JP2019175419A - Data providing system, data providing method, and computer program - Google Patents

Abstract

To provide a new database technique.SOLUTION: A method according to one aspect of the present disclosure includes combining a group of first feature data and a group of second feature data. The first feature data represents features of corresponding one or more first constituents, associated with a first identifier which is an identifier of the corresponding one or more first constituents, among a plurality of first constituents included in a first group. Each piece of the second feature data corresponds to each of a plurality of clusters in a second group. The second feature data includes statistical data associated with a second identifier which is an identifier of two or more second constituents included in a corresponding cluster. The statistical data represents a feature of the two or more second constituent included in the corresponding cluster as a statistic. Combining (S150) is performed on the basis of the first identifier and the second identifier.SELECTED DRAWING: Figure 3

Description

  The present disclosure relates to an information processing system, a data providing system, and related methods.

  Conventionally, a customer's purchasing behavior is analyzed based on product sales data. In order to make use for commercial activities, the customer's contact with mass media and network contents is also analyzed. Various types of information such as customer purchase behavior, mass media / network content contact behavior, and lifestyles are also collected using a questionnaire format or a face-to-face question format.

  In recent years, each company has a huge database with data on such customers. However, each company is reluctant to provide data on these customers to the outside mainly due to the protection of personal information. When these data are provided to the outside by the company that owns them, they are provided in encrypted form, provided with information that leads to the identification of customers being significantly deleted, or intentionally erroneous (noise). It is provided in a state changed to include (see Patent Document 1).

JP 2014-109647 A

  As described above, the provision of data related to customers from data holding companies is conventionally limited from the viewpoint of personal information protection.

  Therefore, in one aspect of the present disclosure, it is desirable to be able to provide a new technique related to data provision and data processing in consideration of personal information protection.

  An information processing system according to an aspect of the present disclosure includes a storage unit, an acquisition unit, and a combining unit. The storage unit is configured to store a group of first feature data relating to the first group. The acquisition unit is configured to acquire a group of second feature data relating to the second group. The first group includes a plurality of first structures. The second group includes a plurality of second structures.

  Each of the first feature data represents a feature of the corresponding one or more first components. Each of the first feature data is associated with a first identifier that is an identifier of the corresponding one or more first constructs.

  Each of the second feature data corresponds to each of the plurality of clusters in the second group. Each of the plurality of clusters includes two or more of the plurality of second structures. Each of the second feature data includes statistical data that represents the features of two or more second constituents included in the corresponding cluster by a statistical amount. The statistical data is associated with a second identifier that is an identifier of two or more second constructs included in the corresponding cluster.

  The combining unit combines each of the first feature data with the statistical data of the second feature data associated with the corresponding second identifier based on the first identifier associated with the first feature data. As described above, the first feature data group and the second feature data group are configured to be combined.

  According to this information processing system, the second feature data related to the second group is acquired as statistical data with information protection, and the structure corresponding to the statistical data is specifically identified based on the second identifier. Thus, it can be combined with the first feature data of the corresponding structure. Therefore, according to one aspect of the present disclosure, it is possible to realize a meaningful data combination suitable for the correspondence between the first group constituent and the second group constituent while realizing the protection of personal information. Can do. Thus, according to one aspect of the present disclosure, a meaningful data combining technique can be provided.

  According to one aspect of the present disclosure, an acquisition unit includes a plurality of second constituents according to a designated constraint condition in a data providing system including a plurality of individual feature data corresponding to each of the plurality of second constituents. The cluster may be requested to be clustered into a plurality of clusters, and a group of second feature data corresponding to the plurality of clusters complying with the constraints may be acquired from the data providing system.

  The data providing system is configured to cluster a plurality of second constructs into a plurality of clusters according to a constraint condition, and to provide a group of second feature data corresponding to the plurality of clusters to the information processing system. Also good. According to one aspect of the present disclosure, the statistical data is obtained by statistically analyzing the characteristics of the two or more second components indicated by the individual feature data of the two or more second components included in the corresponding cluster. It may be generated by converting to a quantity.

  When the information processing system designates the constraint condition for the data providing system, the information processing system can control the clustering of the data providing system. As a result, it is possible to suppress the loss of meaningful information when information included in a plurality of individual feature data corresponding to one cluster is collected as statistical data. That is, it is possible to suppress deterioration of the value of information provided as statistical data from the data providing system due to inappropriate clustering in the data providing system.

  According to one aspect of the present disclosure, the plurality of second constituent members include a plurality of corresponding constituent members corresponding to any of the plurality of first constituent members and a plurality of first constituent members. A plurality of non-corresponding constructs that do not correspond may be included. Each of the second feature data acquired by the acquisition unit may correspond to each of a plurality of clusters defined by clustering a plurality of corresponding constituents in the second group.

  According to one aspect of the present disclosure, the acquisition unit may request the data providing system to cluster a plurality of corresponding constructs into a plurality of clusters. In accordance with this request, the data providing system may be configured to cluster a plurality of corresponding constructs into a plurality of clusters and provide a group of second feature data corresponding to the plurality of clusters to the information processing system. .

  According to one aspect of the present disclosure, the acquisition unit may send a list of constructs to the data providing system. Specifically, the list may be a list of a plurality of first structures or a list of a plurality of corresponding structures. The data providing system may be configured to identify a plurality of corresponding entities in the second group based on the list.

According to one aspect of the present disclosure, the acquisition unit may transmit distance information representing a distance on a feature space between a plurality of constituents included in the list to the data providing system. The data providing system may be configured to cluster a plurality of corresponding constituents in the second group into a plurality of clusters based on the distance information. Providing such distance information is useful for proper clustering of the second group in the data providing system.

  According to one aspect of the present disclosure, the acquisition unit may transmit classification information representing each classification of the plurality of constituents included in the list to the data providing system. The data providing system may be configured to cluster a plurality of corresponding constituents in the second group into a plurality of clusters based on the classification information so that corresponding constituents of different classifications are not mixed in one cluster. Good. Providing such classification information is also useful for proper clustering of the second group in the data providing system.

  According to one aspect of the present disclosure, the plurality of corresponding constituents in the second group may have the same configuration as any of the plurality of first constituents in the first group. Between the first group and the second group, the same identifier may be assigned to each of the plurality of corresponding components. The combining unit may combine each of the first feature data with statistical data of the second feature data associated with the same identifier.

  According to one aspect of the present disclosure, the storage unit may store a correspondence relationship between the first identifier and the second identifier. The combining unit may combine each of the first feature data with statistical data of the second feature data associated with the corresponding second identifier according to the correspondence relationship.

  According to one aspect of the present disclosure, each of the first feature data may correspond to each of a plurality of clusters in the first group. Each of the plurality of clusters may include two or more of the plurality of first structures. The first feature data may include statistical data that represents the features of two or more first constituents included in the corresponding cluster by a statistical amount. The statistical data may be associated with identifiers of two or more first structures included in the corresponding cluster.

  According to one aspect of the present disclosure, in an environment in which the plurality of second constituents include a plurality of corresponding constituents corresponding to any of the plurality of first constituents, the acquisition unit is as follows: It may work. That is, the acquisition unit clusters a plurality of corresponding constituents into a plurality of clusters based on the first feature data of the corresponding first constituents, and selects a cluster to which each of the plurality of corresponding constituents belongs. Providing identifiable cluster information to a data providing system including a plurality of individual feature data corresponding to each of a plurality of second structures, and obtaining a group of second feature data according to the cluster information from the data providing system May be. In this case, for each cluster identified from the cluster information, the data providing system statistically analyzes the characteristics of the two or more second components indicated by the individual feature data of the two or more second components included in the corresponding cluster. It may be configured to generate the second feature data of the corresponding cluster by converting into a quantity, and to provide the group of the generated second feature data to the information processing system.

  According to one aspect of the present disclosure, a data provision system including a storage unit, a clustering unit, a generation unit, and a provision unit may be provided. The storage unit may store a plurality of individual feature data corresponding to a plurality of components in the group.

  The clustering unit may be configured to cluster a plurality of constructs into a plurality of clusters. The generating unit may be configured to generate a group of cluster feature data corresponding to a plurality of clusters. The providing unit may be configured to provide a group of cluster feature data generated by the generating unit to the information processing system.

  Each piece of individual feature data can represent a feature of the corresponding construct. Each piece of individual feature data may be associated with an identifier of a corresponding construct. Each of the cluster feature data may include statistical data that represents the features of two or more constituents included in the corresponding cluster by a statistical amount. The statistical data may be associated with identifiers of two or more constructs included in the corresponding cluster. The generation unit may be configured to generate the statistical data by converting the features of two or more constituents indicated by the individual feature data of the two or more constituents included in the corresponding cluster into statistics.

  This data providing system provides the information processing system with the characteristics of the constructs in the second group as information-protected statistical data, but also associates the second identifier with the statistical data. The structure of the second group corresponding to the data can be specifically identified. Therefore, the data providing system can provide meaningful information useful for data combination to the information processing system.

  According to one aspect of the present disclosure, the clustering unit may cluster a plurality of constructs into a plurality of clusters in accordance with a constraint specified by the information processing system. According to one aspect of the present disclosure, the information processing system may be configured to transmit distance information representing a distance on a feature space between a plurality of components to the data providing system. The clustering unit may cluster a plurality of constructs into a plurality of clusters based on distance information from the information processing system. Based on the distance information, the data providing system can more appropriately cluster a plurality of constructs, and can provide cluster information data having high information value to the information processing system.

  According to one aspect of the present disclosure, the information processing system may be configured to transmit classification information representing each classification of the plurality of constituents to the data providing system. The clustering unit may cluster a plurality of constituents into a plurality of clusters based on classification information from the information processing system so that different classification constituents are not mixed in one cluster. Based on the classification information, the data providing system can more appropriately cluster a plurality of components.

  According to one aspect of the present disclosure, the information processing system may be configured to transmit a list of constructs to the data providing system. The data providing system may include an identification unit configured to identify a plurality of corresponding components corresponding to any of the components included in the list from among the plurality of components. The clustering unit may cluster a plurality of corresponding constituents identified by the identification unit into a plurality of clusters.

  According to one aspect of the present disclosure, an information processing method executed by a computer may be provided. The information processing method acquires a group of first feature data related to a first group including a plurality of first constituents and a second feature related to a second group including a plurality of second constituents Acquiring a group of data, and combining the acquired group of first feature data and the acquired group of second feature data may be included.

  Each of the first feature data can represent a feature of one or more corresponding first components. Each of the first feature data may be associated with a first identifier that is an identifier of one or more corresponding first components.

Each of the second feature data may correspond to each of a plurality of clusters in the second group. Each of the plurality of clusters may include two or more of the plurality of second structures. Each of the second feature data may include statistical data that represents the features of two or more second constituents included in the corresponding cluster by a statistical amount. The statistical data may be associated with a second identifier that is an identifier of two or more second structures included in the corresponding cluster.

  Combining is based on a first identifier associated with the first feature data, and combining each of the first feature data with statistical data of the second feature data associated with the corresponding second identifier. As such, it may include combining a group of first feature data and a group of second feature data.

  According to one aspect of the present disclosure, obtaining the group of first feature data includes reading the group of first feature data from a storage device that stores the group of first feature data. Also good. Acquiring a group of second feature data may include acquiring a group of second feature data from a data providing system that provides the group of second feature data. According to this method, the same effect as the information processing system described above can be obtained.

  According to one aspect of the present disclosure, a data providing method executed by a computer may be provided. The data providing method includes acquiring a plurality of individual feature data corresponding to each of a plurality of constituents in a group, clustering the plurality of constituents into a plurality of clusters, and cluster features corresponding to the plurality of clusters. Generating a group of data and providing the generated group of cluster feature data to an information processing system may be included.

  Each piece of individual feature data can represent a feature of the corresponding construct. Each piece of individual feature data may be associated with an identifier of a corresponding construct. Each of the cluster feature data may include statistical data that represents the features of two or more constituents included in the corresponding cluster by a statistical amount. The statistical data may be associated with identifiers of two or more constructs included in the corresponding cluster.

  Generating may include generating statistical data by converting features of two or more constituents indicated by individual feature data of two or more constituents included in the corresponding cluster into statistics.

  According to one aspect of the present disclosure, clustering may include clustering a plurality of constructs into a plurality of clusters according to a constraint specified by the information processing system.

  According to one aspect of the present disclosure, clustering includes clustering a plurality of constructs into a plurality of clusters based on distance information representing a distance in a feature space between the plurality of constructs from the information processing system. May be included.

  According to one aspect of the present disclosure, clustering is based on classification information representing each classification of a plurality of components from the information processing system, and a plurality of components are classified into one cluster. Clustering into a plurality of clusters may be included so as not to be mixed.

  According to one aspect of the present disclosure, obtaining the plurality of individual feature data may include reading the plurality of individual feature data from a storage device that stores the plurality of individual feature data.

According to one aspect of the present disclosure, a computer program for causing a computer to function may be provided as at least one of the acquisition unit and the coupling unit included in the information processing system described above. According to an aspect of the present disclosure, a computer program for causing a computer to function may be provided as at least one of a clustering unit, a generation unit, an identification unit, and a provision unit included in the data provision system described above.

  According to one aspect of the present disclosure, a computer program for causing a computer to execute the information processing method described above may be provided. According to one aspect of the present disclosure, a computer program for causing a computer to execute the above-described data providing method may be provided. According to one aspect of the present disclosure, a computer-readable non-transitory recording medium storing the above-described computer program may be provided.

It is a block diagram showing the structure of the data processing system of 1st embodiment. FIG. 2A is a diagram illustrating the configuration of the first database, and FIG. 2B is a diagram illustrating the configuration of the second database. It is a flowchart of the process performed by a coupling | bonding system. It is a flowchart of the process performed with a data provision system. It is explanatory drawing regarding the production | generation of cluster characteristic data. It is a figure showing the structure of the database after a process based on a 2nd database. It is a figure showing the structure of a joint database. FIG. 8A is a diagram illustrating a specific example of cluster feature data, and FIG. 8B is a diagram illustrating a specific example of a combined database. It is a block diagram showing the structure of the data processing system of 2nd embodiment. FIG. 10A is a diagram illustrating a configuration of a third database, and FIG. 10B is a diagram illustrating a configuration of a post-processing database based on the third database. It is a figure showing the structure of the extended joint database. FIG. 12A is a diagram illustrating the configuration of the second database in the third embodiment, and FIG. 12B is a diagram illustrating the configuration of the post-processing database in the third embodiment. It is explanatory drawing regarding a correspondence table. It is a figure showing the structure of the joint database of 3rd embodiment. It is a figure showing the structure of the list of common consumers with segment information. It is a flowchart of the process performed with the data provision system of 4th embodiment. It is a flowchart of the process performed with the coupling | bonding system of 5th embodiment. It is a flowchart of the process performed with the data provision system of 5th embodiment. It is a flowchart of the process performed with the coupling | bonding system of 6th embodiment. It is a flowchart of the process performed with the data provision system of 6th embodiment.

  Hereinafter, exemplary embodiments of the present disclosure will be described with reference to the drawings.

[First embodiment]
As shown in FIG. 1, the data processing system 1 of the first embodiment includes a coupling system 10 and a data providing system 30. In this data processing system 1, the combined database 155 is based on the first database 151 regarding the first group of consumers included in the combined system 10 and the second database 351 regarding the second group of consumers included in the data providing system 30. Generated.

  The combined system 10 includes a processor (CPU) 11, a memory 13, and a storage device 15. The coupling system 10 includes a communication interface (not shown) and is communicably connected to the data providing system 30 through the network NT. The processor 11 executes processing according to the computer program stored in the storage device 15. The memory 13 includes a ROM and a RAM. The storage device 15 includes a first database 151.

  As shown in FIG. 2A, the first database 151 has characteristic data F1 of the corresponding consumer individual for each consumer of the first group. The first group of consumers corresponds to the constituents of the first group. In the following, individual consumer feature data is expressed as individual feature data, and individual feature data F1 of the first database 151 is expressed as first individual feature data F1. According to FIG. 2A, the first individual feature data F1 is represented by one line of data. The first individual characteristic data F1 represents the characteristic of one corresponding consumer in association with the consumer's identification code. Specifically, the first individual characteristic data F1 represents the corresponding consumer characteristics by a plurality of parameters X1, X2, and X3.

  According to FIG. 2A, the first individual feature data F1 represents the corresponding consumer feature with three parameters X1, X2, X3. However, this is only a simplified example in view of the drawing size. The first individual feature data F1 may represent consumer features with more than three parameters X1, X2, X3.

  In FIG. 2A, the values of the multiple parameters X1, X2, and X3 of each consumer are represented by “PE”. This expression “PE” indicates that the corresponding value is personal information to be protected. The meaning of the expression “PE” is the same in other drawings. Examples of the plurality of parameters X1, X2, and X3 include parameters related to demographic attributes such as the age, sex, and residential area of the corresponding consumer, as well as parameters related to the consumption behavior of the corresponding consumer. Examples of the parameters related to the consumption behavior include information such as the store used, the purchased product, the purchase time, the number of purchases, and the purchase price.

  The storage device 15 further stores a combined database 155 generated by processing executed by the processor 11 (details will be described later).

  As shown in FIG. 1, the data providing system 30 includes a processor 31, a memory 33, and a storage device 35. The data providing system 30 includes a communication interface (not shown) and is communicably connected to the coupling system 10 through the network NT.

  The processor 31 executes processing according to the computer program stored in the storage device 35. The memory 33 includes a ROM and a RAM. The storage device 35 stores a second database 351.

  The second database 351 has individual characteristic data F2 for each consumer of the second group. The second group of consumers corresponds to the second group of entities. Hereinafter, the individual feature data F2 included in the second database 351 is expressed as second individual feature data F2. According to FIG. 2B, the second individual feature data F2 represents the corresponding feature of one consumer in association with the consumer's identification code, like the first individual feature data F1. Specifically, the second individual feature data F2 represents the corresponding consumer features by a plurality of parameters Y1, Y2, Y3. The number of parameters is not limited as in the first individual feature data F1.

The parameters Y1, Y2, Y3 are parameters X1, X in the first individual feature data F1
2 and X3 include at least partially different parameters. For example, the parameters Y1, Y2, and Y3 may include parameters related to consumer consumption behavior related to products different from the first individual feature data F1.

  The second group of consumers includes some of the same consumers as the first group of consumers. Hereinafter, consumers belonging to both the first group and the second group, that is, consumers common to the first group and the second group are expressed as a common consumer. The first individual feature data F1 and the second individual feature data F2 in the range indicated by the symbol C in FIGS. 2A and 2B correspond to the first individual feature data F1 and the second individual feature data F2 of the common consumer.

  As can be understood from FIGS. 2A and 2B, in the first database 151 and the second database 351, a common identification code is used as a consumer identification code. That is, the same identification code is associated with the first individual feature data F1 and the second individual feature data F2 of the same consumer.

  In order to share the identification code in this way, the company that owns the first database 151 or the operating company of the combined system 10 can issue the identification code to the company that owns the second database 351. Examples of the identification code include a cookie ID. By using the cookie technology or the like to track the consumer terminal device, the identification code can be shared.

  Next, details of processing executed by the combined system 10 and the data providing system 30 will be described. When an instruction for generating the combined database 155 is input from the operator of the combined system 10 through a user interface (not shown), the processor 11 of the combined system 10 executes a combined database generation process shown in FIG. When receiving the request signal from the combined system 10, the processor 31 of the data providing system 30 executes a request receiving process shown in FIG.

  In the combined database generation process, the processor 11 of the combined system 10 reads the first database 151 included in the combined system 10 from the storage device 15 among the first database 151 and the second database 351 to be combined (S110). Furthermore, a common consumer between the first group and the second group corresponding to the combination target is identified (S120).

  Specifically, the processor 11 transmits a list request signal for requesting a consumer list of the second group to the data providing system 30 in order to identify a common consumer (S121). The group consumer list can be received (S123).

  As shown in FIG. 4, when the processor 31 of the data providing system 30 receives the list request signal from the combined system 10 (Yes in S210), the processor 31 refers to the second database 351 and stores the second individual feature data in the second database 351. The consumer list of the second group in which F2 is registered is transmitted to the combined system 10 (S215). The consumer list of the second group includes the identification code of the consumer of the second group, but does not include consumer personal information.

  After receiving this consumer list, the processor 11 of the combined system 10 receives the identification code of the second group of consumers indicated by the received consumer list and the consumption of the first group indicated by the first database 151 read in S110. The common consumer between the first group and the second group can be identified by comparing the identification code of the person (S125).

When the processor 11 identifies the common consumer in S120, the processor 11 transmits a database request signal to the data providing system 30 (S130). The database request signal is accompanied by a list of common consumers identified in S120. The common consumer list includes identification codes of common consumers included in the first group.

  As shown in FIG. 4, when the processor 31 of the data providing system 30 receives the database request signal from the combined system 10 (Yes in S220), the second database 351 is based on the list of common consumers attached to the database request signal. The second individual characteristic data F2 of the common consumer is identified (S230).

  Thereafter, the processor 31 clusters the common consumers into a plurality of clusters based on the second individual feature data F2 of the common consumers (S240). Clustering can be performed based on, for example, the k-means method (k-means method) or other known techniques.

  Clustering is performed so that the number of consumers (that is, cluster size) included in one cluster is equal to or greater than a predetermined number. The predetermined number is determined from the viewpoint of personal information protection. Clustering is performed based on the distribution of common consumers when common consumers are arranged on the feature space so that a plurality of consumers that are close to each other on the feature space are combined into one cluster.

  At the time of clustering, the processor 31 can obtain the distance Dy [i, j] between common consumers by the Euclidean distance. The distance Dy [i, j] between the consumer i and the consumer j can be calculated according to the following equation.

Dy [i, j] = {Σ (yn [i] −yn [j]) ² } ^1/2
Here, Σ (yn [i] −yn [j]) ² is the sum of (yn [i] −yn [j]) ² from n = 1 to n = N. The value N corresponds to the number of parameters Y1, Y2, Y3 representing consumer characteristics in the second individual characteristic data F2. According to FIG. 2B, N = 3.

  yn [i] is the value of parameter Yn for consumer i. yn [j] is a value of the parameter Yn of the consumer j. In the left area of FIG. 5, a second database 351 is shown. Here, the values of the parameters Y1 and Y2 of the second individual feature data F2 relating to the consumer i are represented by y1 [i] and y2 [i]. (I = 1, 2, 3,...).

  If the common consumer in the second group is clustered into a plurality of clusters in S240, the processor 31 moves to S250, and for each of the plurality of clusters, the second individual feature of the plurality of consumers belonging to the corresponding cluster. One cluster feature data FC2 is generated by integrating the data F2.

  Specifically, the processor 31 converts the values of the plurality of parameters Y1, Y2, Y3 indicated by the second individual feature data F2 of the plurality of consumers belonging to the corresponding cluster into one statistic ST for each parameter, One cluster feature data FC2 corresponding to this cluster is generated.

  The right region of FIG. 5 illustrates a group of cluster feature data FC2 generated based on the group of second individual feature data F2 in the second database 351 shown in the left region of FIG.

  According to the example shown in FIG. 5, the consumers with the identification codes ID001, ID002, and ID003 are combined into one cluster, and one cluster feature data FC2 is generated based on the second individual feature data F2 of these consumers. The

According to this example, the values y1 [1], y1 [2], y1 [3] of the parameter Y1 of the identification codes ID001, ID002, ID003 are one statistic ST {y1 [1], y1 [2].
, Y1 [3]}. Here, the expression ST {} may be understood to mean the statistic of the value in parentheses {}. Similarly, the values y2 [1], y2 [2], y2 [3] of the parameter Y2 are converted into one statistic ST {y2 [1], y2 [2], y2 [3]}. Similarly, the values y3 [1], y3 [2], y3 [3] of the parameter Y3 are converted into one statistic ST {y3 [1], y3 [2], y3 [3]}.

  As a result, the second individual feature data F2 of the consumers with the identification codes ID001, ID002, and ID003 belonging to the same cluster is obtained by using the statistics ST {y1 [1] of the parameters Y1, Y2, and Y3 of the consumers that belong to the same cluster. ], Y1 [2], y1 [3]}, ST {y2 [1], y2 [2], y2 [3]}, ST {y3 [1], y3 [2], y3 [3]} It is converted into one cluster feature data FC2 having the statistical data FS2. The cluster feature data FC2 includes identification codes ID001, ID002, and ID003 of consumers belonging to the corresponding cluster in association with the statistical data FS2.

  The statistic ST can be determined in advance for each type of parameters Y1, Y2, Y3. Examples of the statistic ST include ratio, average value, median value, maximum value, standard deviation, and variance. The average value, median value, and maximum value correspond to the representative value of the cluster. The statistic ST may be a combination of a plurality of different statistics.

  For example, when the parameter represents gender, the statistic ST may represent the ratio of males and / or females in the cluster. When the parameter represents age, the statistic ST may represent the average age within the cluster. When the parameter represents product purchase experience, the statistic ST can be the proportion of consumers who have product purchase experience in the cluster. When the parameter represents the number of products purchased, the statistic ST can be one or a combination of the average value, median value, and maximum value of the number of products purchased.

  Similarly, according to FIG. 5, the consumers of the identification codes ID004, ID005, ID006, and ID007 are grouped into one cluster, and one cluster feature data FC2 that represents these consumer features as a statistic ST is generated. The Similarly, according to FIG. 5, the consumers of the identification codes ID008, ID009, and ID0010 are combined into one cluster, and one cluster feature data FC2 that expresses the features of these consumers as a statistic ST is generated.

  After the processing in S250, the processor 31 clusters the non-common consumers in the second group into a plurality of clusters (S260). In this way, in this embodiment, non-common consumers in the second group are clustered separately from common consumers so that common consumers and non-common consumers are not mixed in one cluster. To do.

  Clustering in S260 is also performed in the same manner as S240. That is, clustering is performed so that the number of consumers included in one cluster is equal to or greater than a predetermined number. Clustering is performed so that a plurality of consumers that are close to each other in the feature space are combined into one cluster based on the distribution of the non-common consumers when the non-common consumers are arranged on the feature space.

  After completing the processing in S260, the processor 31 proceeds to S270, and generates one cluster feature data FC2 for each of the plurality of clusters related to the non-common consumer. The generation of the cluster feature data FC2 in S270 is performed in the same manner as the processing in S250.

According to the example of FIG. 5, the non-common consumers with the identification codes ID015 and ID016 are grouped into one cluster, and one cluster feature data FC2 that expresses the features of these consumers as a statistic ST is generated. Similarly, the consumers of the identification codes ID017 and ID0018 are collected into one cluster, and one cluster feature data FC2 that represents the characteristics of these consumers as a statistic ST is generated.

  In the present embodiment, the plurality of second individual feature data F2 are integrated to generate one cluster feature data FC2 in order to protect consumer personal information. Therefore, the number of clusters is preferably greater than two. Here, in order to simplify the explanation and illustration, only an example in which the number of consumers in the cluster is two is shown.

  When the processing in S270 is completed, the processor 31 transmits the group of common consumer and non-common consumer cluster feature data FC2 generated in S250 and S270 as the post-processing database FP2 to the combined system 10 of the database request source. (S280).

  FIG. 6 shows an example of the post-processing database FP2. As can be understood from FIG. 6, the post-processing database FP2 has cluster feature data FC2 for each cluster for each of the common consumer and the non-common consumer. In FIG. 6, the values of the plurality of parameters Y1, Y2, Y3 in the cluster feature data FC2 are represented by “ST”. “ST” indicates that the value of the corresponding parameter is the statistic ST. A thick solid line in FIG. 6 indicates a boundary between a common consumer and a non-common consumer.

  When the processor 11 of the combined system 10 receives the processed database FP2 transmitted from the data providing system 30 in response to the database request signal transmitted in S130 (S140), the process proceeds to S150. In S150, the processor 11 generates the combined database 155 by combining the first database 151 and the processed database FP2 received from the data providing system 30 (S150). FIG. 7 shows an example of the combined database 155.

  In S150, based on the identification code associated with the first individual feature data F1, the processor 11 combines each of the first individual feature data F1 with the statistical data FS2 of the cluster feature data FC2 associated with the corresponding identification code. As described above, the group of first individual feature data F1 and the group of cluster feature data FC2 are combined to generate a combined database 155.

  That is, in S150, each of the first individual feature data F1 is combined with the statistical data FS2 of the cluster feature data FC2 associated with the same identification code, and the combined database 155 is generated. However, the first individual feature data F1 of the non-common consumer in the first group is registered in the combined database 155 without being combined with the cluster feature data FC2. Similarly, the non-common consumer cluster feature data FC2 in the second group is registered in the combined database 155 without being combined with the first individual feature data F1. In this way, the combined database 155 is generated.

  In addition, as shown in FIG. 8A, each of the cluster feature data FC2 provided from the data providing system 30 has, as the statistical data FS2, the statistical data FSI of a plurality of consumers belonging to the corresponding cluster. May be. In this case, the statistical data FSI of each consumer in the corresponding cluster is the same data copied from one statistical data, and shows the same statistical quantity.

  In this case, in the combined database 155, as shown in FIG. 8B, the first individual feature data F1 and the statistical data FSI of the corresponding cluster feature data FC2 are combined for each consumer.

  According to the example shown in FIG. 7, for the sake of simplicity of illustration and description, one cluster is grouped with continuous identification codes. However, the clustering is performed so that a plurality of consumers having a short distance in the feature space, that is, similar features, are combined into one cluster. Therefore, a plurality of consumers belonging to one cluster are usually distributed in the first database 151 and the second database 351.

  If the statistical data FSI is prepared for each consumer, it is not necessary to rearrange the consumer array in the post-processing database FP2 and the combined database 155 so that a plurality of consumers belonging to one cluster are adjacent to each other. is there.

  In this embodiment, the data providing system 30 can provide consumer-related data with a consumer identification code to the coupling system 10 while protecting personal information, and the coupling system 10 is accurate based on the identification code, Multiple types of data about the same consumer can be combined to generate a meaningful combined database 155 about the consumer. Thus, the combined database 155 is very useful for companies that can refer to it to analyze consumer behavior.

[Second Embodiment]
Then, the data processing system 2 of 2nd embodiment is demonstrated. The data processing system 2 of this embodiment is configured such that a further data providing system 50 is provided in the data processing system 1 of the first embodiment. As shown in FIG. 9, the data processing system 2 includes a further data providing system 50 in addition to the combination system 10 and the data providing system 30. Hereinafter, the data providing system 30 is expressed as the first data providing system 30, and the data providing system 50 is expressed as the second data providing system 50. Below, the description regarding the structure similar to 1st embodiment is abbreviate | omitted suitably.

  The second data providing system 50 includes a processor 51, a memory 53, and a storage device 55. The second data providing system 50 is communicably connected to the coupling system 10 through the network NT. The processor 51 executes processing according to the computer program stored in the storage device 55. The storage device 55 stores a third database 551 related to the third group of consumers.

  As shown in FIG. 10A, the third database 551 has individual feature data F3 for each consumer of the third group. Hereinafter, the individual feature data F3 included in the third database 551 is expressed as third individual feature data F3. The third individual characteristic data F3 represents the characteristic of a corresponding one consumer in association with the consumer's identification code. Specifically, the third individual feature data F3 represents the corresponding consumer features by a plurality of parameters Z1, Z2, and Z3. The number of parameters is not limited as in the first and second individual feature data F1, F2.

  The parameters Z1, Z2, and Z3 are parameters that are at least partially different from the parameters X1, X2, and X3 of the first individual feature data F1 and the parameters Y1, Y2, and Y3 of the second individual feature data F2. The parameters Z1, Z2, and Z3 may include parameters related to consumer consumption behavior related to products different from the first individual feature data F1 and the second individual feature data F2.

Moreover, as can be understood from FIG. 10A, the common consumer between the first group and the third group is different from the common consumer between the first group and the second group. According to the example shown in FIG. 2B, the common consumers between the first group and the second group are consumers from the identification code ID001 to the identification code ID010, whereas according to the example shown in FIG. 10A. The common consumer between the first group and the third group has an identification code ID005 as shown in range C.
To the identification code ID014.

  In this embodiment, the processor 11 of the combined system 10 obtained from the second data providing system 50 into the combined database 155 based on the first database 151 and the processed database FP2 obtained from the first data providing system 30. The post-processing database FP3 based on the third database 551 is combined to extend the combined database 155. For that purpose, the processor 11 also executes the processing of S120 to S140 shown in FIG. 3 for the second data providing system 50, and acquires the post-processing database FP3 from the second data providing system 50.

  Similarly to the first data providing system 30, the processor 51 of the second data providing system 50 executes the processing shown in FIG. 4 and when the list request signal is received from the combined system 10 (Yes in S210), Three groups of consumer lists are provided to the combined system 10 (S215).

  When the processor 51 of the second data providing system 50 receives the database request signal from the combined system 10 (Yes in S220), the third individual characteristic data of the common consumer between the first group and the third group. F3 is identified (S230), and based on the third individual feature data F3, common consumers between the first group and the third group are clustered into a plurality of clusters (S240), and cluster feature data for each cluster. FC3 is generated (S250). The cluster feature data FC3 is shown in FIG. 10B.

  The clustering method and the cluster feature data FC3 generation method in the second data providing system 50 are the same as those in the first data providing system 30. That is, for each cluster, the processor 51 converts the values of the plurality of parameters Z1, Z2, and Z3 indicated by the third individual feature data F3 of the plurality of consumers belonging to the corresponding cluster into one statistic ST for each parameter. Thus, one cluster feature data FC3 corresponding to this cluster is generated (S250). The cluster feature data FC3 includes statistical data FS3 including the statistical amount ST for each parameter in association with identification codes of a plurality of consumers belonging to the cluster.

  When a group of cluster feature data FC3 related to the common consumer is generated in this way (S250), the processor 51 further clusters non-common consumers with the first group in the third group into a plurality of clusters (S260), For each cluster, one corresponding cluster feature data FC3 is generated (S270). Then, a group of the cluster feature data FC3 of the common consumer and the non-common consumer is transmitted to the combined system 10 as the processed database FP3 (S280). The thick solid line shown in FIG. 10B indicates the boundary between the common consumer and the non-common consumer between the first group and the third group in the processed database FP3.

  Upon receiving the processed database FP3 from the second data providing system 50 (S140), the processor 11 of the combined system 10 expands the combined database 155 based on the processed database FP3 (S150). That is, the processor 11 expands the combined database 155 by further combining the processed database FP3 based on the third database 551 with the combined database 155 obtained by combining the processed database FP2 with the first database 151 (S150). .

An example of an extended combined database 155 is shown in FIG. In this combined database 155, as shown in FIG. 11, consumer characteristics are represented by a number of parameters. And the 2nd group and the 3rd group have a consumer common between a 2nd group and a 3rd group in a non-common consumer with a 1st group. These consumer statistical data FS2, FS3 are combined in an extended combined database 155. Thus, the expanded combined database 155 is very useful for analyzing consumer behavior.

[Third embodiment]
Subsequently, a third embodiment will be described. The data processing system of the third embodiment corresponds to a modification of the data processing system 1 of the first embodiment. Below, the description regarding the structure similar to 1st embodiment is abbreviate | omitted suitably.

  According to the third embodiment, the first group of consumers is assigned the same identification code as the first embodiment (hereinafter referred to as the first identification code), whereas the second group of consumers. Is assigned a second identification code of a different type from the first identification code. Therefore, in the second database 351 of the present embodiment shown in FIG. 12A, the second individual feature data F4 in the range C is the same consumer data as the first individual feature data F1 shown in the range C in FIG. 2A. However, it does not have the same identification code as the first group. The second identification code illustrated in FIG. 12A exhibits a certain regularity with the first identification code illustrated in FIG. 2A, but this regularity is convenient for easy understanding of the correspondence. Usually, there is no regularity between the first identification code and the second identification code.

  For this reason, in this embodiment, a correspondence table showing the correspondence between the first identification code and the second identification code is prepared, and based on this correspondence table, the common consumption between the first group and the second group is prepared. Is identified. The exemplary correspondence table shown in FIG. 13 has a configuration in which a corresponding second identification code is written for each first identification code.

  For example, the combined system 10 can acquire a correspondence table from an external system. For example, if the first identification code and the second identification code are different website or online store user IDs, these user IDs are linked through the terminal ID and cookie ID of the user terminal device that uses the website. The external system can monitor the wide area network (Internet) to which the user terminal device is connected, and generate a correspondence table based on the terminal ID and the cookie ID. Alternatively, the correspondence table may be acquired from a company that provides ID linkage service or a company that issues valuable points with many affiliated companies.

  In the present embodiment, the processor 11 of the combined system 10 executes the processing shown in FIG. 3, and in S120, acquires a correspondence table from the external system, and based on the acquired correspondence table, the first group and the second group Can identify common consumers. The processor 11 can store the acquired correspondence table in the memory 13 or the storage device 15.

  The processor 11 can transmit the database request signal with the list of the second identification codes of the common consumers to the data providing system 30 (S130). Thereby, the data provision system 30 can identify the common consumer between the first group and the second group based on the second identification code in the list (S230).

  Then, the data providing system 30 may perform clustering on the identified common consumer (S240), and generate and transmit the processed database FP4 shown in FIG. 12B having the cluster feature data FC4 associated with the second identification code. Yes (S250, S280). The post-processing database FP4 is different from the post-processing database FP2 of the first embodiment in that the non-common consumer cluster feature data is not included.

When the processor 11 of the combined system 10 acquires the post-processing database FP4 from the data providing system 30 (S140), the combined database 155 is generated with reference to the correspondence table. That is, referring to the correspondence table, each piece of the first individual feature data F1 is statistical data of the cluster feature data FC4 associated with the second identification code corresponding to the first identification code included in the first individual feature data F1. Combine with FS4 to generate a combined database 155.

  FIG. 14 shows a combined database 155 generated in the present embodiment. This combined database 155 does not have cluster feature data regarding non-common consumers. The configuration of the combined database 155 may be applied to the first and second embodiments. In this case, the data providing systems 30 and 50 in the first embodiment and the second embodiment may be configured not to provide the cluster characteristic data of non-common consumers to the combined system 10.

  According to the third embodiment, even when different identification codes are used between the first group and the second group, the first individual feature data F1 of the same consumer and the cluster feature data FC4 are appropriately displayed. Combined, a meaningful combined database 155 can be generated.

[Fourth embodiment]
Subsequently, a fourth embodiment will be described. The data processing system of the fourth embodiment corresponds to a modification of the data processing system 1 of the first embodiment. Below, the description regarding the structure similar to 1st embodiment is abbreviate | omitted suitably.

  According to the fourth embodiment, the processor 11 of the combined system 10 transmits the database request signal with the common consumer list L4 shown in FIG. 15 to the data providing system 30 in S130. The list L4 shown in FIG. 15 is configured to have segment information representing each affiliation segment of the common consumer. The segment information corresponds to classification information representing each classification of common consumers. Specifically, the list L4 has a configuration in which identification codes of common consumers belonging to the corresponding segment are written for each segment.

  According to FIG. 15, the first segment (SEG1), the second segment (SEG2), and the third segment (SEG3) are defined in the list L4. For each segment, the common consumers belonging to the corresponding segment The identification code is described.

  The segment information is used by the combining system 10 to control clustering in the data providing system 30. The clustering control mentioned here includes controlling clustering in the data providing system 30 so that consumers of different segments are not mixed in one cluster.

  In S <b> 130, the processor 11 can classify the common consumers in the first group into a plurality of segments according to a predetermined rule. For example, when the gender of the common consumer can be determined by referring to the first individual characteristic data F1, the processor 11 includes the list L4 so that common consumers of different genders are not mixed in the same cluster. Male and female segments can be defined. That is, in S130, the processor 11 can generate a list L4 describing the identification code of the corresponding common consumer for each segment for each segment, and transmit the list L4 to the data providing system 30.

Alternatively, when the residential area of the common consumer can be determined by referring to the first individual characteristic data F1, the processor 11 sets the residential area so that common consumers in different residential areas are not mixed in the same cluster. Each segment can be defined. That is, in S130, the processor 11 can generate a list L4 describing the identification code of the common consumer in the corresponding residential area for each segment, and transmit the list L4 to the data providing system 30. To protect consumer personal information, list L4 is generated so that it does not contain segment details, ie, information that can identify specific attributes (gender and residential area) of consumers within the segment. Can be done.

  In S240 and S250, the processor 31 of the data providing system 30 that receives the common consumer list L4 with segment information can execute the process shown in FIG. 16 instead of the process shown in FIG. In the process shown in FIG. 16, based on the list L4, common consumers are clustered into a plurality of clusters for each segment.

  According to the example shown in FIG. 16, the processor 31 first identifies the belonging segment of the common consumer in the second group based on the received list L4 (S241). Thereafter, one of the plurality of segments is selected as a processing target segment (S243), and the common consumers belonging to the selected processing target segment are clustered into a plurality of clusters (S245). After the processing in S245, the processor 31 generates one cluster feature data FC2 for each of the plurality of clusters in the processing target segment based on the second individual feature data F2 of the plurality of consumers belonging to the corresponding cluster. (S247).

  Thereafter, the processor 31 determines whether or not the processes of S245 and S247 have been executed for all segments (S249). If it is determined that all segments have not been executed (No in S249), the processor 31 selects one of the unselected segments as the processing target segment as a new processing target segment (S243), and the processing of S245 and S247 Execute the process. If it is determined that processing has been executed for all segments (Yes in S249), the processing shown in FIG. 16 is terminated.

  According to the present embodiment, the combination system 10 controls the clustering of the data providing system 30 so that the desired information is not lost when the cluster feature data FC2 is generated by meaningfully using the information in the first database 151. can do. Therefore, it is possible to generate a combined database 155 that is more suitable for data analysis.

  This embodiment includes the idea that the combined system 10 controls the clustering of the data providing system 30 and the idea that the data providing system 30 clusters a plurality of consumers according to the constraints specified by the combined system 10.

  Thus, as a further variation, the combined system 10 may be configured to transmit a database request signal requesting clustering by gender and clustering by residential area without providing segment information. . In this case, the processor 31 of the data providing system 30 determines the segment (gender or residential area) of the common consumer and the non-common consumer based on the second individual characteristic data F2 of the second database 351, and determines each segment. Can be performed in response to a request from the combined system 10.

[Fifth embodiment]
Next, a fifth embodiment will be described. The data processing system of the fifth embodiment corresponds to a modification of the data processing system 1 of the first embodiment. Below, the description regarding the structure similar to 1st embodiment is abbreviate | omitted suitably.

  According to the fifth embodiment, the processor 11 of the combined system 10 executes the process shown in FIG. 17 in S130, thereby sending the database request signal with the common consumer list L5 including the distance information to the data providing system 30. Send to.

Specifically, the processor 11 calculates the distance Dx [i, j] on the feature space between the common consumers by referring to the first individual feature data F1 of the first database 151 (S131). The distance Dx [i, j] indicates the distance between the consumer i and the consumer j based on the first individual feature data F1, and this distance is the similarity of the feature between the consumer i and the consumer j. Corresponds to the degree.

  The processor 11 can obtain the distance Dx [i, j] on the feature space between the common consumers by the Euclidean distance. The distance Dx [i, j] between the consumer i and the consumer j can be calculated according to the following equation.

Dx [i, j] = {Σ (xn [i] −xn [j]) ² } ^1/2
Here, Σ (xn [i] −xn [j]) ² is the sum of (xn [i] −xn [j]) ² from n = 1 to n = M. The value M corresponds to the number of parameters X1, X2, and X3 that define consumer characteristics in the first individual characteristic data F1. According to FIG. 2A, M = 3. xn [i] is the value of the parameter Xn of the consumer i. xn [j] is a value of the parameter Xn of the consumer j. In S131, the distance Dx [i, j] is calculated for all combinations of the consumer i and the consumer j corresponding to the common consumer.

  After that, the processor 11 generates a list L5 with distance information, which is the common consumer list L5, and adds the information of the distance Dx calculated in S131 to the common consumer identification code (S133). According to the example shown in FIG. 17, in the list L5, the distance Dx [i, j] for each combination with the consumer i is associated with the identification code of the consumer j.

  The processor 11 transmits the list L5 of common consumers with distance information generated in S133 in this way to the data providing system 30 (S135).

  On the other hand, the processor 31 of the data providing system 30 that receives the list L5 with distance information can execute the process shown in FIG. 18 in S240. That is, the processor 31 calculates a distance Dy [i, j] between consumers to be clustered (that is, between common consumers) based on the second individual feature data F2. The distance Dyi, j] can be calculated by a method similar to the method described in the first embodiment (S310).

Thereafter, the processor 31 calculates a composite distance D [i, j] between consumers according to the following equation (S320). The composite distance D [i, j] is a composite distance between the consumer i and the consumer j.
D [i, j] = (Dx [i, j] ² + D2 [i, j] ² ) ^1/2

  After the processing in S320, the processor 31 collects consumers having a short synthetic distance D [i, j into one cluster based on the distribution of consumers to be clustered specified from the synthetic distance D [i, j]. As described above, the consumers to be clustered are clustered into a plurality of clusters (S330).

  According to the fifth embodiment, the combining system 10 provides distance information between common consumers specified from the first database 151, that is, similarity information, as the data providing system 30 while keeping the personal information of the first database 151 secret. So that the data providing system 30 can perform better clustering. Therefore, the combined database 155 with higher information value can be generated while protecting personal information.

[Sixth embodiment]
Subsequently, a sixth embodiment will be described. The data processing system of the sixth embodiment corresponds to a modification of the data processing system 1 of the first embodiment. Below, the description regarding the structure similar to 1st embodiment is abbreviate | omitted suitably.

According to the sixth embodiment, the processor 11 of the combined system 10 clusters common consumers into a plurality of clusters, provides the cluster information to the data providing system 30, and acquires the cluster feature data FC2 for each cluster. Configured. That is, the sixth embodiment is different from the first embodiment in that the subject of clustering related to common consumers is changed to the combined system 10 instead of the data providing system 30.

  Specifically, the combined system 10 in the present embodiment is configured such that the processor 11 executes a combined database generation process shown in FIG. 19 instead of the process shown in FIG. The data providing system 30 is configured such that the processor 31 executes the processing from S530 to S580 shown in FIG. 20 instead of the processing from S230 to S280 shown in FIG.

  When starting the processing shown in FIG. 19, the processor 11 of the combined system 10 reads the first database 151 from the storage device 15 as in the processing in S110 (S410). Furthermore, the common consumer between groups is identified similarly to the process in S120 (S420).

  Thereafter, the processor 11 clusters the common consumers into a plurality of clusters based on the first individual characteristic data F1 of the common consumers (S425). Here, except for using the first individual feature data F1 instead of the second individual feature data F2, the common consumers can be clustered by the same method as the process in S240.

  After the processing in S425, the processor 11 transmits a database request signal including a list of the second identification codes of the common consumers including the cluster information to the data providing system 30 (S430). The cluster information is configured so that the data providing system 30 can identify the cluster to which each common consumer belongs.

  When receiving the database request signal (Yes in S220), the processor 31 of the data providing system 30 identifies the cluster to which each common consumer belongs, based on the cluster information included in the list, as shown in FIG. The second individual characteristic data F2 of each common consumer is identified on the basis of the second identification code included in (S530).

  Thereafter, the processor 31 generates, for each of the plurality of clusters identified from the cluster information, one cluster feature data FC2 obtained by integrating the second individual feature data F2 of the plurality of consumers belonging to the corresponding cluster ( S550). In S550, the same process as S250 in the first embodiment can be executed.

  When the processing in S550 is completed, the processor 31 executes the processing similar to S260 and S270 in the first embodiment in S560 and S570, and generates cluster feature data FC2 for each cluster related to non-common consumers. it can.

  Thereafter, the processor 31 transmits the group of common consumer and non-common consumer cluster feature data FC2 generated in S550 and S570 to the combined system 10 as a processed database FP2 (S580).

  The processor 11 of the combined system 10 receives the post-processing database FP2 thus transmitted from the data providing system 30 (S440), and generates a combined database 155 (S450). The processor 11 can execute the same processes as S140 and S150 in the first embodiment in S440 and S450.

Also in the sixth embodiment, the same combined database 155 as in the first embodiment can be generated. As can be understood from the sixth embodiment, the common consumer clustering can be executed by either the combined system 10 or the data providing system 30. In this regard, the technical idea that the combining system 10 provides the distance information to the data providing system 30 in the fifth embodiment is the form in which the data providing system 30 provides the distance information to the combining system 10 in the sixth embodiment. It can also be applied to.

  In this case, the processor 31 of the data providing system 30 generates the distance information of the common consumer and the non-common consumer based on the second individual feature data F2 of the second database 351, and combines the generated distance information. System 10 can be provided. The processor 11 of the combined system 10 can use this distance information to cluster common consumers into multiple clusters. In this case, the processor 11 may further perform clustering of non-common consumers using the distance information.

[Others]
As mentioned above, although exemplary embodiment of this indication was described, it cannot be overemphasized that this indication is not limited to the above-mentioned embodiment and can take various modes.

  For example, the first and second group members are not limited to consumers. One or both of the first and second groups may be at least one set of things and places related to human activity, and the members of the first and second groups are elements of these sets. There may be.

  In the above-described embodiment, the combined system 10 acquires the second group consumer list from the data providing system 30 and identifies the common consumer. However, the combined system 10 does not execute the process of S120, and in S130 The consumer list of the entire first group may be configured to be transmitted. In this case, in S230, the data providing system 30 identifies the common consumer by comparing the identification code included in the consumer list of the entire first group with the identification code of the second group of consumers. Can do. Common consumer identification can be performed in a variety of ways, and any of these methods may be employed.

  In addition, the function of one component in the above embodiment may be distributed among a plurality of components. Functions of a plurality of components may be integrated into one component. A part of the configuration of the above embodiment may be omitted. At least a part of the configuration of the embodiment may be added to or replaced with the configuration of the other embodiment. Any aspect included in the technical idea specified from the wording of the claims is an embodiment of the present disclosure.

  Finally, the correspondence between terms will be explained. The coupling system 10 corresponds to an example of an information processing system. The process in which the combined system 10 transmits the database request signal and acquires the processed databases FP2, FP3, and FP4 from the data providing systems 30 and 50 corresponds to an example of the process realized by the acquisition unit of the information processing system. A common consumer corresponds to an example of a corresponding structure. The process in which the data providing systems 30 and 50 identify the individual feature data F2 and F3 of the common consumer in S230 corresponds to an example of a process realized by the identification unit.

  DESCRIPTION OF SYMBOLS 1, 2 ... Data processing system, 10 ... Coupling system, 11 ... Processor, 13 ... Memory, 15 ... Storage apparatus, 30 ... Data provision system, 31 ... Processor, 33 ... Memory, 35 ... Storage apparatus, 50 ... Data provision system , 51 ... Processor, 53 ... Memory, 55 ... Storage device, 151 ... First database, 155 ... Combined database, 351 ... Second database, 551 ... Third database, F1, F2, F3, F4 ... Individual feature data, FC2 , FC3, FC4 ... cluster feature data, FP2, FP3, FP4 ... post-processing database, FS2, FS3, FS4, FSI ... statistical data, L4, L5 ... list.

Claims (30)

A storage unit configured to store a group of first feature data relating to a first group including a plurality of first constructs;
An acquisition unit configured to acquire a group of second feature data relating to a second group including a plurality of second constructs;
A coupling unit;
With
The first feature data is associated with a first identifier that is an identifier of the corresponding one or more first constructs, and represents the features of the corresponding one or more first constructs;
Each of the second feature data corresponds to each of a plurality of clusters in the second group, and each of the plurality of clusters includes two or more of the plurality of second constructs, The second feature data is associated with a second identifier that is an identifier of the two or more second constructs included in the corresponding cluster, and the two or more second configurations included in the corresponding cluster. It has statistical data that expresses body characteristics as statistics,
Based on the first identifier associated with the first feature data, the combining unit converts each of the first feature data to the second feature data associated with the corresponding second identifier. An information processing system configured to combine a group of the first feature data and a group of the second feature data so as to combine with the statistical data. An information processing system according to claim 1,
The acquisition unit is configured to cluster the plurality of second structures into the plurality of clusters according to a specified constraint condition in a data providing system including a plurality of individual feature data corresponding to each of the plurality of second structures. And obtaining a group of the second feature data corresponding to the plurality of clusters according to the constraint condition from the data providing system,
The data providing system clusters the plurality of second structures into the plurality of clusters according to the constraint condition, and the group of the second feature data corresponding to the plurality of clusters is stored in the information processing system. An information processing system configured to provide. An information processing system according to claim 2,
In the statistical data, the data providing system converts the features of the two or more second constituents indicated by the individual feature data of the two or more second constituents included in the corresponding cluster into statistics. Information processing system. An information processing system according to claim 1,
The plurality of second constituent bodies include a plurality of corresponding constituent bodies corresponding to any of the plurality of first constituent bodies, and a plurality of non-corresponding sections not corresponding to any of the plurality of first constituent bodies. A construct, and
Each of the second feature data acquired by the acquisition unit corresponds to each of a plurality of clusters defined by clustering the plurality of corresponding constituents in the second group. An information processing system according to claim 4,
The acquisition unit requests a data providing system having a plurality of individual feature data corresponding to each of the plurality of second structures to cluster the plurality of corresponding structures into the plurality of clusters,
In accordance with the request, the data providing system clusters the plurality of corresponding constructs into the plurality of clusters, and provides the information processing system with the second group of feature data corresponding to the plurality of clusters. An information processing system composed of An information processing system according to claim 5,
The acquisition unit transmits the list of the plurality of first structures or the list of the plurality of corresponding structures to the data providing system as a list of structures,
The data providing system is an information processing system configured to identify the plurality of corresponding constituents in the second group based on the list. An information processing system according to claim 6,
The acquisition unit transmits distance information representing a distance on a feature space between a plurality of constituents included in the list to the data providing system,
The data providing system is an information processing system configured to cluster the plurality of corresponding constituents in the second group into the plurality of clusters based on the distance information. The information processing system according to claim 6 or 7,
The acquisition unit transmits classification information representing each classification of the plurality of constituents included in the list to the data providing system,
The data providing system clusters the plurality of corresponding constituents in the second group into the plurality of clusters based on the classification information so that corresponding constituents of different classifications are not mixed in one cluster. An information processing system composed of An information processing system according to any one of claims 4 to 8,
The plurality of corresponding structural bodies in the second group are the same structural bodies as any of the plurality of first structural bodies in the first group;
Between the first group and the second group, the same identifier is assigned to each of the plurality of corresponding components,
The information processing system, wherein the combining unit combines each of the first feature data with the statistical data of the second feature data associated with the same identifier. An information processing system according to any one of claims 1 to 8,
The storage unit stores a correspondence relationship between the first identifier and the second identifier;
The information processing system, wherein the combining unit combines each of the first feature data with the statistical data of the second feature data associated with the corresponding second identifier according to the correspondence relationship. An information processing system according to any one of claims 1 to 10,
Each of the first feature data corresponds to each of a plurality of clusters in the first group, each of the plurality of clusters including two or more of the plurality of first constructs, The first feature data is a statistic indicating the features of the two or more first components included in the corresponding cluster associated with the identifiers of the two or more first components included in the corresponding cluster. Information processing system with statistical data represented by An information processing system according to claim 1,
The plurality of second constituents include a plurality of corresponding constituents corresponding to any of the plurality of first constituents,
The acquisition unit clusters the plurality of corresponding constituents into a plurality of clusters based on first feature data of the corresponding first constituents, and a cluster to which each of the plurality of corresponding constituents belongs. Is provided to a data providing system including a plurality of individual feature data corresponding to each of the plurality of second structures, and the second feature data according to the cluster information is provided from the data providing system. Get a group,
For each cluster identified from the cluster information, the data providing system statistically analyzes the characteristics of the two or more second structures indicated by the individual feature data of the two or more second structures included in the corresponding cluster. An information processing system configured to generate second feature data of the corresponding cluster by converting into a quantity, and to provide the group of the generated second feature data to the information processing system. A storage unit for storing a plurality of individual feature data corresponding to each of a plurality of components in the group;
A clustering unit configured to cluster the plurality of constructs into a plurality of clusters;
A generating unit configured to generate a group of cluster feature data corresponding to the plurality of clusters;
A providing unit configured to provide a group of the cluster feature data generated by the generating unit to an information processing system;
With
The individual feature data is associated with an identifier of a corresponding construct, and represents the feature of the corresponding construct;
The cluster feature data includes statistical data that represents the features of the two or more constructs included in the corresponding cluster in a statistic associated with identifiers of the two or more constructs included in the corresponding cluster;
The data generation system, wherein the generation unit generates the statistical data by converting the characteristics of the two or more constituents indicated by the individual feature data of the two or more constituents included in the corresponding cluster into the statistics. . 14. The data providing system according to claim 13, wherein
The data providing system, wherein the clustering unit clusters the plurality of constructs into the plurality of clusters in accordance with a constraint specified by the information processing system. 14. The data providing system according to claim 13, wherein
The information processing system is configured to transmit distance information representing a distance on a feature space between the plurality of constructs to the data providing system,
The data providing system, wherein the clustering unit clusters the plurality of constructs into the plurality of clusters based on the distance information from the information processing system. The data providing system according to claim 13 or 15,
The information processing system is configured to transmit classification information representing each classification of the plurality of constructs to the data providing system,
The clustering unit is a data providing system that clusters the plurality of constituents into the plurality of clusters based on the classification information from the information processing system so that constituents of different classifications are not mixed in one cluster. The data providing system according to claim 13 or 15,
The information processing system is configured to transmit a list of constructs to the data providing system;
The data providing system includes an identification unit configured to identify a plurality of corresponding components corresponding to any of the components included in the list from the plurality of components.
The data providing system in which the clustering unit clusters the plurality of corresponding constituents identified by the identification unit into the plurality of clusters. A storage unit for storing a plurality of individual feature data corresponding to each of a plurality of components in the group;
A clustering unit configured to cluster the plurality of constructs into a plurality of clusters;
A generating unit configured to generate a group of cluster feature data corresponding to the plurality of clusters;
A providing unit configured to provide a group of the cluster feature data generated by the generating unit to an information processing system;
With
The individual feature data represents a feature of a corresponding component,
The cluster feature data includes statistical data that expresses features of two or more constituents included in a corresponding cluster by a statistic.
The clustering unit clusters the plurality of constructs into a plurality of clusters according to a constraint specified by the information processing system,
The data generation system, wherein the generation unit generates the statistical data by converting the characteristics of the two or more constituents indicated by the individual feature data of the two or more constituents included in the corresponding cluster into the statistics. . A storage unit for storing a plurality of individual feature data corresponding to each of a plurality of components in the group;
A clustering unit configured to cluster the plurality of constructs into a plurality of clusters;
A generating unit configured to generate a group of cluster feature data corresponding to the plurality of clusters;
A providing unit configured to provide a group of the cluster feature data generated by the generating unit to an information processing system;
With
The individual feature data represents a feature of a corresponding component,
The cluster feature data includes statistical data that expresses features of two or more constituents included in a corresponding cluster by a statistic.
The information processing system is configured to transmit distance information representing a distance on a feature space between the plurality of constructs to the data providing system,
The clustering unit clusters the plurality of constructs into the plurality of clusters based on the distance information from the information processing system,
The data generation system, wherein the generation unit generates the statistical data by converting the characteristics of the two or more constituents indicated by the individual feature data of the two or more constituents included in the corresponding cluster into the statistics. . A storage unit for storing a plurality of individual feature data corresponding to each of a plurality of components in the group;
A clustering unit configured to cluster the plurality of constructs into a plurality of clusters;
A generating unit configured to generate a group of cluster feature data corresponding to the plurality of clusters;
A providing unit configured to provide a group of the cluster feature data generated by the generating unit to an information processing system;
With
The individual feature data represents a feature of a corresponding component,
The cluster feature data includes statistical data that expresses features of two or more constituents included in a corresponding cluster by a statistic.
The information processing system is configured to transmit classification information representing each classification of the plurality of constructs to the data providing system,
The clustering unit, based on the classification information from the information processing system, clusters the plurality of constituents into the plurality of clusters so that constituents of different classifications are not mixed in one cluster,
The data generation system, wherein the generation unit generates the statistical data by converting the characteristics of the two or more constituents indicated by the individual feature data of the two or more constituents included in the corresponding cluster into the statistics. . An information processing method executed by a computer,
Obtaining a group of first feature data relating to a first group comprising a plurality of first constructs;
Obtaining a group of second feature data relating to a second group comprising a plurality of second constructs;
Combining the acquired first feature data group and the acquired second feature data group;
The first feature data is associated with a first identifier that is an identifier of the corresponding one or more first constructs, and represents the features of the corresponding one or more first constructs;
Each of the second feature data corresponds to each of a plurality of clusters in the second group, and each of the plurality of clusters includes two or more of the plurality of second constructs, The second feature data is associated with a second identifier that is an identifier of the two or more second constructs included in the corresponding cluster, and the two or more second configurations included in the corresponding cluster. It has statistical data that expresses body characteristics as statistics,
The combining is based on the first identifier associated with the first feature data, and each of the first feature data is associated with the corresponding second identifier. An information processing method comprising combining the group of the first feature data and the group of the second feature data so as to be combined with the statistical data. The information processing method according to claim 21,
Obtaining the group of first feature data includes reading the group of first feature data from a storage device storing the group of first feature data;
Obtaining the group of second feature data includes obtaining the group of second feature data from a data providing system that provides the group of second feature data. A data providing method executed by a computer,
Obtaining a plurality of individual feature data corresponding to each of a plurality of constructs in the group;
Clustering the plurality of constructs into a plurality of clusters;
Generating a group of cluster feature data corresponding to the plurality of clusters;
Providing a group of the generated cluster feature data to an information processing system;
Including
The individual feature data is associated with an identifier of a corresponding construct, and represents the feature of the corresponding construct;
The cluster feature data includes statistical data that represents the features of the two or more constructs included in the corresponding cluster in a statistic associated with identifiers of the two or more constructs included in the corresponding cluster;
The generating includes generating the statistical data by converting the characteristics of the two or more constituents indicated by the individual feature data of the two or more constituents included in the corresponding cluster into the statistics. Data provision method including. A data providing method executed by a computer,
Obtaining a plurality of individual feature data corresponding to each of a plurality of constructs in the group;
Clustering the plurality of constructs into a plurality of clusters;
Generating a group of cluster feature data corresponding to the plurality of clusters;
Providing a group of the generated cluster feature data to an information processing system;
Including
The individual feature data represents a feature of a corresponding component,
The cluster feature data includes statistical data that expresses features of two or more constituents included in a corresponding cluster by a statistic.
The clustering includes clustering the plurality of constructs into the plurality of clusters according to a constraint specified by the information processing system;
The generating includes generating the statistical data by converting the characteristics of the two or more constituents indicated by the individual feature data of the two or more constituents included in the corresponding cluster into the statistics. Data provision method including. A data providing method executed by a computer,
Obtaining a plurality of individual feature data corresponding to each of a plurality of constructs in the group;
Clustering the plurality of constructs into a plurality of clusters;
Generating a group of cluster feature data corresponding to the plurality of clusters;
Providing a group of the generated cluster feature data to an information processing system;
Including
The individual feature data represents a feature of a corresponding component,
The cluster feature data includes statistical data that expresses features of two or more constituents included in a corresponding cluster by a statistic.
The clustering includes clustering the plurality of constructs into the plurality of clusters based on distance information representing a distance on a feature space between the plurality of constructs from the information processing system,
The generating includes generating the statistical data by converting the characteristics of the two or more constituents indicated by the individual feature data of the two or more constituents included in the corresponding cluster into the statistics. Including data provision methods. A data providing method executed by a computer,
Obtaining a plurality of individual feature data corresponding to each of a plurality of constructs in a group; clustering the plurality of constructs into a plurality of clusters;
Generating a group of cluster feature data corresponding to the plurality of clusters;
Providing a group of the generated cluster feature data to an information processing system;
Including
The individual feature data represents a feature of a corresponding component,
The cluster feature data includes statistical data that expresses features of two or more constituents included in a corresponding cluster by a statistic.
The clustering is based on classification information representing each classification of the plurality of constituents from the information processing system, so that the plurality of constituents are not mixed in different constituents in one cluster. Clustering into the plurality of clusters,
The generating includes generating the statistical data by converting the characteristics of the two or more constituents indicated by the individual feature data of the two or more constituents included in the corresponding cluster into the statistics. Data provision method including. The data providing method according to any one of claims 23 to 26, wherein:
The method of providing data, wherein obtaining the plurality of individual feature data includes reading the plurality of individual feature data from a storage device that stores the plurality of individual feature data.   The computer program for functioning a computer as the said acquisition unit with which the information processing system as described in any one of Claims 1-12 is provided, and the said coupling | bonding unit.   A computer program for causing a computer to function as the clustering unit, the generating unit, and the providing unit provided in the data providing system according to any one of claims 13 to 16.   A computer program for causing a computer to function as the clustering unit, the generating unit, and the providing unit provided in the data providing system according to any one of claims 18 to 20.

Images