JP2024043181A - Information processing device, information processing method and program - Google Patents

Abstract

[Problem] To enable accurate evaluation of predicted values of a prediction model. To provide an evaluation method applicable to learning models constructed by various machine learning algorithms, not limited to neural networks. [Solution] An information processing device includes an identification unit that identifies important explanatory variables from multiple explanatory variables included in learning data used to generate a prediction model, an extraction unit that extracts similar learning data similar to the prediction target data from the multiple learning data based on the values of the important explanatory variables of the learning data and the values of the important explanatory variables of the prediction target data, and an evaluation unit that evaluates the reliability of a predicted value output from a prediction model using the prediction target data as input, based on the similar learning data extracted by the extraction unit. [Selected Figure] Figure 3

Description

The present invention relates to an information processing device, an information processing method, and a program.

2. Description of the Related Art Conventionally, a technique is known in which a learning model constructed by machine learning using learning data is used to output a predicted value for new data. Patent Document 1 describes a technique for evaluating prediction results for new input cases by a trained neural network constructed using learning cases. In the technique described in Patent Document 1, similar cases are extracted from the learning cases based on the degree of similarity between the learning cases and the input cases, and the prediction results by the neural network are evaluated using the similar cases.

Japanese Patent Application Publication No. 2006-236367

However, in the technology described in Patent Document 1, all explanatory variables of input examples and learning examples are used to evaluate the prediction result, and the more important the explanatory variable is in the evaluation of the prediction result, the greater the contribution is made. However, if there are many unimportant explanatory variables, they may influence the evaluation. For this reason, with the technique described in Patent Document 1, there were cases in which prediction results could not be accurately evaluated.

The present invention has been made in view of these circumstances, and one of its exemplary objects is to provide an information processing device, an information processing method, and a program that make it possible to accurately evaluate predicted values of a prediction model. It's about doing.

In order to solve the above problem, an information processing device according to one aspect of the present invention includes an identification unit that identifies important explanatory variables from a plurality of explanatory variables included in the learning data used to generate a prediction model, an extraction unit that extracts similar learning data that is similar to the data to be predicted from the plurality of learning data based on the values of the important explanatory variables of the learning data and the values of the important explanatory variables of the data to be predicted, and an evaluation unit that evaluates the reliability of a predicted value output from the prediction model using the data to be predicted as input based on the similar learning data extracted by the extraction unit.

Yet another aspect of the present invention is an information processing method. This information processing method involves identifying important explanatory variables from multiple explanatory variables included in the training data used to generate the predictive model, and comparing the values of the important explanatory variables in the training data with the important explanatory variables in the data to be predicted. extracting similar learning data that is similar to the prediction target data from multiple learning data based on the value, and determining the reliability of the predicted value output from the prediction model using the prediction target data as input based on the similar learning data. including evaluating the

Yet another aspect of the present invention is a program. This program uses a computer to identify important explanatory variables from multiple explanatory variables included in the training data used to generate the prediction model, and to calculate the values of the important explanatory variables in the training data and the important explanatory variables of the data to be predicted. extracting similar learning data that is similar to the prediction target data from multiple training data based on the value of , and increasing the reliability of the predicted value output from the prediction model using the prediction target data as input based on the similar learning data. The objective is to evaluate the gender and carry out the following.

In addition, any combination of the above components, and any transformation of the present invention into a method, device, system, recording medium, computer program, etc., are also valid aspects of the present invention.

According to the present invention, it is possible to provide an information processing device, an information processing method, and a program that enable accurate evaluation of predicted values of a prediction model.

1 is a diagram showing a hardware configuration of an information processing device according to an embodiment of the present invention. FIG. 2 is a functional block diagram of the information processing device according to the embodiment. FIG. 2 is a functional block diagram of a processing unit according to the embodiment. FIG. 6 is a diagram for explaining an example of a method in which the extraction unit calculates the degree of similarity of learning data to prediction target data. 3 is a flowchart illustrating an example of an operation by an information processing apparatus according to an embodiment of the present invention. 3 is a flowchart illustrating an example of an operation by an information processing apparatus according to an embodiment of the present invention.

[Embodiment]
DESCRIPTION OF EMBODIMENTS Hereinafter, embodiments for carrying out the present invention will be described in detail with reference to the drawings. In addition, in the description of the drawings, the same elements are given the same reference numerals, and redundant description will be omitted as appropriate. Further, the configuration described below is an example and does not limit the scope of the present invention in any way.

FIG. 1 is a diagram showing a hardware configuration of an information processing device 1 according to an embodiment of the present invention. The information processing device 1 includes a processor 10, a storage device 12, an input device 14 that receives input operations, and an output device 16 that outputs information. The processor 10 includes a CPU (Central Processing Unit), a GPU (Graphical Processing Unit), and the like. The storage device 12 includes a memory, an HDD (Hard Disk Drive), an SSD (Solid State Drive), and the like. Input device 14 includes, for example, a keyboard, touch panel, mouse, microphone, and the like. Output device 16 includes, for example, a display, a touch panel, and a speaker.

Figure 2 is a functional block diagram of an information processing device 1 according to one embodiment of the present invention. The information processing device 1 according to this embodiment includes an input unit 20, a storage unit 22, a processing unit 24, and an output unit 26.

The input unit 20 receives various types of information and transmits the information to the processing unit 24. The input unit 20 is realized by the input device 14 included in the information processing device 1.

The storage unit 22 stores various information. The storage unit 22 may store, for example, programs for the processing unit 24 to execute various information processes, information regarding a prediction model, learning data used to generate the prediction model, prediction target data, and the like. The storage unit 22 is realized by the storage device 12 included in the information processing device 1.

The prediction model according to this embodiment is generated using training data based on various known machine learning algorithms. The prediction model may be composed of, for example, a neural network and a regression tree.

The learning data according to this embodiment includes a plurality of explanatory variables and a target variable. The explanatory variables according to this embodiment include variables related to the vehicle, and may indicate, for example, the type, model, color of the vehicle body, number of driven wheels (two-wheel drive or four-wheel drive), mileage, and years of use. Further, the objective variable according to this embodiment indicates the price of the vehicle. Note that the explanatory variables and objective variables of the learning data are not limited to variables related to the vehicle.

A prediction model generated using such training data takes the prediction target data as input and outputs the price of the vehicle as a predicted value. The prediction target data in this embodiment includes multiple explanatory variables similar to the training data. Therefore, the prediction model in this embodiment takes information about the vehicle as input and outputs the price of the vehicle as a predicted value.

The processing unit 24 executes various information processing and transmits the results to the storage unit 22 and the output unit 26. The functions of the processing unit 24 are realized by the processor 10 of the information processing device 1 executing a program stored in the storage device 12.

The output unit 26 outputs various information. The output unit 26 may display, for example, the result of information processing by the processing unit 24, and specifically, may display the evaluation result of the reliability of the predicted value output from the prediction model. The output unit 26 is realized by the output device 16 included in the information processing device 1.

FIG. 3 is a functional block diagram of the processing unit 24 according to this embodiment. The processing unit 24 according to this embodiment includes a generation processing unit 240, an evaluation processing unit 260, and a prediction processing unit 280.

The generation processing unit 240 generates a predictive model using the learning data, and determines the importance of explanatory variables forming the learning data based on the generation result. The functions of the generation processing section 240 are realized by the cooperation of the model generation section 242 and the determination section 244.

The model generation unit 242 generates a predictive model using learning data based on various known machine learning algorithms. Information regarding the generated prediction model is transmitted to the determination unit 244 and the prediction processing unit 280. Although the configuration of the prediction model generated by the model generation unit 242 is not particularly limited, in this embodiment, an example will be described in which the prediction model is configured with a regression tree.

The determining unit 244 determines the importance of each explanatory variable of the learning data. The determining unit 244 may determine the importance of each explanatory variable using the feature importance calculated based on various known machine learning algorithms. More specifically, the determination unit 244 may determine the importance of each explanatory variable based on the feature importance in the regression tree.

The determination unit 244 may determine the importance of the explanatory variable based on, for example, feature importance defined by CatBoost (registered trademark) or the like. For information on how to determine feature importance, please refer to the CatBoost URL below (Feature importance - Model analysis | CatBoost). Note that the feature value importance may be calculated by the model generation unit 242.
https://catboost.ai/en/docs/concepts/fstr#fstr__regular-feature-importance

When the learning data is divided by explanatory variables in the regression tree of the prediction model, the determination unit 244 determines the closeness of the target variable in the divided set of training data (more specifically, the Gini impurity based on the distribution of the target variable). The importance of the explanatory variable may be determined based on . Specifically, the determination unit 244 may determine the importance of the explanatory variable based on the difference in Gini impurity before and after the division.

For example, assume that the Gini impurity before and after dividing the learning data using a certain explanatory variable A is a difference A, and the Gini impurity before and after dividing the learning data using a certain explanatory variable B is a difference B. The determination unit 244 can determine the importance of the explanatory variable A and the explanatory variable B by comparing the difference A and the difference B. Specifically, the determining unit 244 may determine that the greater the amount of decrease in Gini impurity before and after dividing the learning data, the higher the importance of the corresponding explanatory variable.

It should be noted that for a certain explanatory variable, the learning data is divided according to the first condition regarding explanatory variable A, then the learning data is divided according to explanatory variable B, and then the learning data is further divided according to the second condition regarding explanatory variable A. The training data may be divided multiple times. In this case, regarding at least the explanatory variable A, the determination unit 244 considers the difference in Gini impurity when dividing the learning data under the first and second conditions.

Note that when the prediction model is configured by a neural network, the determination unit 244 may determine the importance of each explanatory variable based on the connection state of the neural network.

The evaluation processing unit 260 evaluates the reliability of the predicted value output from the prediction model based on the prediction model generated by the generation processing unit 240 and the determined importance of the explanatory variable. The functions of the evaluation processing section 260 are realized by the cooperation of the identification section 262, the extraction section 264, the counting section 266, the calculation section 268, and the evaluation section 270.

The identifying unit 262 identifies important explanatory variables from a plurality of explanatory variables included in the learning data used to generate the predictive model. Specifically, the identifying unit 262 may identify important explanatory variables based on the importance of the explanatory variables determined by the determining unit 244. For example, the identifying unit 262 may identify an explanatory variable having a degree of importance equal to or higher than a predetermined threshold as an important explanatory variable. The identification unit 262 transmits information regarding important explanatory variables to the extraction unit 264.

The extraction unit 264 extracts similar learning data similar to the prediction target data from the plurality of learning data based on the value of the important explanatory variable of the learning data and the value of the important explanatory variable of the prediction target data. Specifically, the extraction unit 264 calculates the similarity of the learning data to the prediction target data based on the value of the important explanatory variable of the learning data and the value of the important explanatory variable of the prediction target data, and uses the similarity. Similar learning data may be extracted using For example, if the similarity of the learning data exceeds a predetermined threshold, the extraction unit 264 may extract the learning data as similar learning data. The extraction unit 264 transmits information regarding the extracted similar learning data to the counting unit 266, the calculation unit 268, and the prediction processing unit 280.

An example of a method by which the extraction unit 264 calculates the similarity of learning data to prediction target data will be described with reference to FIG. 4. FIG. 4 shows the values of three important explanatory variables (X1, X2, and X3) for prediction target data and one piece of learning data. Here, the importance of X1, X2 and X3 is 0.7, 0.2 and 0.1.

As shown in FIG. 4, the values of the important explanatory variables X1, X2, and X3 of the data to be predicted are 0.7, 0.5, and 1, respectively. The values of the important explanatory variables X1, X2, and X3 of the training data are 0.2, 0.1, and 0, respectively. In this example, the similarity is calculated based on the values of these important explanatory variables by performing the following processes (1) to (3).

(1) Calculating the absolute value of the difference between the values of important explanatory variables First, the absolute value of the difference between the value of the important explanatory variable of the learning data and the value of the important explanatory variable of the prediction target data is calculated. As shown in FIG. 4, the calculation results are 0.5, 0.4, and 1 for the important explanatory variables X1, X2, and X3, respectively. In this example, we will explain an example in which similar training data is extracted using the absolute value of the difference between the value of the important explanatory variable of the training data and the value of the important explanatory variable of the prediction target data. The product (also referred to as "cosine similarity") of the value of the explanatory variable and the value of the important explanatory variable of the prediction target data may be calculated, and similar learning data may be extracted based on this product.

In addition, if the important explanatory variable is a quantitative variable, before calculating the absolute value of the difference between the values of the important explanatory variable, the minimum value of the important explanatory variable in all training data becomes 0, and Min-max normalization may be performed so that the maximum value of the explanatory variable is 1.

Further, when the important explanatory variable is a qualitative variable, the value of the important explanatory variable may be set to 1 when a predetermined condition is satisfied, and the value of the important explanatory variable may be set to 0 when the predetermined condition is not satisfied. For example, if the predetermined condition is that the color of the vehicle is black, when the color of the vehicle indicated by the value of the important explanatory variable is black, the value of the important explanatory variable is set to 1, and the value of the important explanatory variable is set to 1. When the color of the vehicle shown is a color other than black, the value of the important explanatory variable may be set to 0.

(2) Multiplying the absolute values and the importance The absolute values of the differences between the values of the important explanatory variables calculated in (1) above are multiplied by the importance of the important explanatory variables. As shown in Figure 4, the multiplication results are 0.35, 0.08, and 0.1 for the important explanatory variables X1, X2, and X3, respectively.

(3) Calculating the degree of similarity The degree of similarity of the learning data is calculated using the sum of the values obtained by integrating in (2) above. Specifically, 0.47, which is the value obtained by subtracting the total integrated value (0.35+0.08+0.1=0.53) from 1, is calculated as the similarity of the learning data. Similarly, the extraction unit 264 can calculate the degree of similarity for all of the learning data, and similar learning data is extracted from all of the learning data based on the degree of similarity thus calculated.

The method for calculating the similarity of learning data is not limited to the example described here. When the important explanatory variable is a quantitative variable, the extraction unit 264 may calculate the degree of similarity based on the distance between the learning data and the prediction target data. Specifically, the extraction unit 264 may calculate the degree of similarity based on the square root of the sum of squares using the differences in the values of the important explanatory variables. Furthermore, when the important explanatory variables are qualitative variables, the extraction unit 264 may calculate the degree of similarity based on the number of matching important explanatory variables between the learning data and the prediction target data. For example, the extraction unit 264 determines the similarity so that the greater the number of matching important explanatory variables, or the greater the ratio of the number of matching important explanatory variables to the total number of important explanatory variables, the greater the similarity. You can calculate it.

Returning to FIG. 3, the functions of the evaluation processing section 260 will be explained. The counting unit 266 of the evaluation processing unit 260 counts the number of similar learning data extracted by the extraction unit 264 and transmits the counting result to the evaluation unit 270. Further, the calculation unit 268 calculates the standard deviation of the target variable of the plurality of similar learning data extracted by the extraction unit 264, and transmits the calculated standard deviation to the evaluation unit 270.

The evaluation unit 270 evaluates the reliability of the predicted value output from the prediction model when the prediction target data is input, based on the similar learning data extracted by the extraction unit 264. The evaluation unit 270 may transmit the evaluation results to the output unit 26. Thereby, the evaluation result is displayed by the output unit 26.

In this embodiment, the evaluation unit 270 evaluates the reliability of the predicted value based on the number of similar learning data counted by the counting unit 266. Specifically, the evaluation unit 270 may evaluate the reliability of the predicted value higher as the number of similar learning data increases. Alternatively, the evaluation unit 270 may evaluate the reliability of the predicted value higher as the ratio of the number of similar learning data to the number of all similar learning data is larger. It is considered that the more similar learning data is used to generate a predictive model, the higher the reliability of the predicted value output from the predictive model. Therefore, by using the number of similar learning data for evaluation, it is possible to evaluate the reliability of predicted values with higher accuracy.

Furthermore, the evaluation unit 270 evaluates the reliability of the predicted value based on the standard deviation of the target variable of the similar learning data calculated by the calculation unit 268. Specifically, the evaluation unit 270 may evaluate the reliability of the predicted value higher as the standard deviation of the target variable of the similar learning data is smaller. On the other hand, the evaluation unit 270 may evaluate the reliability of the predicted value lower as the standard deviation of the target variable of the similar learning data is larger. The evaluation unit 270 may calculate the standard deviation or a value based on the standard deviation as the reliability.

The smaller the standard deviation of the objective variable, the smaller the variation in the values of the objective variable. A prediction model generated using similar learning data with target variable values with small variations can be expected to output highly accurate predicted values. On the other hand, the accuracy of predicted values output by a predictive model generated using similar learning data with target variable values with large variations is considered to be low. Therefore, by using the standard deviation of the target variable of similar learning data for evaluation, it is possible to evaluate the reliability of predicted values with higher accuracy.

The prediction processing unit 280 generates a predicted value using the prediction model generated by the model generation unit 242, and corrects the predicted value. The functions of the prediction processing section 280 are realized by the cooperation of the prediction generation section 282 and the correction section 284.

The prediction generation unit 282 generates a predicted value for the prediction target data using the prediction model, and transmits the predicted value to the correction unit 284. Specifically, the prediction generation unit 282 inputs the explanatory variables of the prediction target data into the prediction model and obtains the predicted value.

The correction unit 284 corrects the predicted value generated by the prediction generation unit 282 and transmits the corrected result to the output unit 26. Thereby, the predicted value is displayed on the output section 26. Specifically, the correction unit 284 may correct the predicted value based on the plurality of similar learning data extracted by the extraction unit 264. By correcting the predicted value using similar learning data, it becomes possible to correct the predicted value more accurately.

Even if the vehicle has the same make, model, mileage, and other conditions, the value of the vehicle often changes from year to year. Therefore, when there is a plurality of similar learning data, the values of these objective variables (vehicle prices) may vary greatly depending on the year corresponding to each similar learning data. In this embodiment, the correction unit 284 corrects the predicted value in consideration of such changes in the value of the target variable over time.

In this embodiment, each of the multiple learning data is associated with time information. The time information may be, for example, information indicating the year and date when the price of the vehicle corresponding to the value of the objective variable was determined by appraisal and when the vehicle was sold at the price corresponding to the value of the objective variable. The correction unit 284 can correct the predicted value based on the relationship between the value of the objective variable and the time information.

For example, suppose the value of the objective variable in the learning data tends to increase over time. In this case, the price of a vehicle that includes conditions similar to the values of the explanatory variables in the learning data based on past information is presumed to be higher than the price of the vehicle predicted based on the learning data. However, because the prediction model is generated using the learning data, the time dependency of the objective variable values is not reflected in the prediction model. For this reason, when the data to be predicted is input into the prediction model, there is a possibility that a price lower than the accurate vehicle price will be output as the predicted value.

Therefore, the correction unit 284 may correct the predicted value output from the prediction model based on the time dependence of the value of the objective variable. Specifically, if the value of the target variable is increasing over time, a price depending on the rate of increase in the value of the target variable may be added to the predicted value. Furthermore, if the value of the target variable is decreasing over time, a price corresponding to the rate of decline in the value of the target variable may be subtracted from the predicted value. By correcting the predicted value using the time dependence of the value of the objective variable in this way, it is possible to more accurately correct the predicted value when the objective variable is something whose price fluctuates from year to year, such as the price of a vehicle. It becomes possible to do so.

FIG. 5 is a flowchart illustrating an example of the operation of the information processing device 1 according to an embodiment of the present invention. The flow of operations according to this embodiment will be described below along the flowchart shown in FIG.

First, the model generation unit 242 generates a predictive model using a plurality of learning data (S101). At this time, the feature value importance may be calculated along with the generation of the predictive model. Next, the determining unit 244 determines the importance of the explanatory variables of the learning data (S103). Here, the determination unit 244 determines the importance of a plurality of explanatory variables included in the learning data used to generate the predictive model in S101. Next, the identifying unit 262 identifies important explanatory variables based on the importance of the explanatory variables determined in S103 (S105).

Next, the extraction unit 264 extracts similar learning data from the plurality of learning data used to generate the prediction model in S101, based on the value of the important explanatory variable of the learning data and the value of the important explanatory variable of the prediction target data. Extract (S107). The important explanatory variables used here are the important explanatory variables specified in S105. Next, the counting unit 266 counts the number of similar learning data extracted in S107 (S109). Next, the calculation unit 268 calculates the standard deviation of the target variable of the similar learning data extracted in S107 (S111).

Next, the evaluation unit 270 evaluates the predicted value output from the prediction model when the prediction data is input, based on the similar learning data (S113). Here, the evaluation unit 270 evaluates the reliability of the predicted value based on the number of similar learning data counted in S109 and the standard deviation of the target variable calculated in S111. When the prediction unit 252 evaluates the reliability of the predicted value, the evaluation process shown in FIG. 5 ends.

FIG. 6 is a flowchart illustrating an example of the operation of the information processing device 1 according to an embodiment of the present invention. The flow of the evaluation process according to this embodiment will be described below along the flowchart shown in FIG.

First, the processes of S201 to S207 are carried out, but since these processes are substantially the same as the processes of S101 to S107 described with reference to FIG. 5, they will not be described here. When similar learning data is extracted in S207, the prediction generation unit 282 uses a prediction model to generate a predicted value by inputting the prediction target data (S209). Next, the correction unit 284 corrects the predicted value generated in S209 based on the relationship between the value of the objective variable of the similar learning data extracted in S207 and the time information (S211). When the correction unit 284 corrects the predicted value, the correction process shown in FIG. 6 ends.

The configuration and operation of the information processing device 1 according to an embodiment of the present invention have been described above. According to the information processing device 1 according to the present embodiment, similar learning data similar to prediction target data is extracted based on an important explanatory variable identified from a plurality of explanatory variables, and a prediction model is output using the similar learning data. The reliability of predicted values can be evaluated. Therefore, even if the learning data includes unimportant explanatory variables, such explanatory variables are not used to evaluate the reliability of the predicted value, so the predicted value of the predictive model can be evaluated with high accuracy.

Further, according to the information processing device 1 according to the present embodiment, it is possible to evaluate the reliability of predicted values of predictive models constructed not only by neural networks but also by various models such as regression trees. Therefore, the information processing device 1 according to the present embodiment can provide a highly versatile method for evaluating reliability of predicted values.

Further, according to the information processing device 1 according to the present embodiment, the evaluation unit 270 evaluates the reliability of the predicted value based on the number of similar learning data. Therefore, the evaluation unit 270 can evaluate that the more similar learning data there is, the higher the reliability is, and it becomes possible to evaluate the reliability with higher accuracy.

Further, according to the information processing device 1 according to the present embodiment, the evaluation unit 270 evaluates the reliability of the predicted value based on the standard deviation of the target variable of the similar learning data. Therefore, the evaluation unit 270 can evaluate that the reliability of the predicted value is higher as the variation in the target variable of the similar learning data is smaller, and it is possible to evaluate the reliability with higher accuracy.

The information processing device 1 according to the present embodiment can also be used to evaluate the reliability of the predicted value of a prediction model for predicting the price at which a used vehicle will be sold at auction. The sales price of a used vehicle is generally assessed by an appraiser based on his or her own sense, experience, and price experience. According to the present embodiment, by using the prediction model generated by the information processing device 1, it is possible to predict the price of a vehicle without relying on the sense of an appraiser, and to evaluate the reliability of the predicted value. is possible.

[supplement]
The present invention has been described above based on the embodiments. Those skilled in the art will understand that this embodiment is merely an example, and that various modifications can be made to the combinations of the constituent elements and processing processes, and that such modifications are also within the scope of the present invention. be.

The processing of each step by the information processing apparatus 1 described with reference to the above flowchart does not necessarily have to be executed in the illustrated order. As long as there is no logical contradiction, the processing of the steps may be changed as appropriate, or the processing of a plurality of steps may be executed in parallel. For example, the processes of S109 (counting the number of similar learning data) and S111 (calculating the standard deviation of the objective variable of similar learning data) may be performed in parallel.

1 information processing device, 240 generation processing unit, 242 model generation unit, 244 determination unit, 252 prediction unit, 260 evaluation processing unit, 262 identification unit, 264 extraction unit, 266 count unit, 268 calculation unit, 270 evaluation unit, 282 prediction Generation unit, 284 Correction unit, 30 Regression tree.

Claims (8)

An identification unit that identifies important explanatory variables from a plurality of explanatory variables included in the learning data used to generate the prediction model;
an extraction unit that extracts similar learning data similar to the prediction target data from a plurality of the learning data based on values of important explanatory variables of the learning data and values of important explanatory variables of the prediction target data;
and an evaluation unit that evaluates the reliability of a predicted value output from the prediction model using the prediction target data as an input based on the similar learning data extracted by the extraction unit.
Information processing device. The evaluation unit evaluates reliability of the predicted value based on the number of similar learning data extracted by the extraction unit.
The information processing device according to claim 1 . The extraction unit extracts a plurality of similar learning data that are respectively similar to the prediction target data from the plurality of learning data,
The evaluation unit evaluates the reliability of the predicted value based on the standard deviation of the objective variable of the plurality of similar learning data.
The information processing device according to claim 1. The explanatory variables include variables related to vehicles,
the objective variable of the learning data indicates the price of the vehicle;
The information processing device according to claim 1. The prediction model includes a regression tree,
The identifying unit identifies the important explanatory variable based on the feature importance in the regression tree.
The information processing device according to claim 4. a prediction generation unit that uses the prediction model to generate the predicted value by inputting the prediction target data;
further comprising a correction unit that corrects the predicted value generated by the prediction generation unit based on the similar learning data,
The extraction unit extracts a plurality of similar learning data that are respectively similar to the prediction target data from the plurality of learning data,
Each of the plurality of similar learning data is associated with time information,
The correction unit corrects the predicted value based on the relationship between the value of the objective variable and the time information.
The information processing device according to claim 4 or 5. Identifying important explanatory variables from multiple explanatory variables included in the training data used to generate the predictive model;
Extracting similar learning data similar to the prediction target data from a plurality of the learning data based on the value of the important explanatory variable of the learning data and the value of the important explanatory variable of the prediction target data;
Evaluating the reliability of a predicted value output from the prediction model using the prediction target data as input, based on the similar learning data,
Information processing method. to the computer,
Identifying important explanatory variables from multiple explanatory variables included in the training data used to generate the predictive model;
Extracting similar learning data similar to the prediction target data from a plurality of the learning data based on the value of the important explanatory variable of the learning data and the value of the important explanatory variable of the prediction target data;
Evaluating the reliability of a predicted value output from the prediction model using the prediction target data as input, based on the similar learning data;
A program to run.

Images