as input.

[0071] In this way, the information processing device 100 of this embodiment inputs into a learning model one or more pieces of information, such as a questionnaire about the project status and information about working hours, information about the income and expenditures of the project, and the results of clustering comments about the project status, to predict the probability of a development deficit, thereby achieving the effect of reflecting the opinions of the personnel participating in the project in the prediction and making it possible to more accurately predict projects involving risk.

[0072] 2. Configuration of Information Processing Device Next, an example of the configuration of the information processing device 100 will be described with reference to Fig. 11. Fig. 11 is a diagram showing an example of the functional configuration of the information processing device 100. As shown in Fig. 11, the information processing device 100 has a communication unit 110, a control unit 120, and a storage unit 130. Note that each of these units may be held in a distributed manner by a plurality of devices. The processing of each of these units will be described below.

[0073] (Control unit 120) The control unit 120 is realized using a CPU, an NP, an FPGA, or the like, and executes a processing program stored in a memory. As shown in Fig. 11, the control unit 120 has a learning unit 121, an acquisition unit 122, a prediction unit 123, an output unit 124, and a calculation unit 125. Each unit of the control unit 120 will be described below.

[0074] (Learning Section 121) The learning unit 121 learns a learning model using at least one of a questionnaire on the project status and information on working hours, information on the income and expenditure of the project, comments on the project status, and information on the development deficit of the actual project as correct answer data. For example, the learning unit 121 learns the relationship between at least one of a questionnaire on the project status and information on working hours, information on the income and expenditure of the project, comments on the project status, and information on the development deficit of the actual project, and learns a learning model that outputs information on the development deficit of the project using at least one of a questionnaire on the project status and information on working hours, information on the income and expenditure of the project, and comments on the project status as input.

[0075] (Acquisition part 122) The acquiring unit 122 further acquires comments on the project status. For example, the acquiring unit 122 acquires information on working hours "overtime hours", information on the income and expenditure of the project "sales profit of the project, name of the client of the project, month of execution of the project, outsourcing cost of the project", and a comment on the project status "anxiety about the project".

[0076] (Calculation unit 125) The calculation unit 125 classifies the comments acquired by the acquisition unit 122 into a plurality of groups, and calculates the probability that each comment belongs to each group. For example, the calculation unit 125 classifies the comments acquired by the acquisition unit 122 into four groups, A, B, C, and D, and calculates the probability that each comment belongs to each of the groups A, B, C, and D.

[0077] Here, an example of the calculation process according to the embodiment will be described with reference to FIG. 12. FIG. 10 is a diagram for explaining an example of the calculation process according to the embodiment. The calculation unit 125 classifies comments on the status of a certain project into a plurality of groups, and calculates the probability that the comment belongs to each of the classified groups. For example, as shown in FIG. 12, the calculation unit 125 predicts that the probability that the comment "Most of the main functions of the business have been finalized..." belongs to each cluster (group) is "A: 0.1, B: 0.2, C: 0.88, D: 0.2, E: 0.1, F: 0.1, F: 0.1, G: 0.2, H: 0.2".

[0078] The calculation unit 125 can label each group by receiving an input from a user or an external device. For example, the user interprets the result of classification by the calculation unit 125, and receives an input of a correspondence between a group name and a label, such as "Group A: BP dissatisfaction" and "Group B: Management dissatisfaction," to label each group.

[0079] The calculation unit 125 also vectorizes the comments acquired by the acquisition unit 122 using the frequency of occurrence or distributed representation of words included in the comments, and classifies the vectorized comments into a plurality of groups by clustering. Here, an example of a method for vectorization using the frequency of occurrence of words included in comments is TF-IDF (Term Frequency-Inverse Document Frequency). Here, an example of a method for vectorization using distributed representation of words included in comments is Word2Vec (Word to Vector) or BERT (Bidirectional Encoder Representations from Transformers).

[0080] In addition, dimensionality reduction of vectorized comments is performed using methods such as PCA (Principal Component Analysis), t-SNE (t-Distributed Stochastic Neighbor Embedding), and UMAP (Uniform Manifold Approximation and Projection). Examples of clustering algorithms include the K-Means method and HDBSCAN (Hierarchical Density Based Spatial Clustering of Applications with Noise).

[0081] Here, an example of the calculation process according to the embodiment will be described with reference to FIG. 13. FIG. 13 is a diagram for explaining an example of the calculation process according to the embodiment. For example, the calculation unit 125 vectorizes the comments acquired by the acquisition unit 122 using BERT. Next, the calculation unit 125 performs dimensional compression on the vectorized comments using UMAP, and then clusters the comments using HBDSCAN to classify them into a plurality of groups. For example, the calculation unit 125 classifies the comments into 62 groups as shown in FIG. 13.

[0082] Then, the calculation unit 125 reduces the number of groups by classifying the comments classified into 62 groups into groups of 40, 20, and 10. Here, the number of groups into which the comments are classified can be changed arbitrarily depending on the purpose. Note that the vectorization method, dimensionality reduction method, and clustering method are not limited to those described above, and known methods can be combined and used depending on the characteristics of the data and the purpose.

[0083] (Prediction section 123) The prediction unit 123 receives as input one or more pieces of information on the project status acquired by the acquisition unit 122, information on the income and expenditures of the project, and the probability that the comment belongs to a specified group calculated by the calculation unit 125, and predicts information on the development deficit of the project using a learning model.

[0084] For example, the prediction unit 123 predicts information regarding the development deficit of the project by inputting one or more pieces of information regarding the project status and information about working hours acquired by the acquisition unit 122, information regarding the income and expenditure of the project, and the probability that the comment calculated by the calculation unit 125 belongs to a specified group into a learning model that outputs information regarding the development deficit of the project.

[0085] [3. Flowchart] Next, an example of the flow of processing by the information processing device 100 will be described with reference to Fig. 14. Fig. 14 is a flowchart showing an example of the flow of processing by the information processing device 100 according to the embodiment. Note that the following steps can be executed in a different order, and some processing may be omitted.

[0086] First, the acquisition unit 122 acquires at least one of a questionnaire about the project status and information about working hours, information about the income and expenditure of the project, and a comment about the project status (step S201). For example, the acquisition unit 122 acquires questionnaire information about the project status "health check", working hours information "overtime hours", information about the income and expenditure of the project "sales profit of the project, client name of the project, month of execution of the project, outsourcing cost of the project", and a comment about the project status "anxiety about the project".

[0087] Next, the calculation unit 125 vectorizes the comments acquired by the acquisition unit 122 (step S202). For example, the calculation unit 125 vectorizes the comments acquired by the acquisition unit 122 by Word2Vec.

[0088] Next, the calculation unit 125 performs dimensional compression on the vectorized comments (step S203). For example, the calculation unit 125 performs dimensional compression on the vectorized comments using UMAP. Next, the calculation unit 125 classifies the vectorized comments into a plurality of groups using clustering, and calculates the probability that each comment belongs to a predetermined group (step S204). For example, the calculation unit 125 performs clustering on the vectorized comments using HBDSCAN, classifies each comment into a plurality of groups, and calculates the probability that each comment belongs to a predetermined group.

[0089] Next, the calculation unit 125 labels each group classified as a result of the clustering (step S205). For example, the calculation unit 125 accepts the content of the user's interpretation of the clustering result and labels each group.

[0090] Next, the prediction unit 123 inputs one or more pieces of information on the project status acquired by the acquisition unit 122, information on the income and expenditure of the project, and the probability that the comment belongs to a specified group calculated by the calculation unit 125, and predicts information on a development deficit for the project using a learning model (step S206).

[0091] For example, the prediction unit 123 predicts information regarding the development deficit of the project by inputting one or more pieces of information regarding the project status and information about working hours acquired by the acquisition unit 122, information regarding the income and expenditure of the project, and the probability that the comment calculated by the calculation unit 125 belongs to a specified group into a learning model that outputs information regarding the development deficit of the project.

[0092] Next, the prediction unit 123 judges whether or not information on development deficits has been predicted for all projects (step S207). If the prediction unit 123 judges that information on development deficits has not been predicted for all projects (step S207: No), the process of step S206 is performed again.

[0093] On the other hand, if it is determined that the prediction unit 123 has predicted information on the development deficit for all projects (step S207: Yes), the output unit 124 outputs information on the development deficit of the projects predicted by the prediction unit 123 (step S208). For example, the output unit 124 outputs the unprofitability coefficient and unprofitability label of the projects for each project to the user's terminal device 200 as information on the development deficit of the projects predicted by the prediction unit 123.

[0094] Next, the output unit 124 determines whether or not a project selection from the user has been accepted (step S209). If the output unit 124 determines that a project selection from the user has not been accepted (step S209: No), the output unit 124 performs the process of step S209 again.

[0095] On the other hand, if the output unit 124 determines that it has accepted a project selection from the user (step S209: Yes), the output unit 124 outputs, for the project whose selection has been accepted, the results of a survey on the project status, a list of the working hours of those participating in the project, and information regarding the income and expenditure of the project, as well as information regarding the deficit predicted by the prediction unit 123 (step S210).

[0096] For example, for a project for which a selection has been accepted, the output unit 124 outputs to the user's terminal device 200 the negative response rate for each question as a result of a questionnaire on the project status, the average overtime hours of the personnel participating in the project and the number of people whose overtime hours exceeded 60 hours, and information on the project's periodic profit and loss, together with the unprofitability coefficient predicted by the prediction unit 123.

[0097] 4. Effects The acquisition unit 122 of the information processing device 100 according to the embodiment further acquires comments on the project status, classifies the comments acquired by the acquisition unit 122 into multiple groups, and further includes a calculation unit 125 that calculates the probability that the comment belongs to a specific group among the multiple groups. The prediction unit 123 predicts information regarding the development deficit of the project using a learning model, using as input one or more pieces of information on the questionnaire and working hours on the project status acquired by the acquisition unit 122, information regarding the income and expenditure of the project, and the probability that the comment calculated by the calculation unit 125 belongs to a specific group.

[0098] As a result, when predicting information regarding development deficits, the information processing device 100 can further use the quantified information of the characteristics of each comment as input to the model, thereby reflecting the opinions of the personnel participating in the project in the prediction and making more accurate predictions of projects that involve risk.

[0099] The calculation unit 125 of the information processing device 100 according to the embodiment vectorizes the comments acquired by the acquisition unit 122 using the frequency of occurrence or distributed representation of words contained in the comments, and classifies the vectorized comments into multiple groups by clustering.

[0100] As a result, when predicting information regarding development deficits, the information processing device 100 can more accurately predict projects at risk by further using as input the results of clustering comments that have been vectorized using a method using word frequency or distributed representation.

[0101] [others] [1. Hardware configuration] Next, a hardware configuration example of the information processing device 100 according to the embodiment will be described. FIG. 15 is a block diagram showing a hardware configuration example of the information processing device 100 according to the embodiment. Referring to FIG. 15, the information processing device 100 has, for example, a processor 801, a ROM 802, a RAM 803, a host bus 804, a bridge 805, an external bus 806, an interface 807, an input device 808, an output device 809, a storage 810, a drive 811, a connection port 812, and a communication device 813. Note that the hardware configuration shown here is an example, and some of the components may be omitted. In addition, components other than those shown here may be further included.

[0102] (Processor 801) The processor 801 functions, for example, as an arithmetic processing device or control device, and controls all or part of the operation of each component based on various programs recorded in the ROM 802, the RAM 803, the storage 810, or the removable recording medium 901.

[0103] (ROM802, RAM803) The ROM 802 is a means for storing the programs to be read into the processor 801, data to be used for calculations, etc. The RAM 803 temporarily or permanently stores, for example, the programs to be read into the processor 801, various parameters that change appropriately when the programs are executed, etc.

[0104] (host bus 804, bridge 805, external bus 806, interface 807) The processor 801, ROM 802, and RAM 803 are connected to each other via, for example, a host bus 804 capable of high-speed data transmission. On the other hand, the host bus 804 is connected to an external bus 806 having a relatively low data transmission speed via, for example, a bridge 805. In addition, the external bus 806 is connected to various components via an interface 807.

[0105] (Input device 808) Examples of the input device 808 include a mouse, a keyboard, a touch panel, a button, a switch, and a lever. Furthermore, a remote controller capable of transmitting control signals using infrared rays or other radio waves may be used as the input device 808. The input device 808 also includes an audio input device such as a microphone.

[0106] (Output Device 809) The output device 809 is a device capable of visually or audibly notifying the user of acquired information, such as a display device such as a CRT (Cathode Ray Tube), LCD, or organic EL, an audio output device such as a speaker or a headphone, a printer, a mobile phone, or a facsimile. The output device 809 according to this embodiment also includes various vibration devices capable of outputting tactile stimulation. The output device 809 may also include an AI speaker or a wearable device.

[0107] (Storage 810) The storage 810 is a device for storing various types of data. For example, a magnetic storage device such as a hard disk drive (HDD), a semiconductor storage device, an optical storage device, or a magneto-optical storage device may be used as the storage 810.

[0108] (Drive 811) The drive 811 is a device that reads information recorded on a removable recording medium 901 such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory, and writes information to the removable recording medium 901.

[0109] (Connection port 812) The connection port 812 is a port for connecting an external device 902, such as a Universal Serial Bus (USB) port, an IEEE1394 port, a Small Computer System Interface (SCSI), an RS-232C port, or an optical audio terminal.

[0110] (Communication device 813) The communication device 813 is a communication device for connecting to a network, such as a communication card for wired or wireless LAN, Bluetooth (registered trademark), or WUSB (Wireless USB), a router for optical communications, a router for ADSL (Asymmetric Digital Subscriber Line), or a modem for various types of communications.

[0111] (Removable recording media 901) The removable recording medium 901 is, for example, a DVD medium, a Blu-ray (registered trademark) medium, an HD DVD medium, various semiconductor storage media, etc. Of course, the removable recording medium 901 may be, for example, an IC card equipped with a non-contact type IC chip, or an electronic device.

[0112] (External connection device 902) The external connection device 902 is, for example, a printer, a portable music player, a digital camera, a digital video camera, or an IC recorder.

[0113] The storage unit 130 is realized by the ROM 802, the RAM 803, and the storage 810. The control unit 120 realized by the processor 801 reads out and executes each control program (for example, an information processing program according to the embodiment) that realizes the learning unit 121, the acquisition unit 122, the calculation unit 125, the prediction unit 123, and the output unit 124 from the ROM 802, the RAM 803, etc.

[0114] [2. Other] Among the above-mentioned processes, all or part of the processes described as being performed automatically may be performed manually. Also, all or part of the processes described as being performed manually may be performed automatically by a known method. In addition, the information including the processing procedures, specific names, various data and parameters shown in the above documents and drawings may be changed arbitrarily unless otherwise specified. For example, the various information shown in each drawing is not limited to the illustrated information.

[0115] In addition, each component of each device shown in the figure is a functional concept, and does not necessarily have to be physically configured as shown in the figure. In other words, the specific form of distribution and integration of each device is not limited to that shown in the figure. Furthermore, each component may be configured by distributing and integrating functionally or physically in any unit, in whole or in part, depending on various loads and usage conditions. Furthermore, each process described above may be executed in appropriate combination within a range that does not contradict.

[0116] The above describes the embodiments of the present application in detail with reference to several drawings. However, these are merely examples, and the present invention can be embodied in other forms that incorporate various modifications and improvements based on the knowledge of those skilled in the art, including the aspects described in the Disclosure of the Invention section. [Explanation of symbols]

[0117] 100 Information processing device 110 Communications Department 120 Control section 121 Learning Department 122 Acquisition Department 123 Prediction Department 124 Output section 125 Calculation Section 130 Storage section 131 Learning data storage unit 132 Project information storage unit 200 Terminal Equipment

Claims

1. an acquisition unit for acquiring a questionnaire on the project status indicating responses to a question about whether the project is causing problems for personnel participating in the project, and information regarding the income and expenditure of the project; a prediction unit that predicts a development deficit of the project by inputting the questionnaire regarding the project status acquired by the acquisition unit and information regarding the income and expenditure of the project into a learning model that has been trained to predict a development deficit of the project; an output unit that outputs information about the development deficit of the project predicted by the prediction unit; having The acquiring unit further acquires a plurality of comments on the project status indicating written responses to a question of whether the personnel participating in the project feel there are any problems with the project, a calculation section for classifying the plurality of comments acquired by the acquisition section into a plurality of groups related to anxieties felt in proceeding with the project, and calculating a probability that the plurality of comments belong to each of the plurality of groups; and The prediction unit is The questionnaire regarding the project status acquired by the acquisition unit, information regarding the income and expenditure of the project, and the probability of belonging to each of the plurality of groups calculated by the calculation unit are input to the learning model, and a development deficit of the project is predicted.

23. An information processing apparatus comprising:

2. A learning unit that learns the learning model using the questionnaire regarding the project status, information regarding the income and expenditure of the project, and information regarding the development deficit of the actual project as correct answer data.

2. The information processing apparatus according to claim 1, further comprising:

3. The prediction unit is a questionnaire regarding the status of the project and information regarding the income and expenditure of the project are inputted into the learning model to predict the probability of the project incurring a loss; The output unit outputs the deficit occurrence probability in order of the projects with the highest deficit occurrence probability predicted by the prediction unit.

2. The information processing apparatus according to claim 1,

4. The prediction unit is The learning model is input with at least one of the following information regarding the income and expenditure of the project: the sales profit of the project, the customer name of the project, the month of execution of the project, and the outsourcing cost of the project, to predict the development deficit of the project.

2. The information processing apparatus according to claim 1,

5. The acquisition unit periodically collects a questionnaire regarding the project status and information regarding the income and expenditure of the project, The output unit is The acquisition unit outputs the results of the questionnaire regarding the project status collected by the acquisition unit, and information regarding the income and expenditure of the project, as well as information regarding the development deficit of the project predicted by the prediction unit.

2. The information processing apparatus according to claim 1,

6. The calculation unit is The plurality of comments acquired by the acquisition unit are vectorized using an appearance frequency or a distributed representation of words included in the plurality of comments, and the vectorized plurality of comments are classified into the plurality of groups by clustering.

2. The information processing apparatus according to claim 1,

7. A method executed by an information processing device, comprising: an acquisition step of acquiring a questionnaire on the project status indicating the answers to a question about whether the personnel participating in the project feel there are any problems with the project, and information on the income and expenditure of the project; a prediction step of predicting a development deficit of the project by inputting the questionnaire regarding the project status acquired by the acquisition step and information regarding the income and expenditure of the project into a learning model trained to predict a development deficit of the project; an output step of outputting information regarding the development deficit of the project predicted by the prediction step; Including, The acquiring step further acquires a plurality of comments on the project status indicating written responses to a question of whether the personnel participating in the project feel there are any problems with the project, A calculation step of classifying the plurality of comments acquired by the acquisition step into a plurality of groups related to anxieties felt in proceeding with the project, and calculating the probability that the plurality of comments belong to each of the plurality of groups. Further comprising: The prediction step includes: The questionnaire regarding the project status acquired in the acquiring step, information regarding the income and expenditure of the project, and the probability of belonging to each of the plurality of groups calculated in the calculating step are input to the learning model, and a development deficit of the project is predicted.

23. An information processing method comprising:

8. an acquisition step of acquiring a questionnaire on the project status indicating the answers to a question about whether the personnel participating in the project feel there are any problems with the project, and information on the income and expenditure of the project; a prediction step of predicting a development deficit of the project by inputting the questionnaire regarding the project status acquired by the acquisition step and information regarding the income and expenditure of the project into a learning model trained to predict a development deficit of the project; an output step of outputting information regarding the development deficit of the project predicted by the prediction step; Run the following on your computer: The acquiring step further acquires a plurality of comments on the project status indicating written responses to a question of whether the personnel participating in the project feel there are any problems with the project, A calculation step of classifying the plurality of comments acquired by the acquisition step into a plurality of groups related to anxieties felt in proceeding with the project, and calculating the probability that the plurality of comments belong to each of the plurality of groups. Further executing the following on the computer, The prediction step comprises: The questionnaire regarding the project status acquired in the acquisition step, information regarding the income and expenditure of the project, and the probability of belonging to each of the plurality of groups calculated in the calculation step are input to the learning model, and a development deficit of the project is predicted.

2. An information processing program comprising:

Images