Information processing system, information processing method, and information processing program

By using a regression model through an information processing system to gradually eliminate candidate components and perform phased optimization, the problem of excessive computation time caused by the large number of candidate components is solved, and the effect of rapidly exploring the characteristic quantities of the composition is achieved.

CN119013678APending Publication Date: 2024-11-22RESONAC CORP
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
CN202380033328.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2022-02-18
Filing Date
2023-02-15
Publication Date
2024-11-22

AI Technical Summary

Technical Problem

When the number of candidate components is large, existing technologies require a significant amount of computation time to explore the characteristic quantities that constitute the desired composition.

Method used

The information processing system uses a regression model for optimization, gradually eliminating candidate components and repeatedly executing feature vector calculations to determine the physical properties of the composition, thus narrowing down the candidate components and performing optimization in stages.

Benefits of technology

It enables faster exploration of characteristic quantities of candidate components for constituting the desired composition, reducing computation time, especially when there are many candidate components.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119013678A_ABST
    Figure CN119013678A_ABST
Patent Text Reader

Abstract

An information processing system according to an embodiment includes at least one processor. At least one processor executes a process of executing an exclusion process of excluding at least one candidate component from a candidate component group based on a plurality of candidate components and repeating an optimization using a regression model that calculates a physical property value of a composition from a feature quantity vector representing a feature quantity of each of the candidate component groups, the feature amount of each of the one or more candidate components remaining after the repetition is determined. The exclusion process includes the steps of: selecting a feature quantity vector corresponding to a solution of the physical property value obtained by the optimization; and excluding at least one candidate component from the candidate component group according to the selected feature quantity vector.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] One aspect of the present disclosure relates to an information processing system, an information processing method, and an information processing program. Background Art

[0002] In the past, there are known methods for exploring the characteristic quantities of candidate components for forming a desired composition. For example, Patent Document 1 describes a method for preparing a blending plan for mixing a plurality of blending raw materials using a mathematical programming method. Patent Document 2 describes a method for planning the blending ratio of raw materials by solving a normalized raw material blending plan as a mixed integer programming problem containing nonlinear variables.

[0003] Previous technical literature

[0004] Patent Literature

[0005] Patent Document 1: Japanese Patent Application Publication No. 2009-175804

[0006] Patent Document 2: Japanese Patent Application Publication No. 2020-149607 Summary of the invention

[0007] Technical issues to be solved by the invention

[0008] In the methods described in Patent Documents 1 and 2, if the number of candidate components is large, a lot of computation time is required to obtain the feature values ​​of the candidate components for constituting a desired composition. Therefore, a method for more quickly searching for the feature values ​​is desired.

[0009] Means for solving technical problems

[0010] An information processing system according to one aspect of the present disclosure includes at least one processor. The at least one processor performs the following processing: performing an exclusion process of excluding at least one candidate component from a candidate component group based on a plurality of candidate components and repeatedly performing optimization, wherein the optimization uses a regression model that calculates a physical property value of a composition from a feature vector representing each feature value of the candidate component group, and determines each feature value of one or more candidate components remaining after the iteration. The exclusion process includes the following steps: selecting a feature vector corresponding to a solution of the physical property value obtained by optimization; and excluding at least one candidate component from the candidate component group based on the selected feature vector.

[0011] An information processing method according to one aspect of the present disclosure is performed by an information processing system having at least one processor. The information processing method includes the following steps: performing an exclusion process of excluding at least one candidate component from a candidate component group based on a plurality of candidate components and repeatedly performing optimization, wherein the optimization uses a regression model that calculates the physical property value of the composition from a feature vector representing each feature value of the candidate component group; and determining each feature value of one or more candidate components remaining after the iterative step. The exclusion process includes the following steps: selecting a feature vector corresponding to a solution of the physical property value obtained by optimization; and excluding at least one candidate component from the candidate component group based on the selected feature vector.

[0012] An information processing program according to one aspect of the present disclosure causes a computer to execute the following steps: performing an exclusion process of excluding at least one candidate component from a candidate component group based on a plurality of candidate components and repeatedly performing optimization using a regression model that calculates a physical property value of a composition from a feature quantity vector representing each feature quantity of the candidate component group; and determining each feature quantity of one or more candidate components remaining after the repeated step. The exclusion process includes the following steps: selecting a feature quantity vector corresponding to a solution of the physical property value obtained by the optimization; and excluding at least one candidate component from the candidate component group based on the selected feature quantity vector.

[0013] In such an aspect, the optimization of the regression model for calculating the physical property value of the composition from the feature vector of the candidate component group is performed repeatedly to reduce the types of candidate components. By narrowing down the candidate components and performing the optimization in stages, the feature values ​​of the candidate components for forming the desired composition can be quickly explored.

[0014] Effects of the Invention

[0015] According to an aspect of the present invention, it is possible to more quickly search for characteristic quantities of candidate components for constituting a desired composition. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a diagram showing an example of a functional configuration of an information processing system according to an embodiment.

[0017] Figure 2 This is a flowchart showing an example of the operation of the information processing system involved in the embodiment.

[0018] Figure 3 This is a flowchart showing an example of optimization. DETAILED DESCRIPTION

[0019] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. In the description of the drawings, the same or equivalent elements are denoted by the same reference numerals, and repeated descriptions are omitted.

[0020] [System Structure]

[0021] The information processing system 10 involved in the embodiment is a computer system for exploring the characteristic quantities of candidate components for constituting a desired composition. A composition refers to a substance constituted by processing one or more components. Typically, a composition is obtained using two or more components. For example, the composition may be an inorganic composition, an organic composition such as a resin composition, or a feed composition. The components of a composition refer to substances intentionally used to generate the composition. The components may or may not retain their original form when the composition is completed. In the present disclosure, components that can constitute a desired composition are also referred to as "candidate components". Individual candidate components are prepared corresponding to the composition, for example, various materials such as inorganic materials and organic materials.

[0022] In one example, the information processing system 10 repeatedly performs optimization using a pre-learned regression model, and ultimately determines the characteristic quantities of more than two candidate components that constitute the desired composition. In the present disclosure, "desired composition" refers to a composition having desired physical property values. As an example of the characteristic quantity of a candidate component, the amount of compounding can be cited. The amount of compounding can be expressed in absolute terms, or in terms of a ratio (combination ratio), or in terms of other indicators. Individual characteristic quantities can be expressed as a weighted average, or as a linear load and. For example, the amount of compounding can be expressed as a weighted average. As examples of the physical property values ​​of a composition, glass transition temperature, elastic modulus, and expansion coefficient can be cited.

[0023] A regression model refers to a computational model used to find the value of one or more response variables y based on the value of one or more explanatory variables x. In one example, the regression model accepts a feature vector based on the respective feature values ​​of a candidate component group based on multiple candidate components as an explanatory variable, and calculates the physical property value of the composition as a response variable. In the present disclosure, it should be noted that the term "multiple candidate components" refers to multiple types of candidate components. The candidate component group can also be a collection of multiple candidate components. The feature vector is an n-dimensional vector that represents the respective feature values ​​of n candidate components as vector elements and can be expressed in a one-dimensional arrangement.

[0024] The information processing system 10 changes at least a part of the feature vector in one optimization, and inputs each feature vector into the regression model to obtain the physical property value of the composition. The information processing system 10 selects a temporary solution from a plurality of physical property values, and selects a feature vector corresponding to the solution, that is, a feature vector input into the regression model in order to obtain the solution. The information processing system 10 performs an exclusion process to exclude at least one candidate component from the candidate component group based on the selected feature vector. Moreover, the information processing system 10 performs the optimization again on the premise of excluding some candidate components. The information processing system 10 performs the exclusion process and iterates the optimization, and outputs the respective feature values ​​of the one or more candidate components remaining after the repetition as the final processing result.

[0025] In this way, the information processing system 10 narrows down the candidate components and performs optimization in stages. As the number of candidate components considered in the optimization problem gradually decreases, the time required for individual optimization is shortened each time the optimization is repeated. Therefore, compared with the optimization that always considers all candidate components, the time required to obtain the final processing result can be shortened. This high-speed processing becomes significant in the case where there are many types of candidate components, such as material development.

[0026] The information processing system 10 is composed of one or more computers. When a plurality of computers are used, these computers are connected via a communication network such as the Internet or an intranet, thereby logically constructing one information processing system 10.

[0027] The computer constituting the information processing system 10 generally includes a processor, a memory, and a communication interface as hardware devices. The processor is, for example, a CPU, and the memory is composed of a flash memory, a hard disk, etc., but the types of hardware devices constituting the information processing system 10 are not limited thereto and can be selected arbitrarily. Each function of the information processing system 10 is realized by executing a program stored in the memory by the processor.

[0028] The information processing program for causing the computer to function as the information processing system 10 includes program codes for realizing each functional module of the information processing system 10. The information processing program can be provided on the basis of a tangible recording medium such as a CD-ROM, a DVD-ROM or a semiconductor memory in a non-temporary manner. Alternatively, the information processing program can also be provided via a communication network as a data signal superimposed on a carrier. The provided information processing program is stored in a memory, for example.

[0029] Figure 11 is a diagram showing an example of the functional structure of the information processing system 10. In this example, the information processing system 10 includes a processor 101. In one example, the processor 101 functions as an acquisition unit 11, an optimization unit 12, an exclusion unit 13, a determination unit 14, and an output unit 15. The acquisition unit 11 is a functional module for acquiring desired physical property values, i.e., reference physical property values, and a candidate component group. The optimization unit 12 is a functional module for performing optimization using a regression model. The exclusion unit 13 is a functional module for excluding at least one candidate component from the candidate component group. The determination unit 14 is a functional module for determining whether to end the iteration of optimization. The output unit 15 is a functional module for outputting the processing result.

[0030] In one example, the information processing system 10 accesses the candidate component database 21 via a given communication network. The communication network may be composed of at least one of the Internet and an intranet. The communication network may be composed of at least one of a wired network and a wireless network. The candidate component database 21 may be a component of the information processing system 10, or may be set in a computer system other than the information processing system 10.

[0031] The candidate component database 21 is a database that stores candidate component data representing individual candidate components. In one example, the candidate component data represents a plurality of available components.

[0032] [System Action]

[0033] refer to Figure 2 and Figure 3 , the operation of the information processing system 10 is described, and the information processing method involved in this embodiment is described. Figure 2 This is a flowchart showing an example of processing in the information processing system 10 . Figure 3 This is a flowchart showing an example of optimization processing.

[0034] In step S11, the acquisition unit 11 acquires reference property values. In one example, the acquisition unit 11 receives reference property values ​​input by a user. Alternatively, the acquisition unit 11 may read reference property values ​​from a pre-prepared property value database, or may receive reference property values ​​sent from another computer.

[0035] In step S12, the acquisition unit 11 acquires a candidate component group. In one example, the acquisition unit 11 accesses the candidate component database 21, reads out candidate component data representing a plurality of candidate components, and acquires the plurality of candidate components as a candidate component group. The acquisition unit 11 may acquire all candidate components recorded in the candidate component database 21 as a candidate component group, or may acquire at least two of the candidate components as a candidate component group. Alternatively, the acquisition unit 11 may accept a plurality of candidate components input by a user as a candidate component group without accessing the candidate component database 21. Alternatively, the acquisition unit 11 may receive a candidate component group sent from another computer.

[0036] In step S13, the optimization unit 12 optimizes the feature value of the candidate component using the regression model. "Optimization of the feature value of the candidate component" refers to a method of exploring one or more candidate components and the feature value of each candidate component to obtain a composition having a physical property value that is consistent with or close to a reference physical property value.

[0037] refer to Figure 3 , an example of optimization processing is described. In step S131, the optimization unit 12 generates a feature vector. In one example, the optimization unit 12 generates a feature vector representing the feature of each candidate component group. The dimension of the feature vector corresponds to the number of types of candidate components. That is, when the number of types of candidate components is n, the generated feature vector is an n-dimensional vector representing n feature as vector elements. The optimization unit 12 sets the feature of each candidate component using, for example, random search.

[0038] In step S132, the optimization unit 12 calculates the physical property values ​​of the composition from the feature vector using a regression model. In the present disclosure, the physical property values ​​calculated by the regression model are also referred to as "estimated physical property values". The optimization unit 12 inputs the feature vector into the regression model to calculate the estimated physical property values. The regression model used by the optimization unit 12 can be a support vector regression (Support Vector Regression (SVR)) or a Gaussian Process Regression (Gaussian Process Regression (GPR)). The optimization unit 12 can use a regression model stored in a given memory, or can access a regression model existing outside the information processing system 10 via a given communication network.

[0039] In step S133, the optimization unit 12 determines whether the termination condition of the calculation of the estimated physical property value is satisfied. For example, the termination condition can be set according to the number of executions of the regression model. The number of executions can be set by the user or can be set in advance.

[0040] If the end condition is not satisfied (No in step S133), the process returns to step S131, and the optimization unit 12 generates a new feature vector. The optimization unit 12 changes at least one vector element of the feature vector to generate a new feature vector. In order to generate a new feature vector, the optimization unit 12 can use a random search, a combination of a random search and a gradient method, or a combination of a random search and a Newton method. Alternatively, the optimization unit 12 can use a Bayesian optimization acquisition function to generate a new feature vector.

[0041] If the end condition is satisfied (Yes in step S133), the process proceeds to step S134, and the optimization unit 12 obtains at least one estimated property value corresponding to the reference property value among the plurality of estimated property values ​​as a solution. The "estimated property value corresponding to the reference property value" may be, for example, an estimated property value that is consistent with the reference property value, or an estimated property value that is most similar to the reference property value.

[0042] Return to Figure 2 . In step S14, the exclusion unit 13 excludes at least one candidate component from the candidate component group. That is, step S14 is an exclusion process. The exclusion unit 13 selects a feature vector corresponding to the estimated physical property value obtained as a solution in step S13. Next, the exclusion unit 13 applies at least one exclusion requirement to each feature represented by its feature vector, and excludes at least one candidate component that satisfies any exclusion requirement from the candidate component group. For example, the exclusion unit 13 may exclude a candidate component that satisfies an exclusion requirement that the feature is below a given threshold. Hereinafter, the exclusion requirement is also referred to as the "first exclusion requirement". Alternatively, the exclusion unit 13 may exclude a candidate component having the smallest feature in the feature vector. Hereinafter, the exclusion requirement is also referred to as the "second exclusion requirement". Alternatively, the exclusion unit 13 performs a process of excluding at least one candidate component using the first exclusion requirement, and if one candidate component is not excluded in the process, the second exclusion requirement may be used to exclude at least one candidate component. Alternatively, the exclusion unit 13 may exclude a candidate component having the largest feature in the feature vector. In one example, the exclusion unit 13 excludes the candidate component from the candidate component group by fixing the feature amount of the excluded candidate component in the feature amount vector to zero.

[0043] In step S15, the determination unit 14 determines whether the termination condition of the optimization iteration is satisfied. For example, the termination condition can be set according to the number of types of candidate components remaining in the candidate component group, that is, the number of candidate components whose feature quantities are not fixed to zero in the feature quantity vector. At this time, the determination unit 14 determines that the termination condition is satisfied when the number is below a given threshold. Alternatively, the termination condition may be that the optimization process and the elimination process are executed only a given number of times.

[0044] When the end condition is not satisfied (No in step S15), the process returns to step S13. In step S13, the optimization unit 12 performs the optimization process again, and in step S14, the elimination unit 13 performs the elimination process again.

[0045] In the repeated step S13, the feature quantity of at least a part of the feature quantity vector is fixed to zero. In step S131, in the state where the feature quantity of the excluded candidate component is fixed to zero, the optimization unit 12 sets the feature quantity of each of the remaining multiple candidate components by various methods such as random search, gradient method, Newton method, and acquisition function to generate a new feature quantity vector. At this time, the dimension of the generated feature quantity vector is essentially the number of excluded candidate components subtracted from the initial dimension of the feature quantity vector. For example, when the initial feature quantity vector is n-dimensional and 3 candidate components are excluded, the newly generated feature quantity vector is essentially (n-3)-dimensional. Therefore, the dimension of the newly generated feature quantity vector is substantially reduced during repeated optimization and exclusion processing.

[0046] When the end condition is met (yes in step S15), the process enters step S16, and the output unit 15 outputs the processing result. For example, the output unit 15 outputs the characteristic values ​​of the remaining one or more candidate components as the processing result. Alternatively, the output unit 15 can output the estimated physical property value together with the characteristic value as the processing result. For example, the output unit 15 can store the processing result in a given database, send it to other computers or computer systems, or display it on a display device. Alternatively, the output unit 15 can output the processing result to other functional modules for subsequent processing by the information processing system 10.

[0047] As described above, the information processing system 10 performs an exclusion process of excluding at least one candidate component from a candidate component group based on a plurality of candidate components and repeatedly performs optimization until the end condition in step S15 is satisfied, wherein the optimization uses a regression model that calculates the physical property value of the composition from a feature vector representing each feature of the candidate component group. In the case of exploring p or less candidate components constituting the desired composition from a candidate component group that is a set of n candidate components (where p < n), the information processing system 10 considers all n candidate components in the first optimization to find a temporary solution. It can be said that it is a process of exploring a relaxed solution by calculating a relaxed problem in which the constraints related to the feature vector are relaxed. Excluding a part of the candidate components by an exclusion process and fixing the corresponding part of the feature vector to zero means imposing constraints. That is, performing an exclusion process and repeatedly optimizing can be said to be a process of gradually increasing constraints and exploring the optimal solution.

[0048] [Effect]

[0049] As described above, an information processing system involved in one aspect of the present disclosure includes at least one processor. The at least one processor performs the following processing: performing an exclusion process of excluding at least one candidate component from a candidate component group based on a plurality of candidate components and repeatedly performing optimization, wherein the optimization uses a regression model that calculates the physical property value of the composition from the feature vector representing the respective feature values ​​of the candidate component group, and determines the respective feature values ​​of one or more candidate components remaining after the iteration. The exclusion process includes the following steps: selecting a feature vector corresponding to a solution of the physical property value obtained by optimization; and excluding at least one candidate component from the candidate component group based on the selected feature vector.

[0050] An information processing method according to one aspect of the present disclosure is performed by an information processing system having at least one processor. The information processing method includes the following steps: performing an exclusion process of excluding at least one candidate component from a candidate component group based on a plurality of candidate components and repeatedly performing optimization, wherein the optimization uses a regression model that calculates the physical property value of the composition from a feature vector representing each feature value of the candidate component group; and determining each feature value of one or more candidate components remaining after the iterative step. The exclusion process includes the following steps: selecting a feature vector corresponding to a solution of the physical property value obtained by optimization; and excluding at least one candidate component from the candidate component group based on the selected feature vector.

[0051] An information processing program according to one aspect of the present disclosure causes a computer to execute the following steps: performing an exclusion process of excluding at least one candidate component from a candidate component group based on a plurality of candidate components and repeatedly performing optimization using a regression model that calculates a physical property value of a composition from a feature quantity vector representing each feature quantity of the candidate component group; and determining each feature quantity of one or more candidate components remaining after the repeated step. The exclusion process includes the following steps: selecting a feature quantity vector corresponding to a solution of the physical property value obtained by the optimization; and excluding at least one candidate component from the candidate component group based on the selected feature quantity vector.

[0052] In such an aspect, the optimization of a regression model for calculating the physical property value of a composition from a feature vector of a candidate component group is performed repeatedly to reduce the types of candidate components. By narrowing down the candidate components and performing the optimization in stages, it is possible to quickly explore the feature values ​​of the candidate components used to form the desired composition. For example, even compared with metaheuristic methods such as Ant Colony Optimization (ACO) and traditional methods such as random search, the information processing system involved in the present disclosure can explore feature values ​​more quickly.

[0053] In an information processing system according to another aspect, optimization may include the steps of: for each of a plurality of feature vectors, inputting the feature vector into a regression model to calculate a physical property value of the composition as an estimated physical property value; and obtaining at least one of the plurality of estimated physical property values ​​as a solution. The solution is obtained from the plurality of estimated physical property values ​​obtained from the plurality of feature vectors, thereby improving the accuracy of optimization.

[0054] In the information processing system according to another aspect, at least one processor may perform the following processing: obtaining a reference property value used as a reference in optimization, and obtaining at least one estimated property value corresponding to the reference property value among a plurality of estimated property values ​​as a solution. By using the estimated property value corresponding to the reference property value as the solution for optimization, it is possible to more accurately explore the characteristic amount of the candidate component for constituting the desired composition.

[0055] In the information processing system according to another aspect, at least one processor can terminate iteration of optimization according to the number of candidate components remaining after elimination processing. By setting the termination condition of iteration according to the number of candidate components remaining, it is possible to explore the feature amount of the candidate components in consideration of the preparation of the candidate components for obtaining the desired composition, the production process of the composition, etc.

[0056] In the information processing system according to another aspect, at least one processor may exclude candidate components having feature quantities less than a threshold value in the step of excluding. By using the threshold value, candidate components can be effectively excluded.

[0057] In the information processing system according to another aspect, at least one processor may exclude the candidate component having the smallest feature amount in the step of excluding. By excluding the candidate component having the smallest feature amount, the excluding process can be reliably performed.

[0058] In the information processing system according to another aspect, the feature value may represent the blending amount of the candidate component for constituting the composition. By setting the feature value in this way, it is possible to search for the blending amount of each candidate component that should be prepared in order to constitute a desired composition.

[0059] [Modifications]

[0060] The above is a detailed description based on the embodiments of the present disclosure. However, the present disclosure is not limited to the above embodiments. The present disclosure can be modified in various ways without departing from the gist of the present disclosure.

[0061] For example, the information processing method executed by at least one processor is not limited to the above example. For example, part of the above steps or processes may be omitted, or the steps may be executed in another order. Furthermore, any two or more of the above steps may be combined, or part of the steps may be modified or deleted. Alternatively, in addition to the above steps, other steps may be executed.

[0062] When comparing the magnitude relationship of two numerical values ​​in the information processing system 10, either of the two benchmarks of "above" and "greater than" may be used, or either of the two benchmarks of "below" and "less than" may be used. The choice of such a benchmark does not change the technical significance of the process for comparing the magnitude relationship of two numerical values.

[0063] In the present disclosure, the expression "at least one processor executes a first process, executes a second process, ... executes an nth process." or an expression corresponding thereto indicates a concept including a case where the processor executing n processes from the first process to the nth process is changed in the middle. That is, the expression indicates a concept including both a case where all n processes are executed by the same processor and a case where the processor is changed according to an arbitrary policy among the n processes.

[0064] Explanation of symbols

[0065] 10- information processing system, 11- acquisition unit, 12- optimization unit, 13- elimination unit, 14- determination unit, 15- output unit, 21- candidate component database.

Claims

1. An information processing system comprising at least one processor, The at least one processor performs the following processing: performing an elimination process of eliminating at least one candidate component from a candidate component group based on a plurality of candidate components and repeatedly performing optimization using a regression model for calculating a physical property value of a composition from a feature quantity vector representing each feature quantity of the candidate component group, determining the feature value of each of the one or more candidate components remaining after the repetition, The exclusion process comprises the following steps: selecting the feature quantity vector corresponding to the solution of the physical property value obtained by the optimization; and The at least one candidate component is excluded from the candidate component group based on the selected feature vector.

2. The information processing system according to claim 1, wherein: The optimization comprises the following steps: For each of the plurality of feature vectors, the feature vector is input into the regression model to calculate the physical property value of the composition as an estimated physical property value; and At least one of the plurality of estimated physical property values ​​is acquired as a solution.

3. The information processing system according to claim 2, wherein: The at least one processor performs the following processing: obtaining a reference physical property value used as a reference in the optimization, The at least one estimated physical property value corresponding to the reference physical property value among the plurality of estimated physical property values ​​is acquired as the solution.

4. The information processing system according to any one of claims 1 to 3, wherein: The at least one processor ends the iteration of the optimization according to the number of types of the candidate components remaining after the elimination process.

5. The information processing system according to any one of claims 1 to 4, wherein: In the excluding step, the at least one processor excludes candidate components whose feature amount is equal to or less than a threshold value.

6. The information processing system according to any one of claims 1 to 5, wherein: The at least one processor excludes the candidate component having the smallest feature amount in the excluding step.

7. The information processing system according to any one of claims 1 to 6, wherein: The characteristic amount indicates the blending amount of the candidate component for constituting the composition.

8. An information processing method, which is performed by an information processing system having at least one processor, the information processing method comprising the following steps: performing an elimination process of eliminating at least one candidate component from a candidate component group based on a plurality of candidate components and repeatedly performing optimization using a regression model that calculates a physical property value of a composition from a feature quantity vector representing each feature quantity of the candidate component group; and determining the feature value of each of the one or more candidate components remaining after the iterative step, The exclusion process comprises the following steps: selecting the feature quantity vector corresponding to the solution of the physical property value obtained by the optimization; and The at least one candidate component is excluded from the candidate component group based on the selected feature vector.

9. An information processing program causing a computer to execute the following steps: performing an elimination process of eliminating at least one candidate component from a candidate component group based on a plurality of candidate components and repeatedly performing optimization using a regression model that calculates a physical property value of a composition from a feature quantity vector representing each feature quantity of the candidate component group; and determining the feature value of each of the one or more candidate components remaining after the iterative step, The exclusion process comprises the following steps: selecting the feature quantity vector corresponding to the solution of the physical property value obtained by the optimization; and The at least one candidate component is excluded from the candidate component group based on the selected feature vector.

Citation Information

Patent Citations

  • Combination selecting method and system

    CN105741183A

  • Polymer composition and process for production of the polymer composition

    CN109415448A

  • An information processing method, an information processing system, and a computer readable recording medium

    CN113342403A

  • Signal selection device, learning device, and signal selection method and program

    CN113646786A

  • Logistic regression trees for drug analysis

    EP1111533A2