Diversity concentration device, diversity concentration method, and diversity concentration program

The diversity enrichment device and method address the challenge of maintaining data diversity by selectively removing data that minimally contributes to it, ensuring adaptive and efficient data retention.

JP7585447B1Active Publication Date: 2024-11-18早坂壮大
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
JP2023212621
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Filing Date
2023-12-18
Publication Date
2024-11-18
Estimated Expiration
2043-12-18

AI Technical Summary

Technical Problem

Existing data management techniques struggle to maintain data diversity efficiently, as they often require fixed retention intervals and multiple parameters, making it difficult to adapt to changes in operational status, leading to loss of data diversity.

Method used

A diversity enrichment device and method that identifies data to be excluded based on a specified distribution, either agreeing or disagreeing with an ideal distribution, to maintain data diversity by selectively removing data that contributes least to it.

Benefits of technology

Effectively maintains data diversity by identifying and removing data that minimally contributes to it, allowing for adaptive retention strategies that preserve data age and operational relevance.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007585447000001_ABST
    Figure 0007585447000001_ABST
Patent Text Reader

Abstract

Provided are a diversity concentration device, method, and program that can suitably identify data to be excluded from a group of data and efficiently maintain the diversity of the group of data. [Solution] The system includes a means for identifying data to be excluded from a group of data, which is a collection of data including attribute values, so that the distribution of attribute values ​​in the group of data after the data to be excluded is excluded approximates a specific distribution, and an output means for outputting information indicating the data to be excluded identified by the identification means.
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to a technique for updating a data set. [Background technology]

[0002] When adding new data to a group of mutually related data, such as a group of data obtained through experiments or observation, or a group of backup data obtained by modifying the original data, some of the data may be deleted (i.e., thinned out) to avoid putting pressure on the storage capacity of the storage medium.

[0003] One technique that can be used to delete some data from a set of data is to delete data in ascending order of time of generation. Such techniques are based on the assumption that older data is generally less useful. However, when such techniques are applied, data from before a certain point in time is lost, which reduces the diversity of the set of data as a whole, such as the age of the data.

[0004] The technology of deleting new data while retaining a certain percentage of old data is based on the assumption that old data still has a certain degree of usefulness. When applying this type of technology to update a group of data, one of the important points is to thin out the data without losing as much diversity as possible, such as the age of the data, in the group as a whole.

[0005] As a related technique, Non-Patent Document 1 describes a technique for retaining a group of data so that the data ages are at non-equidistant intervals. Specifically, page 14 of Non-Patent Document 1 describes a technique for setting the intervals for retaining file versions for generations saved by backup from 1 day ago to 7 days ago, 8 days ago to 90 days ago, 91 days ago to 365 days ago, and 366 days or later. [Prior art documents] [Non-patent literature]

[0006] [Non-Patent Document 1] “Getting Started Welcome to BackStore by PROe! Operation Manual Ver 5.0.1”, [online], Nekojarashi Co., Ltd., [Searched on September 28, 2023], Internet<URL:https: / / www.backstore.jp / dl / BackStore_GettingStarted_PROe.pdf> Summary of the Invention [Problem to be solved by the invention]

[0007] However, in the technology described in Non-Patent Document 1, the interval for retaining file versions can only be set for a fixed period determined in advance. In other words, in the technology described in Non-Patent Document 1, the period for which the interval for retaining file versions can be set is limited. In addition, in the technology described in Non-Patent Document 1, multiple parameter settings must be performed in advance, making it difficult to optimize the retention interval in response to changes in operational conditions over time. Therefore, in the technology described in Non-Patent Document 1, it is difficult to efficiently maintain the diversity of a group of data.

[0008] The present invention aims to provide a diversity concentration device, a diversity concentration method, and a diversity concentration program that can preferably identify data to be excluded from a group of data and efficiently maintain the diversity of the group of data. [Means for solving the problem]

[0009] The diversity concentration device according to the present invention is a device for concentrating diversity by extracting a set of data, which is a set of data including attribute values. When one of the data is excluded Distribution of attribute values Of the given distributions where specific distribution The data is removed by the specified distribution that is most consistent with the specified distribution or the specified distribution that is least consistent with the specified distribution. Exclusion targets As The data processing device is characterized in that it includes a specifying means for specifying the data to be excluded, and an output means for outputting information indicating the data to be excluded that is specified by the specifying means.

[0010] The diversity enrichment method according to the present invention is a method for enriching diversity by a computer. When one of the data is excluded Distribution of attribute values Of the given distributions where specific distribution The data is removed by the specified distribution that is most consistent with the specified distribution or the specified distribution that is least consistent with the specified distribution. Exclusion targets As The method is characterized in that it identifies the data to be excluded and outputs information indicating the identified data to be excluded.

[0011] The diversity enrichment program according to the present invention is a program for causing a computer to extract a set of data, which is a set of data including attribute values, from the data. When one of the data is excluded Distribution of attribute values Of the given distributions where specific distribution The data is removed by the specified distribution that is most consistent with the specified distribution or the specified distribution that is least consistent with the specified distribution. Exclusion targets As The present invention is characterized in that it executes a specification process for specifying data to be excluded, and an output process for outputting information indicating the data to be excluded specified in the specification process. Effect of the Invention

[0012] According to the present invention, data that contributes little to diversity can be suitably identified as data to be removed from a group of data, and the diversity of the group of data can be efficiently maintained. [Brief description of the drawings]

[0013] [Figure 1] FIG. 1 is a block diagram showing an example of a system to which a diversity concentration device is applied. [Diagram 2] FIG. 1 is an explanatory diagram showing an example of an ideal distribution composed of a proportional phase and an exponential phase. [Diagram 3] FIG. 11 is an explanatory diagram showing an example of identifying data to be excluded; [Figure 4] 13 is a flowchart showing an example of a processing operation of a diversity concentration device. [Diagram 5] FIG. 13 is a diagram showing a comparative example between a generation retained when a backup method in which data is deleted from old data is applied and a generation retained when a diversity concentration method is applied. [Figure 6] FIG. 11 is a diagram showing a comparative example of the transition of the amount of information retained when a backup method that deletes old data is applied and the transition of the amount of information retained when a diversity concentration method is applied. [Figure 7] FIG. 13 is a diagram showing a comparative example between the distribution of data held when a backup method that deletes old data is applied and the distribution of data held when a diversity concentration method is applied. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0014] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0015] Fig. 1 is a block diagram showing an example of a system to which a diversity concentration device is applied. The diversity concentration system to which the diversity concentration device illustrated in Fig. 1 is applied includes a storage unit 1, a monitoring means 2, an extraction means 3, an exclusion means 4, a diversity concentration device 10, and a user terminal 20. The diversity concentration device 10 includes an input means 5, an exclusion target identification means 6, and an output means 7. The exclusion target identification means 6 includes an ideal distribution identification means 61, and a post-change distribution evaluation means 62. Note that the arrows in Fig. 1 simply indicate the direction of signal (data) flow, but do not exclude bidirectionality.

[0016] The storage unit 1 is realized by, for example, a storage device such as a semiconductor memory or a hard disk. The functions of the monitoring means 2, the extraction means 3, and the elimination means 4 are realized, for example, by a central processing unit (CPU) of an information processing device executing processes according to a program stored in a storage medium. The input means 5, the elimination target identification means 6 (including an ideal distribution identification means 61 and a post-change distribution evaluation means), and the output means are realized, for example, by a CPU of the diversity concentration device 10 executing processes according to a program stored in a storage medium.

[0017] The monitoring means 2, the extraction means 3, and the exclusion means 4 may be included in, for example, the same information processing device, or may be included in different information processing devices. Also, the monitoring means 2, the extraction means 3, and the exclusion means 4 may be included in, for example, an information processing device (not shown) including a storage unit 1, or may be included in a diversity concentration device 10. Also, the storage unit 1, the monitoring means 2, the extraction means 3, and the exclusion means 4 may be included in the diversity concentration device 10.

[0018] The storage unit 1 stores a group of data to be managed. The storage unit 1 may store various data related to the group of data. For example, an information processing device (not shown) in the diversity enrichment system or outside the diversity enrichment system can store new data as the group of data in the storage unit 1. Also, for example, an information processing device (not shown) in the diversity enrichment system or outside the diversity enrichment system can update the data stored in the storage unit 1 as the group of data.

[0019] The group of data to be managed is a collection of data each including one or more attribute values. The attribute values ​​may have one or more inherent reference values ​​and may be characterized in that the attribute values ​​can be compared to the reference values.

[0020] A group of data may be a collection of data itself, or a collection of metadata indicating the attributes, characteristics, structure, meaning, relationships, etc. of the data. A group of data may also be a collection of data indicating the attributes, characteristics, structure, meaning, relationships, etc. of objects or events that are not expressed by data (hereinafter also referred to as catalog data). A group of data may also include a mixture of data in different formats, such as text data, spreadsheet data, video data, and audio data.

[0021] In this embodiment, the expression "data including attribute values" is used, but the attribute values ​​included in the data may indicate attributes of the data itself or may indicate attributes of other data. Also, the attribute values ​​included in the data may indicate attributes of objects, events, etc. that are not expressed by data.

[0022] The attribute value included in the data may be, for example, a value indicating a date and time associated with the data. Specifically, the attribute value included in the data may be determined based on the date and time when the data was accessed or the date and time when the data was generated.

[0023] In this embodiment, each data included in the group of data to be managed is stored in association with order information (hereinafter also referred to as rank) for identifying the order in which the data was stored in the group of data. Note that, without being limited to the configuration of this embodiment, for example, the order information corresponding to data may be configured to be generated based on the attribute value of the data, as necessary.

[0024] Next, a specific example of a group of data to be managed will be described. For example, the group of data D is composed of c text files D1 to Dc. Each data (i.e., text file) in the group of data D has information indicating the time T at which the file was last accessed as an attribute value. The group of data D is a collection of document data generated each time a specific document data is modified. When each text file is plotted on a one-dimensional time axis based on the attribute value (time T), the distribution state of each text file (i.e., attribute value (time T)) is related to the diversity of the last access times of the group of data D. For example, when the distribution state of the text files (i.e., attribute value (time T)) is close to the ideal distribution state, the group of data D is determined to have high diversity. Conversely, when the distribution state of the text files (i.e., attribute value (time T)) is far from the ideal distribution state, the group of data D is determined to have low diversity.

[0025] In the group of data D, the text file D1 is the most recently accessed (i.e., the newest) text file. On the other hand, the text file Dc is the oldest text file. In this case, the text file D1 is associated with rank 1, which is a rank value of 1, as the order information. Also, the text file Dc is associated with rank c, which is a rank value of c, as the order information. That is, in this embodiment, the larger the rank value, the older the data is, and the longer the period of time that the data has been stored as a group of data is indicated. Also, in this embodiment, for example, when new data is stored in the group of data, the new data is associated with rank 1, and the rank value of each of the data already stored is incremented by 1. With this configuration, in this embodiment, the order in which each of the group of data to be managed was stored can be identified. Hereinafter, the rank with the smallest rank value (i.e., associated with the newest data) is also referred to as the minimum rank or the latest rank. Also, the rank with the largest rank value (i.e., associated with the oldest data) is also referred to as the maximum rank or the oldest rank.

[0026] In this embodiment, in order to simplify the explanation, a configuration will be described in which data such as one text file is stored in association with one rank. However, for example, a configuration may be used in which multiple pieces of data having common storage timing, attribute values, etc. are stored in association with one rank.

[0027] In this embodiment, each data included in a group of data is stored in association with a group value that allows identification of which of a plurality of groups the data belongs to. In this embodiment, as data groups, a first group that is a monitoring target and a target to be excluded, and a second group that is a monitoring target but not a target to be excluded are provided. As data groups, for example, a third group that is not a monitoring target may be provided. Data belonging to the first group is stored in association with a group value=1. Data belonging to the second group is stored in association with a group value=2. Data belonging to the third group is stored in association with a group value=3. The group value of each data is designated, for example, based on a designation operation performed by a user using the user terminal 20. The group value of each data may be automatically designated according to a predetermined algorithm, for example, based on the attributes, characteristics, structure, meaning, and relationship of the data.

[0028] The monitoring means 2 has a function of monitoring the group of managed data. The monitoring means 2 monitors, for example, whether or not it is necessary to remove some data from the group of managed data. Specifically, the monitoring means 2 monitors the amount of data of group value=1 and group value=2 among the group of data stored in the storage unit 1. When the monitoring means 2 determines that the amount of data will exceed a predetermined capacity due to the addition of new data to the group of data, it notifies the extraction means 3 to that effect.

[0029] In this embodiment, in order to simplify the explanation, a case will be explained in which the upper limit of the number of ranks for a group of data to be managed is fixedly determined in advance. In this case, when the total number of data with group value=1 or 2 (for example, the total number of text files described above) among the group of data to be managed exceeds the predetermined upper limit of ranks (for example, 100), the monitoring means 2 notifies the extraction means 3 of that fact.

[0030] In this embodiment, the upper limit of the number of ranks for a group of data to be managed (i.e., the upper limit of the total number of data) is fixedly determined in advance. However, the upper limit of the number of ranks may be variable. For example, the number of ranks for a group of data to be managed may be fixedly determined according to the amount of resources that can be supplied related to storage (hereinafter also referred to as the supply amount), or may be variable according to changes in the supply amount.

[0031] The supply amount may be determined based on, for example, the total amount or number of data that can be stored in a specified storage medium. The supply amount may also be determined based on, for example, the amount of communication communicated in a predetermined period in a communication means (not shown) included in the diversity concentration device 10. The supply amount may also be determined based on the amount of calculation calculated in a predetermined period in a calculation means (not shown) such as a CPU included in the diversity concentration device 10. The supply amount may also be determined based on, for example, a value input by a user via a user input means (not shown) such as a keyboard.

[0032] The extraction means 3 has a function of extracting data from the storage unit 1. For example, based on a notification from the monitoring means 2, the extraction means 3 extracts data with a group value = 1 or 2 from a group of data to be managed. The extraction means 3 also transmits the extracted data to the diversity concentration device 10. Note that the extraction means 3 may transmit only metadata of the data with group value 1 or 2 to the diversity concentration device 10, rather than the data itself with group value 1 or 2. When the group of data is a collection of inventory data, the extraction means 3 transmits the inventory data with group value 1 or 2 to the diversity concentration device 10.

[0033] In this embodiment, the extraction means 3 is configured to perform data extraction processing and transmission processing based on a notification from the monitoring means 2. However, for example, the extraction means 3 may perform data extraction processing and transmission processing based on an instruction operation performed by a user using the user terminal 20. That is, the extraction means 3 may be configured to be able to issue an instruction to exclude some data from a group of data at the user's discretion. For example, the user may use the user terminal 20 to input extraction instruction information indicating the range and conditions of data to be extracted from a group of data, and exclusion target information indicating the amount of data to be excluded. In this case, the extraction means 3 transmits the exclusion target information to the diversity concentration device 10 together with the data extracted based on the extraction instruction information. In addition, the diversity concentration device 10 repeats the process of identifying data to be excluded until the exclusion target is achieved.

[0034] The input means 5 of the diversity concentration device 10 has a function of inputting data. The input means 5 inputs, for example, a group of data transmitted by the extraction means 3.

[0035] The exclusion target identification means 6 has a function of identifying data to be excluded from a group of data input by the input means 5. The exclusion target identification means 6 identifies data to be excluded from, for example, data with group value=1 in the group of data input by the input means 5.

[0036] The ideal distribution specifying means 61 included in the removal target specifying means 6 has a function of specifying an ideal distribution that shows an ideal distribution state with high diversity for the distribution of attribute values ​​of a group of data. Note that the ideal distribution state with high diversity is not uniquely determined, but may be any desirable distribution state. The ideal distribution may be determined based on information indicating conditions input by a user via a user input means such as a keyboard, for example.

[0037] In this embodiment, various distributions can be applied as the ideal distribution. For example, the ideal distribution may be composed of a single phase whose distribution is specified by one function, or may be composed of multiple phases whose distributions are specified by mutually different functions.

[0038] For example, an ideal distribution in the case where the upper limit total number of data stored as a group of data is 9 will be described. In this case, when the range of attribute values ​​is 1 to 260 and the minimum increment of attribute values ​​is 1, applying exponential distribution to the entire region as the ideal distribution can efficiently maintain a desired distribution state such as {1, 2, 4, 8, 16, 32, 64, 128, 256}. However, for example, when the range of attribute values ​​is 1 to 14 and the minimum increment of attribute values ​​is 1, this is not the case. Specifically, the ideal distribution includes decimals such as 1.1, but since the significant increment is 1, rounding to 1 times results in overlaps such as {1, 1, 2, 3, 4, 5, 6, 10, 14}. In this way, when the range is small compared to the increment of attribute values, applying exponential distribution to the entire region as the ideal distribution cannot efficiently maintain a desired distribution state.

[0039] In this embodiment, in order to efficiently maintain a desired distribution state, an ideal distribution composed of a proportional phase to which a proportional distribution is applied and an exponential phase to which an exponential distribution is applied is applied. The ideal distribution composed of a proportional phase and an exponential phase will be described below.

[0040] FIG. 2 is an explanatory diagram showing an example of an ideal distribution composed of a proportional phase and an exponential phase. In FIG. 2, FIG. 2(1) shows a state in which a group of data is held at a first time point. FIG. 2(2) shows a state in which a group of data is held at a second time point after a predetermined period has elapsed from the first time point shown in FIG. 2(1). In the example shown in FIG. 2(1)(2), a process of removing (i.e. thinning out) data from the group of data and adding new data (hereinafter also referred to as an update process for the group of data) is performed multiple times during the predetermined period from the first time point to the second time point.

[0041] The horizontal axis (x-axis) of the graph shown in FIGS. 2(1) and 2(2) represents the rank value of the data as Rank. The vertical axis (y-axis) of the graph represents, on a logarithmic scale, the data age (e.g., the number of days or hours from the generation date and time to the current date and time, etc.) specified as the attribute value of the data as Age. The graph shown in FIGS. 2(1) and 2(2) is a phase diagram showing the relationship between Rank and Age. In this example, the rank k of a group of data is set as x k and the logarithm of Age corresponding to the rank k is set as y k . Also, assume that a group of data contains c pieces of data. Therefore, k takes values from 1 to c. As shown in FIGS. 2(1) and 2(2), the coordinates of the data with rank c corresponding to the oldest rank are represented by (x c , y c ). Hereinafter, the coordinates (x c , y c ) of the data with rank c are also referred to as the oldest point c.

[0042] In each graph shown in FIGS. 2(1) and 2(2), the region above the transition curve corresponds to the proportional phase, and the region below the transition curve corresponds to the exponential phase. In this embodiment, various methods can be applied to determine the point (hereinafter, also referred to as the transition point t) at which the transition occurs from the proportional phase to the exponential phase. For example, the method shown below may be applied.

[0043] As shown in FIGS. 2(1) and 2(2), the transition point t is represented by (x t , y t ). In this case, the ideal distribution in the proportional phase (x ≤ x t ) and the ideal distribution in the exponential phase (x t < x) are smoothly connected at the transition point t. Also, the ideal distribution in the exponential phase passes through the transition point t and the oldest point c.

[0044] In the semi-logarithmic graph shown in FIG. 2, the ideal distribution in the proportional phase is expressed by a logarithmic distribution. The ideal distribution in the proportional phase is expressed by the following formula (1). In the semi-logarithmic graph shown in FIG. 2, the ideal distribution in the exponential phase is expressed by a linear distribution. The ideal distribution in the exponential phase is expressed by the following formula (2) since it is a straight line passing through the oldest point c and the transition point t. Since the slope of this straight line is equal to the slope at the transition point t, the transition point t exists on the transition curve of the following formula (3). Therefore, the intersection of the line connecting the transition curve and each data point of the group of data can be specified as the transition point t. Note that if the intersection cannot be uniquely determined, the point closest to the oldest point c among the multiple intersections may be specified as the transition point t.

[0045]

number

[0046]

number

[0047]

number

[0048] As shown in Figure 2 (1) and (2), the position of the transition point t changes depending on the state of holding the group of data. That is, the position of the transition point t changes when an update process is performed on the group of data. Therefore, the ranges of the proportional phase and the exponential phase also change when an update process is performed on the group of data.

[0049] In addition, as shown in Figures 2(1) and 2(2), the ideal distribution changes depending on the retention state of a group of data. In other words, the ideal distribution shown in Figures 2(1) and 2(2) changes when an update process is performed on a group of data. Therefore, by using an ideal distribution such as that shown in Figure 2, the distribution state of the data retention interval, etc. can be automatically optimized according to changes in the operational status over time, etc.

[0050] The transition point t and the ideal distribution are determined based on a group of data. In this embodiment, the upper limit of the total number of data set for the group of data is the oldest rank value (x c ) corresponds to the oldest rank value (x c ) is 100. Therefore, in this embodiment, the transition point t is determined based on the upper limit total number of data set for a group of data, that is, the oldest rank value (x c ) and the attribute value (y c The ideal distribution is calculated based on the transition point t and the earliest point c calculated in this way.

[0051] The ideal distribution is preferably specified by the outer edge obtained by connecting each data point in the attribute value space. For example, in a data group on a one-dimensional age space, the oldest data and the latest data (or 0) are the outer edge. In this embodiment, the ideal distribution is a state in which they are evenly distributed on a logarithmic scale. In a state in which all data points are on the outer edge (e.g., this is likely to occur when the dimension of the attribute value is high relative to the number of data points), the ideal distribution may be specified by the degree of deviation from the center of gravity or median of the data points. For example, the ideal distribution may be a state in which the average score vector in the five-subject score space is distributed by standard distribution. In addition, if a cluster of data points can be drawn, this may be done for each cluster. For example, if there is a science cluster and a humanities cluster, the distribution can be specified in the same way for each and then added together to specify an ideal distribution that is closer to reality.

[0052] 2, the ideal distribution specifying means 61 first specifies a transition point t. Next, the ideal distribution specifying means 61 specifies an ideal distribution based on the specified transition point t.

[0053] In this embodiment, the ideal distribution is composed of a plurality of phases (i.e., proportional phases and exponential phases) having different distribution characteristics, which allows a desirable distribution state to be expressed appropriately.

[0054] In this embodiment, the ideal distribution specifying means 61 specifies an ideal distribution consisting of two phases, a proportional phase and an exponential phase. However, for example, the ideal distribution specifying means 61 may specify an ideal distribution consisting of three or more phases with different distribution characteristics. In this case, the ideal distribution specifying means 61 may determine whether or not data belonging to each phase is to be excluded.

[0055] Also, the ideal distribution specifying means 61 may specify the ideal distribution by, for example, a histogram. Also, when a group of data is expressed by a two-dimensional distribution, the ideal distribution specifying means 61 may specify an approximation curve by the least squares method, and set the specified approximation curve as the ideal distribution.

[0056] The post-change distribution evaluation means 62 included in the exclusion target identification means 6 has a function of identifying the distribution of attribute values ​​of a group of data after the change (hereinafter also referred to as the post-change distribution) when specific data is removed from the group of data input by the input means 5. Note that, for example, the post-change distribution may be the distribution of attribute values ​​of each data when specific data is removed from the group of data and the latest data is added to the group of data.

[0057] Generally, new data is more likely to be useful than old data. In addition, if the oldest data is removed, the diversity of the entire group of data may decrease. Therefore, in this embodiment, the new data stored after the rank corresponding to the transition point t (hereinafter also referred to as the transition rank) and the oldest data corresponding to the oldest rank are configured not to be removed. Therefore, the post-change distribution evaluation means 62 specifies the post-change distribution when each data, except for the new data stored after the transition rank (i.e., data included in the proportional phase) and the oldest data corresponding to the oldest rank, among the data that can be removed (i.e., data with a group value of 1), is removed from the group of data. Note that the configuration is not limited to this embodiment, and the new data stored after the transition rank (i.e., data included in the proportional phase) and the oldest data corresponding to the oldest rank may also be configured to be removed.

[0058] The post-change distribution evaluation means 62 has a function of evaluating the degree of mismatch (or the degree of match) with the ideal distribution for each identified post-change distribution.

[0059] The post-change distribution evaluation means 62 obtains, for example, the difference between the ideal distribution and the post-change distribution as a deviation. Next, the post-change distribution evaluation means 62 evaluates that the degree of mismatch is high (or the degree of agreement is low) when the deviation is large, and evaluates that the degree of mismatch is small (or the degree of agreement is high) when the deviation is small. Note that the evaluation of the degree of mismatch (or agreement) between the ideal distribution and the post-change distribution is not limited to the exemplified method, and various other methods can be applied.

[0060] The exclusion target identification means 6 compares the degree of inconsistency (or degree of agreement) evaluated for each post-change distribution, and identifies the post-change distribution with the lowest degree of inconsistency (or highest degree of agreement). That is, the exclusion target identification means 6 identifies the post-change distribution that is most similar to the ideal distribution from multiple post-change distributions. The exclusion target identification means 6 also identifies data excluded when generating the identified post-change distribution as data to be excluded. That is, the exclusion target identification means 6 identifies data to be excluded so that the distribution of a group of data after excluding the data to be excluded is most similar to the ideal distribution.

[0061] An example of identifying data to be excluded will be conceptually described with reference to FIG. 3. FIG. 3 is an explanatory diagram showing an example of identifying data to be excluded. In the graph shown in FIG. 3, for each rank other than the oldest rank included in the exponential phase, the degree to which the post-change distribution when the data of that rank is excluded is close to the ideal distribution is shown as Gain. In the graph shown in FIG. 3, the scale on the right side of the graph corresponds to the degree of Gain. The value of Gain can be obtained, for example, by the exclusion target identifying means 6 identifying the post-change distribution when the data corresponding to each rank is excluded, and evaluating the degree of inconsistency between the post-change distribution and the ideal distribution.

[0062] In the graph shown in Fig. 3, the data with the rank with the maximum Gain value is specified as the data to be excluded. The data with the rank with the maximum Gain value is data that, when the data is excluded from a group of data, most closely approximates the distribution of the attribute values ​​of the group of data to the ideal distribution. In other words, the data with the rank with the maximum Gain value is data that contributes little to the diversity of the group of data.

[0063] In this embodiment, the post-change distribution evaluation means 62 does not specify a post-change distribution when each of the data that cannot be excluded (i.e., data with a group value of 2) is removed from the group of data. However, the post-change distribution evaluation means 62 may specify a post-change distribution when each of the data in the exponential phase, excluding the data with the oldest rank, is removed from the group of data, regardless of the group value of the data. In this case, the exclusion target identification means 6 determines whether or not the data removed when generating the post-change distribution is data with a group value of 1 (i.e., data that can be removed) after specifying the post-change distribution that is most similar to the ideal distribution. If the data is data with a group value of 1, the exclusion target identification means 6 sets the data as a target for exclusion. If the data is not data with a group value of 1, the exclusion target identification means 6 next specifies a post-change distribution that is similar to the ideal distribution. Next, the exclusion target identification means 6 determines whether or not the data removed when generating the specified post-change distribution is data with a group value of 1 (i.e., data that can be removed). The exclusion target identification means 6 repeats the process of identifying a post-change distribution that approximates the ideal distribution and the process of determining whether the data has a group value=1, until data with a group value of 1 is identified.

[0064] The output means 7 has a function of outputting information indicating the data to be excluded that is identified by the exclusion target identification means 6 .

[0065] The output means 7 outputs, for example, information indicating the data to be excluded to the user terminal 20. In this case, the user can use the user terminal 20 to check the identified data to be excluded. Furthermore, the user can use the user terminal 20 to instruct whether or not to exclude the identified data to be excluded. Furthermore, in the case where multiple exclusion methods are provided, the user can use the user terminal 20 to instruct the exclusion method. When the user uses the user terminal 20 to perform an input operation to instruct the execution of exclusion, the user terminal 20 transmits, to the exclusion means 4, information indicating the data to be excluded, information instructing the execution of exclusion, information specifying the exclusion method, and the like.

[0066] The output means 7 may be configured to output information indicating the data to be excluded to the exclusion means 4, for example, without going through the user terminal 20. That is, the output means 7 may be configured to automatically perform a process of excluding (i.e., thinning) data from a group of data. In this case, the output means 7 may also output information specifying an exclusion method (for example, a predetermined exclusion method, or an exclusion method that is automatically determined according to a predetermined algorithm based on the attributes, characteristics, structure, meaning, and relationship of the data to be excluded). In addition, for example, the output means 7 may be configured to output data to be excluded whose group value has been updated to a value other than 1 (for example, group value=3) to the storage unit 1, without going through the user terminal 20 and the exclusion means 4, to update the group of data.

[0067] The user terminal 20 is realized by an information processing device such as a personal computer, a smartphone, a tablet computer, etc. For example, the user terminal 20 displays information indicating the data to be excluded that is output from the output means 7 on a display unit. Also, for example, the user terminal 20 transmits information indicating the data to be excluded, information instructing the execution of exclusion, information specifying the exclusion method, etc. to the exclusion means 4 based on an input operation by the user.

[0068] The exclusion means 4 has a function of excluding data to be excluded from a group of data stored in the storage unit 1.

[0069] In this embodiment, the expression "removing data from a group of data" is used, which includes not only deleting data but also changing the management form of data. For example, the process of removing specific data from a group of data includes processes such as changing the group value of specific data in the group of data to a value other than 1 (for example, group value = 3), transferring the storage destination to another storage medium, performing lossless or lossy compression, and lowering the resolution.

[0070] The process of removing data from a group of data will be described by taking as an example a case where a group of data is stored containing a video consisting of frame image data corresponding to each of successive frames. In this case, the process of removing data from a group of data includes a process of deleting a frame image data item identified as the item to be removed, a process of reducing the resolution of the frame image data item, and a process of lossless or lossy compression of the frame image data item.

[0071] The exclusion means 4 excludes data to be excluded, for example, based on information output from the output means 7 or information transmitted from the user terminal 20. In addition, the exclusion means 4 excludes data to be excluded, for example, by a preset exclusion method or an exclusion method designated by information transmitted from the output means 7 or the user terminal 20.

[0072] Next, a description will be given of the processing operation of the diversity concentrating device 10. FIG 4 is a flowchart showing an example of the processing operation of the diversity concentrating device 10.

[0073] When the extraction means 3 transmits the group of data extracted to the diversity concentration device 10, the input means 5 receives and inputs the group of data (Y in step S1).

[0074] Next, the ideal distribution specifying means 61 specifies a transition point t between the proportional phase and the exponential phase in the ideal distribution corresponding to the group of data input by the input means 5 (step S2).

[0075] Next, the ideal distribution specifying means 61 specifies an ideal distribution based on the specified transition point t (step S3).

[0076] Next, the post-change distribution evaluation means 62 assigns a value obtained by subtracting 1 from the value of the oldest rank c to the variable k (step S4). That is, the post-change distribution evaluation means 62 specifies the rank k corresponding to the data that is one newer than the oldest rank c.

[0077] Next, the post-change distribution evaluation means 62 judges whether the data of rank k is data that can be excluded (i.e., data with group value=1) (step S5). If the data is not data that can be excluded, the process proceeds to step S8.

[0078] If the data of rank k is removable, the post-change distribution evaluation means 62 specifies the post-change distribution when the data corresponding to rank k is removed from the group of data (step S6).

[0079] Next, the post-change distribution evaluation means 62 evaluates the degree of mismatch between the specified post-change distribution and the ideal distribution (step S7).

[0080] Next, the post-change distribution evaluation means 62 subtracts 1 from the value of the variable k (step S8). Next, the post-change distribution evaluation means 62 determines whether the value of the variable k coincides with the value of the transition rank t (step S9).

[0081] If the value of the variable k does not match the value of the transition rank t (N in step S9), the process proceeds to step S5. By repeatedly executing the processes in steps S5 to S9, a post-change distribution is identified in which data corresponding to a rank one newer than the oldest rank c to data corresponding to a rank one older than the transition rank t is excluded.

[0082] If the value of the variable k matches the value of the transfer rank t (Y in step S9), the exclusion target identification means 6 compares the degree of inconsistency evaluated for each post-change distribution and identifies the post-change distribution with the lowest degree of inconsistency (step S10).

[0083] Next, the exclusion target identification means 6 identifies the data excluded when generating the identified post-change distribution as data to be excluded (step S11).

[0084] Next, the output means 7 outputs information indicating the data to be excluded (step S12).

[0085] In the example shown in FIG. 4, the diversity concentration device 10 ends the process based on outputting information indicating the identified data to be excluded. However, for example, the diversity concentration device 10 may be configured to end the process when a termination condition such as achieving the exclusion goal (i.e., identifying data to be excluded that has a data amount designated as the exclusion goal) is satisfied. That is, the diversity concentration device 10 may be configured to continue the process of identifying data to be excluded until the termination condition is satisfied. In this case, for example, the diversity concentration device 10 may be configured to output information indicating the data each time data to be excluded is identified. Also, for example, the diversity concentration device 10 may be configured to output information indicating the data collectively after identifying a plurality of data to be excluded that achieve the exclusion goal.

[0086] In this embodiment, the exclusion target identification means 6 identifies data to be excluded from a group of data (e.g., a group of data itself, a group of metadata, a group of catalog data, etc.) that is a group of data including attribute values, such that the distribution of attribute values ​​in the group of data after the data to be excluded is approximate to an ideal distribution. With this configuration, in this embodiment, data that contributes little to diversity can be suitably identified as data to be excluded from the group of data, and the diversity of the group of data can be efficiently maintained.

[0087] In this embodiment, the ideal distribution specifying means 61 specifies an ideal distribution based on a predetermined total amount determined for a group of data (for example, an upper limit total number of the group of data; the oldest point c is specified by the upper limit total number). In addition, the post-change distribution evaluation means 62 specifies a post-change distribution, which is the distribution of attribute values ​​when any one data is removed from the group of data. Next, the exclusion target specifying means 6 specifies, as data to be excluded, data removed when generating a post-change distribution that is most consistent with the ideal distribution or a post-change distribution that is least consistent with the ideal distribution. With this configuration, in this embodiment, it is possible to specify data to be excluded so as to most closely approximate a desired distribution state of attribute values.

[0088] Next, we will explain the difference between managing backup data by applying a backup method that deletes old data and managing backup data by applying a diversity concentration method using the diversity concentration device 10 of this embodiment.

[0089] FIG. 5 is a diagram showing a comparative example between a generation that is retained when a backup method that deletes data starting from old data is applied and a generation that is retained when a diversity concentration method is applied.

[0090] FIG. 5(1) is a diagram showing an update history by backup. The blocks shown in FIG. 5(1) represent all generations that have been backed up. The area marked with the symbol a in the blocks shown in FIG. 5(1) represents data of the 41st to 50th generations. The area marked with the symbol b in the blocks shown in FIG. 5(1) represents data of the 31st to 40th generations. The area marked with the symbol c in the blocks shown in FIG. 5(1) represents data of the 21st to 30th generations. The area marked with the symbol d in the blocks shown in FIG. 5(1) represents data of the 11th to 20th generations. The area marked with the symbol e in the blocks shown in FIG. 5(1) represents data of the 10th to 10th generations.

[0091] Fig. 5(2) is a diagram showing an example of data that is retained when a backup method that deletes data from the oldest is applied. Fig. 5(2) shows an example of a case where data for up to 10 generations can be retained.

[0092] When applying the backup method of deleting data starting from the oldest one shown in FIG. 5(2), it is assumed that the user voluntarily limits the backup frequency to half in order to increase the number of generations that can be retained.

[0093] FIG. 5(2-1) of FIG. 5(2) is a diagram showing an example of data that is retained when 20 generations have passed since the start of backup. FIG. 5(2-1) shows an example in which data from roughly the 1st to 20th generations is retained, skipping one generation. FIG. 5(2-2) of FIG. 5(2) is a diagram showing an example of data that is retained when 50 generations have passed since the start of backup. FIG. 5(2-2) shows an example in which data from roughly the 31st to 40th generations is retained, skipping one generation.

[0094] As shown in Figure 5 (2-2), if a backup method that deletes old data is applied, even if the backup frequency is limited to half in order to increase the number of generations that can be retained, all data prior to a certain point in time (in this example, data from generations prior to the 30th generation) will be lost.

[0095] Fig. 5(3) is a diagram showing an example of data stored when the diversity concentration method using the diversity concentration device 10 of the present embodiment is applied. Fig. 5(3) shows an example of a case where data for up to 10 generations can be stored, similar to Fig. 5(2).

[0096] When the diversity concentration method is applied, the diversity concentration device 10 automatically holds data in a state of high diversity. Therefore, when the diversity concentration method is applied, unlike the case of applying a backup method in which data is deleted from the oldest data, the user does not need to voluntarily limit the backup frequency.

[0097] FIG. 5(3-1) of FIG. 5(3) is a diagram showing an example of data that is retained when 20 generations have passed since the start of backup. FIG. 5(3-1) shows an example in which data from roughly the 1st to 20th generations is retained, skipping one generation. FIG. 5(3-2) of FIG. 5(3) is a diagram showing an example of data that is retained when 50 generations have passed since the start of backup. FIG. 5(3-2) shows an example in which data from roughly the 1st to 50th generations is retained, skipping four generations.

[0098] As shown in FIG. 5 (3-2), when the diversity concentration method is applied, the first generation of data at the beginning of backup can be retained without voluntarily limiting the backup frequency.

[0099] Furthermore, as shown in FIG. 5 (3-2), when the diversity concentration method is applied, even when 50 generations have passed since the start of backup, data from roughly the 1st to 50th generations is still retained, skipping four generations. Therefore, for example, even if infected with malware with a long incubation period, data from roughly 0 to 4 generations before the infection is always retained. In this way, when the diversity concentration method is applied, the diversity of the retained generations can be amplified and diversity can be preserved.

[0100] Next, the transition of the amount of information to be stored will be described. Fig. 6 is a diagram showing a comparative example of the transition of the amount of information to be stored when a backup method that deletes old data is applied and the transition of the amount of information to be stored when a diversity concentration method is applied.

[0101] FIG. 6(1) is a diagram showing the transition of the amount of information held when a backup method of deleting old data is applied. FIG. 6(2) is a diagram showing the transition of the amount of information held when a diversity concentration method using the diversity concentration device 10 of this embodiment is applied. FIG. 6(3) is a diagram showing a comparison between the example shown in FIG. 6(1) and the example shown in FIG. 6(2). As shown in FIG. 6(3), the total amount of information S1 held in the example shown in FIG. 6(1) and the total amount of information S2 held in the example shown in FIG. 6(2) are the same amount of information.

[0102] As shown in FIG. 6(1), when a backup method that deletes old data is applied, all information stored before the time T1 has elapsed is retained, but no information stored after the time T1 has elapsed is retained.

[0103] As shown in FIG. 6(2), when the diversity concentration method is applied, the amount of information retained gradually decreases with time, but information after time T1 is also retained.

[0104] When the diversity concentration method is applied, it is possible to reproduce a forgetting curve in which memory is gradually lost over time, as shown in Figure 6(2), unlike the memory characteristic in which all memories prior to a certain point are lost, as shown in Figure 6(1).

[0105] The backup method of deleting old data first has the following characteristics (1) and (2). (1) The retroactive period is the number of generations that can be stored in the storage x c x [times] and the backup interval s [h] c ·s[h]. Data from before that period cannot be obtained. (2) If you need data from a time t[h] or earlier within the retroactive period, you can obtain data from t to t+s[h]. The range of the time difference does not depend on t.

[0106] The diversity enrichment method according to the present invention has the following features (1) and (2). (1) Storage capacity x c It is possible to trace back to the beginning of operation, regardless of the backup interval s [h]. (2) If you need data from a time t [h] or earlier, you can get data from approximately t to t (1 + ε) [h]. The range of the time difference depends on t.

[0107] The time difference rate ε described in the feature (2) of the diversity concentration method according to the present invention can be calculated, for example, as follows.

[0108] Number of generations that can be stored in storage x c For example, we will explain the case where the backup interval s is 5 hours and the system is operated for 4096 hours. t is 21 generations ago. Therefore, the ideal distribution is calculated by the following formula (4). The base in formula (4) corresponds to 1+ε. Therefore, the time difference rate ε in this case is 4.75%.

[0109]

number

[0110] Next, the number of generations that can be stored in the storage x c For example, we will explain the case where the backup interval s is 5 hours and the system is operated for 262144 hours. t is 10 generations ago. Therefore, the ideal distribution is calculated by the following formula (5). Therefore, the time difference rate ε in this case is 9.98%.

[0111]

number

[0112] In this way, when the diversity concentration method according to the present invention is applied, even if the operation time or backup frequency increases by 64 times, the time difference rate does not increase by 64 times. The impact of increasing the operation time or backup frequency by 64 times is only a slight increase in the time difference rate. For example, assume that the incubation period of malware is about six months. When the diversity concentration method according to the present invention is applied, data within a time difference range of the time when the malware was lurking + a few percent can be obtained. If this time difference rate is acceptable, it can be said that the backup method applying the diversity concentration method according to the present invention, which has no constraints on the retroactive period, is extremely practical.

[0113] FIG. 7 is a diagram showing a comparative example between the distribution of data held when a backup method that deletes data starting from old data is applied and the distribution of data held when a diversity concentration method is applied.

[0114] Fig. 7(1-1) is a diagram showing an example in which 4096 generations have passed since the start of backup in a configuration in which a backup method that deletes old data is applied. Fig. 7(1-2) is a diagram showing an example in which 4096 generations have passed since the start of backup in a configuration in which a diversity concentration method is applied.

[0115] Fig. 7(2-1) is a diagram showing an example in which 262144 generations have passed since the start of backup in a configuration in which a backup method that deletes old data is applied. Fig. 7(2-2) is a diagram showing an example in which 262144 generations have passed since the start of backup in a configuration in which a diversity concentration method is applied.

[0116] As shown in Figure 7, when the diversity concentration method is applied, data can be kept in a state that is always close to the ideal distribution, compared to when the backup method of deleting old data is applied. In addition, when time passes from the 4096th generation to the 262144th generation, when the backup method of deleting old data is applied, the deviation from the ideal distribution becomes larger. However, when the diversity concentration method is applied, there is no significant change in the state that is close to the ideal distribution.

[0117] A part or all of the above-described embodiment can be described as, but is not limited to, the following (1) to (2).

[0118] (1) A diversity concentration device including: a means for identifying data to be excluded (e.g., implemented by an exclusion target identifying means 6) that identifies data to be excluded from a group of data (e.g., a collection of data itself, a collection of metadata, a collection of catalog data, etc.) that is a collection of data including attribute values ​​(e.g., values ​​indicating attributes of the data itself, values ​​indicating attributes of other data, values ​​indicating attributes of objects or events not expressed by data, etc.) so that the distribution of attribute values ​​in the group of data after the data to be excluded is excluded approximates a specific distribution (e.g., an ideal distribution); and an output means (e.g., implemented by an output means 7) that outputs information indicating the data to be excluded identified by the identifying means. With this configuration, data that contributes little to diversity can be appropriately identified as data to be excluded from the group of data, and the diversity of the group of data can be efficiently maintained.

[0119] (2) In the diversity concentration device described in (1) above, the specifying means specifies a specific distribution based on a predetermined total amount (for example, the upper limit total number of the group of data. The oldest point c is specified by the upper limit total number) determined for the group of data (for example, the ideal distribution specifying means 61 corresponds to the part that executes the process of step S3 shown in FIG. 4), specifies a predetermined distribution (for example, a post-change distribution) that is a distribution of attribute values ​​when any one data is removed from the group of data (for example, the post-change distribution evaluation means 62 corresponds to the part that executes the process of step S6 shown in FIG. 4), and specifies the data removed when specifying the predetermined distribution that has the highest degree of agreement with the specific distribution or the predetermined distribution that has the lowest degree of disagreement with the specific distribution as data to be excluded (for example, the exclusion target specifying means 6 corresponds to the part that executes the processes of steps S10 to S11 shown in FIG. 4). With this configuration, it is possible to specify data to be excluded so as to most closely approximate the distribution state of the desired attribute values.

[0120] Although the present invention has been described above with reference to the embodiment, the present invention is not limited to the above embodiment. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present invention within the scope of the present invention. [Industrial Applicability]

[0121] The present invention is applicable to applications such as version management for CAD (computer-assisted / aided drafting), programming, and game save data. For example, application of the present invention makes it possible to extend the storage period and diversify the storage time in version management. Furthermore, by enabling the extension of the storage period and diversification of the storage time, when an irreversible operational error occurs, it becomes easy to restore to a state before the operational error occurred and with minimal rework. Therefore, the present invention is also preferably applicable to generation management of backup data in preparation for latent malware.

[0122] The present invention is applicable to applications for managing data recorded by security cameras and event data recorders. By applying the present invention, unlike the conventional usage pattern in which the storage period was limited, it is possible to extend the storage period. Therefore, it is possible to prevent the occurrence of a de facto statute of limitations in which investigations are difficult due to the absence of recorded data. In addition, by applying the present invention, the conventional usage pattern of recording recent situations can be changed to a usage pattern in which normal situations can be recorded. It is expected that these effects will lead to improved deterrence against crimes and harassment.

[0123] Furthermore, by applying the present invention, even in the case of recorders for which there is a strong demand for recording normal conditions, such as life logs and animal logs, it becomes possible to preserve normal condition records without spending time on data management.

[0124] The present invention can be used in applications where artificial intelligence can achieve memory characteristics that are more similar to those of humans. For example, a memory system to which the present invention is applied can reproduce a forgetting curve in which memories are gradually lost over time, unlike the conventional memory characteristics in which memories from a certain period of time are completely lost.

[0125] The present invention can be applied to applications in which, by using the overwriting of a file as a trigger, each time the file is saved, the file is automatically backed up and the backup data is automatically deleted.

[0126] The present invention is preferably applied to software for storing and managing various data, information processing devices such as versioning file systems and version control systems, digital cameras, and event data recorders. In this case, even if the free space of a hard disk that was previously wasted is allocated to these data, free space can be secured at any time as needed, making it possible to make the most of storage resources.

[0127] The present invention can be applied to applications that provide decision support functions that maximize diversity in groups allowed by limited resources other than memory resources. For example, by adopting this method as decision support for production and inventory management, it is possible to enrich the variety of limited inventory.

[0128] The present invention is applicable to applications that provide a diversification mechanism that does not necessarily depend on probability to information processing devices and the like. For example, heuristics that optimize evaluation values, such as genetic algorithms, have the problem that the solution group impairs diversity and it is difficult to obtain a globally optimal solution (initial convergence problem). A mechanism to which the present invention is applied can output the contribution of each element of the solution group to the diversity of an arbitrary attribute value. This contribution can be considered as an additional evaluation value to the original evaluation value of the heuristic. Therefore, it is expected that the mechanism to which the present invention is applied can preserve diversity and reduce the initial convergence problem. Such a mechanism can be applied to applications such as circuit design, material development, and drug development (e.g., drug discovery and pharmaceuticals). In other words, when searching for explosives or alloys that meet the intended purpose, it is expected that a globally optimal solution can be obtained from a diverse solution group by treating a parameter set such as optimal arrangement, shape, and mixture ratio as a data group. At the same time, if applied to in silico drug discovery, in which new drugs are searched for by simulating molecular structures, etc. on a computer, it can be used to research drugs with higher activity and fewer side effects. [Explanation of symbols]

[0129] 1 Storage section 2 Monitoring means 3 Extraction means 4 Elimination means 5. Input methods 6. Methods for identifying targets for exclusion 61 Ideal distribution identification means 62 Post-change distribution assessment method 7 Output Method 10 Diversity concentration device 20 User terminal

Claims

1. a specifying means for specifying, as a target for elimination, data excluded from a predetermined distribution that is a distribution of attribute values ​​when any one data item is excluded from a group of data items that is a set of data items including attribute values, the data item that has the highest degree of agreement with the predetermined distribution or the data item that has the lowest degree of disagreement with the predetermined distribution; An output means for outputting information indicating the data to be excluded identified by the identification means.

2. The identification means is identifying the particular distribution based on a predetermined amount determined for the set of data; Identifying the predetermined distribution, which is a distribution of attribute values ​​when any one data item is removed from the group of data items; Data excluded when identifying the predetermined distribution having the highest degree of agreement with the specific distribution or the predetermined distribution having the lowest degree of disagreement with the specific distribution is identified as data to be excluded. The diversity concentrator of claim 1.

3. The computer identifying, as a target for elimination, data excluded from a predetermined distribution that is a distribution of attribute values ​​when any one piece of data is excluded from a group of data that is a collection of data including attribute values, the data excluded from the predetermined distribution that has the highest degree of agreement with the predetermined distribution or the data excluded from the predetermined distribution that has the lowest degree of disagreement with the predetermined distribution; Output information about the identified data to be excluded A method for enriching diversity comprising:

4. On the computer, a process of identifying, as a target for elimination, data excluded from a predetermined distribution that is a distribution of attribute values ​​when any one data item is excluded from a group of data items that is a set of data items including attribute values, the data item that is excluded from the predetermined distribution that has the highest degree of agreement with the predetermined distribution or the data item that has the lowest degree of disagreement with the predetermined distribution; an output process for outputting information indicating the data to be excluded that has been identified in the identification process; A diversity enrichment program for carrying out the above.

Citation Information

Patent Citations

  • Information processor, information processing method, and program

    JP2017219896A

  • An end-user behavior modeling system

    WO2023129108A1