Information processing device, information processing method and information processing system

Abstract

To provide an information processing device capable of predicting what type of data content is combined with data that has not been provided for data combination.SOLUTION: An information processing device 1 includes: a generation unit 132 that, when either first data obtained from a first business operator or second data obtained from a second business operator is received as input, generates a data generation model having a plurality of latent variables, the data generation model for outputting pseudo-integrated data, the integrated data associating the input first data or second data with the pseudo data corresponding to non-input data, the data that has not been input, by generating values that can be taken by latent variables corresponding to the non-input data; and a transmission unit 133 that transmits the data generation model generated by the generation unit 132 to at least any of a first device 2 used by the first business operator and a second device 3 used by the second business operator.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present invention relates to an information processing apparatus, an information processing method, and an information processing system.

Background Art

[0002] Conventionally, it has been practiced to collect user information, which is information about users, from multiple operators and perform data analysis. For example, Patent Document 1 discloses a system that combines data related to the personal information of users corresponding to each of multiple operators based on a combination key for combining multiple pieces of user information.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] Data that an operator can provide to a system for data combination may be some records among the data the operator has. In this case, the operator can receive from the system the combined data corresponding to the data provided to the system side. On the other hand, on the operator side, there may be a case where it is desired to predict what kind of data will be combined with the data related to personal information not provided to the system side.

[0005] Therefore, the present invention has been made in view of these points, and an object thereof is to make it possible to predict what kind of data will be combined with the data not provided for data combination.

Means for Solving the Problems

[0006] The information processing apparatus according to the first aspect of the present invention is a data generation model having a plurality of latent variables determined based on input data and used for determining output data. When receiving an input of either the first data acquired from the first operator or the second data acquired from the second operator, by generating values that the latent variables corresponding to the uninput data, which is the data that was not input among the first data or the second data, can take, a pseudo-integrated data, which is integrated data associating the input first data or second data with pseudo-data corresponding to the uninput data, is output. The information processing apparatus has a generation unit that generates a data generation model, and a transmission unit that transmits the data generation model generated by the generation unit to at least one of a first device used by the first operator and a second device used by the second operator.

[0007] The generation unit may generate a data generation model that is set to output the integrated data for an input of integrated data, which is data associating the first data and the second data, and that can update parameters by a loss function set to output the input integrated data in response to at least one of the first data and the second data being input as the integrated data.

[0008] The information processing apparatus has a reception unit that receives, from the first operator, data identification information for identifying the first data and the second data, which is data identification information for associating the first data and the second data, and the first data, and receives, from the second operator, the data identification information and the second data. The generation unit may generate integrated data, which is data associating the first data and the second data based on the data identification information, and generate a data generation model in which the parameters are set by the loss function set to output the generated integrated data for an input of the integrated data.

[0009] The receiving unit may receive the first data or the second data with noise added thereto at a predetermined probability from at least one of the first device used by the first operator and the second device used by the second operator.

[0010] The information processing apparatus includes: a receiving unit that receives, from at least one of the first device and the second device, an updated model that is a data generation model in which the parameter is updated by the loss function; and an updating unit that updates the parameter provided in the data generation model generated by the generation unit based on the parameter included in the updated model, thereby updating the data generation model. The transmitting unit may transmit the data generation model updated by the updating unit to at least one of the first device and the second device.

[0011] The parameter includes a first unique parameter that is not updated based on the second data but is updated based on the first data, and a second unique parameter that is not updated based on the first data but is updated based on the second data. The information processing apparatus may include a generation unit that generates a data generation model capable of updating the first unique parameter when the first data is input and capable of updating the second unique parameter when the second data is input.

[0012] The information processing apparatus includes: a receiving unit that receives the pseudo-integrated data from at least one of the first device and the second device; and an updating unit that updates the data generation model by updating the parameter so that the pseudo-integrated data is output in response to the input of the pseudo-integrated data received by the receiving unit. The transmitting unit may transmit the data generation model updated by the updating unit to at least one of the first device and the second device. The receiving unit may receive the pseudo-integrated data with noise added thereto at a predetermined probability from at least one of the first device and the second device.

[0013] The information processing apparatus includes a receiving unit that receives, from at least one of the first device and the second device, intermediate data for updating the parameters, which is generated by the data generation model in the process of generating the pseudo-integration data, and an updating unit that updates the parameters based on the intermediate data received by the receiving unit. The transmitting unit may transmit the data generation model updated by the updating unit to at least one of the first device and the second device. The receiving unit may receive, from at least one of the first device and the second device, the intermediate data to which noise is added with a predetermined probability.

[0014] The receiving unit includes a receiving unit that receives, from at least one of the first device and the second device, first pseudo-integration data and the latent variable obtained in the process of outputting the first pseudo-integration data, and a transmitting unit that transmits the latent variable received by the receiving unit to the other device different from the source of the latent variable among the first device and the second device. The receiving unit receives second pseudo-integration data output from the data generation model using the latent variable transmitted by the transmitting unit from the other device. The information processing apparatus has an updating unit that updates the data generation model by updating the parameters of the data generation model generated by the generation unit so that the first pseudo-integration data is output from the data generation model based on the comparison result between the first pseudo-integration data received by the receiving unit and the second pseudo-integration data. The transmitting unit may transmit the data generation model updated by the updating unit to at least one of the first device and the second device.

[0015] The information processing apparatus may include a receiving unit that receives, from at least one of the first device and the second device, usage status information indicating the usage status of the data generation model, and a charging unit that charges the operator who uses the data generation model based on the usage status information received by the receiving unit.

[0016] The information processing method according to the second aspect of the present invention is a data generation model having a plurality of latent variables determined based on input data and used for determining output data, which is executed by a computer. When receiving an input of either the first data acquired from a first operator or the second data acquired from a second operator, a value that the latent variable corresponding to the uninput data, which is the data that was not input among the first data or the second data, can take is generated, thereby outputting pseudo-integrated data, which is integrated data associating the input first data or second data with pseudo data corresponding to the uninput data. The method includes a step of generating a data generation model; and a step of transmitting the generated data generation model to at least one of a first device used by the first operator and a second device used by the second operator.

[0017] The information processing system according to the third aspect of the present invention is an information processing system including a first device used by a first operator, a second device used by a second operator, and an information processing device communicably connected to the first device and the second device. The information processing device includes a generation unit that generates a data generation model having a plurality of latent variables determined based on input data and used for determining output data. When receiving an input of either the first data acquired from the first device or the second data acquired from the second device, the generation unit generates a value that the latent variable corresponding to the uninput data, which is the data that was not input among the first data or the second data, can take, thereby outputting pseudo-integrated data, which is integrated data associating the input first data or second data with pseudo data corresponding to the uninput data. The information processing device also includes a transmission unit that transmits the data generation model generated by the generation unit to at least one of the first device and the second device. At least one of the first device and the second device includes a reception unit that receives the data generation model, and an acquisition unit that inputs at least one of the first data and the second data as the integrated data to the data generation model and acquires the pseudo-integrated data output from the data generation model. [Advantages of the Invention]

[0018] According to the present invention, it is possible to predict what kind of data will be combined with data that has not been provided for data combination, which has the effect of enabling such prediction. [Brief Description of the Drawings]

[0019]

Figure 1

Figure 2

Figure 3

Figure 4

[0020] [Outline of Information Processing System S] FIG. 1 is a diagram for explaining the outline of the information processing system S. The information processing system S includes an information processing apparatus 1, a first apparatus 2 that manages first data, and a second apparatus 3 that manages second data, and is a system that generates a data generation model that outputs integrated data obtained by integrating the first data and the second data.

[0021] The information processing apparatus 1 is, for example, a computer used by an aggregation business operator that provides a service for aggregating data and providing the aggregated data. The information processing apparatus 1 is communicably connected to external apparatuses such as the first apparatus 2 and the second apparatus 3 via a communication network (not shown) such as the Internet or a mobile phone line.

[0022] The first device 2 is, for example, a computer used by a first operator. The first device 2 manages a plurality of first records associating a data ID (Identification) as data identification information for identifying data with first data. The first data is, for example, information about users collected by the first operator from users who use the services provided by the first operator.

[0023] The second device 3 is, for example, a computer used by a second operator. The second device 3 manages a plurality of second records associating a common data ID with the data ID associated with the first data and second data. The second data is, for example, information about users collected by the second operator from users who use the services provided by the second operator.

[0024] The information processing device 1 acquires a plurality of first records from the first device 2 and acquires a plurality of second records from the second device 3. The information processing device 1 generates integrated data by integrating the first data included in the acquired first records and the second data included in the second records based on the data ID. The information processing device 1 generates a data generation model that outputs integrated data in response to an input of the integrated data. The data generation model is, for example, a model capable of updating parameters by a loss function set so that the input integrated data is output in response to an input of at least any one of the first data and the second data as the integrated data.

[0025] Further, the data generation model is a data generation model having a plurality of latent variables determined based on the input data and used for determining the output data, and when receiving an input of either the first data or the second data constituting the integrated data, it is a model capable of outputting pseudo-integrated data, which is integrated data associating pseudo-data corresponding to the uninput data, which is the data that has not been input among the first data and the second data, with the input data.

[0026] The information processing apparatus 1 transmits the generated data generation model to the first apparatus 2 and the second apparatus 3. As a result, in the first apparatus 2 and the second apparatus 3, only one of the first data and the second data is input to the data generation model, and pseudo-integrated data in which the uninput data is complemented can be obtained from the data generation model. As a result, in the first apparatus 2 and the second apparatus 3, it is possible to predict what kind of data will be combined with the data that was not provided for data combination.

[0027] [Functional Configuration of Information Processing Apparatus 1] Subsequently, the functional configuration of the information processing apparatus 1 will be described. FIG. 2 is a diagram showing the functional configuration of the information processing apparatus 1.

[0028] As shown in FIG. 2, the information processing apparatus 1 includes a communication unit 11, a storage unit 12, and a control unit 13. The communication unit 11 is a communication interface for transmitting and receiving data to and from the first apparatus 2 and the second apparatus 3 via a communication network.

[0029] The storage unit 12 is a storage medium that stores various types of data, and includes a ROM (Read Only Memory), a RAM (Random Access Memory), a hard disk, an SSD (Solid State Drive), and a flash memory, etc. The storage unit 12 stores the programs executed by the control unit 13. The storage unit 12 stores programs that cause the control unit 13 to function as a receiving unit 131, a generating unit 132, a transmitting unit 133, an updating unit 134, and a charging unit 135.

[0030] The control unit 13 is, for example, a CPU (Central Processing Unit). The control unit 13 functions as a receiving unit 131, a generating unit 132, a transmitting unit 133, an updating unit 134, and a charging unit 135 by executing the programs stored in the storage unit 12.

[0031] [Description of Data Generation Model] In the following description of the functions of the control unit 13, a data generation model generated by the information processing apparatus 1 will be described. FIG. 3 is a diagram for explaining the data generation model.

[0032] The data generation model is an application program that implements a neural network having an input layer, an intermediate layer composed of a plurality of layers, and an output layer, and is set so that the input integrated data is output for the input of the integrated data. The data generation model has a plurality of nodes that constitute layers, calculate values of latent variables based on the input data, and output the calculated values of the latent variables. Each of the plurality of nodes is connected to one or more other nodes. The data generation model is provided with parameters indicating the strength (weights) of the connection relationship between the nodes and other nodes.

[0033] As shown in FIGS. 3(A) to 3(C), the data generation model is set so that the same data as the input data is output as output data in response to the input of at least one of the first data and the second data as input data.

[0034] Further, the data generation model can update the parameters by a loss function set so that the input integrated data is output in response to the input of at least one of the first data and the second data as integrated data. For example, the data generation model can update the parameters so that the value of the loss function indicating the difference between the output data and the input data becomes 0.

[0035] The parameters in the data generation model include a first unique parameter that is not updated based on the second data but is updated based on the first data, a second unique parameter that is not updated based on the first data but is updated based on the second data, and a common parameter that is updated based on the first data and the second data.

[0036] For example, when the first data is input to the data generation model and the second data is not input, the common parameter and the first specific parameter are updated. Also, when the second data is input to the data generation model and the first data is not input, the common parameter and the second specific parameter are updated. Further, when the first data and the second data, that is, the integrated data, are input to the data generation model, the common parameter, the first specific parameter, and the second specific parameter are updated.

[0037] Also, as described above, the data generation model has a plurality of latent variables. The latent variables are internal variables that are determined based on the input data and are used to determine the output data. As shown in FIGS. 3(D) and 3(E), when either the first data or the second data is input to the data generation model, the data generation model can automatically generate the value of the latent variable corresponding to the uninput data, which is the data that was not input, among the first data or the second data. For example, the data generation model specifies in advance the possible values that the latent variable can take, and based on the possible values, generates the value of the latent variable corresponding to the uninput data. Thereby, the data generation model can output pseudo-integrated data, which is integrated data associating the input first data or second data with pseudo-data corresponding to the uninput data.

[0038] For example, the data generation model receives an input of a setting parameter indicating whether to output pseudo-integrated data. When the value of the input setting parameter indicates that pseudo-integrated data is to be output, if either the first data or the second data is input, the data that was not input is complemented to output pseudo-integrated data. In the first operator and the second operator, there are cases where, when there is first data or second data that is not provided to the information processing apparatus 1, it is desired to confirm what second data or first data is associated with these first data or second data. In contrast, the first operator and the second operator can confirm the second data or first data associated with the first data or second data by inputting the first data or second data into the data generation model and obtaining pseudo-integrated data from the data generation model, using only the first data or second data they have.

[0039] [Generation and Update of Data Generation Model] Subsequently, the functions of the control unit 13 related to the generation and update of the data generation model will be described. The receiving unit 131 obtains, from the first operator, a data ID for identifying the first data and the second data, a data ID for associating the first data and the second data, and the first data, and receives, from the second operator, the data ID and the second data. For example, the receiving unit 131 receives a plurality of first records associating the data ID and the first data from the first device 2, and receives a plurality of second records associating the data ID and the second data from the second device 3.

[0040] The receiving unit 131 receives first data or second data with noise added thereto from at least one of the first device 2 and the second device 3 with a predetermined probability. Here, the predetermined probability is the noise addition rate in the first data and the second data when noise is added to the integrated data so as to satisfy ε-local differential privacy when the first data and the second data are integrated to generate integrated data. Further, the noise indicates, for example, that other data that the actual data can take is substituted for the actual data. By adding noise to the first data and the second data, a part of the user information included in each of the first data and the second data is anonymized, and the privacy of the user can be enhanced.

[0041] The generation unit 132 is a data generation model set to output integrated data in response to an input of integrated data, which is data associating first data acquired from a first operator and second data acquired from a second operator, and is a data generation model capable of updating parameters by a loss function set so that the input integrated data is output in response to at least one of the first data and the second data being input as the integrated data.

[0042] Specifically, first, the generation unit 132 generates integrated data, which is data associating first data acquired from the first device 2 used by the first operator and second data acquired from the second device 3 used by the second operator, based on the data ID received by the receiving unit 131 together with the first data and the second data.

[0043] Then, the generation unit 132 generates a data generation model with parameters set by a loss function set to output the integrated data for the input of the integrated data. For example, the generation unit 132 uses the generated integrated data as teacher data with the input data and the output data. Then, the generation unit 132 uses the teacher data to learn the parameters indicating the strength of the connection relationship between the nodes constituting the layer and other nodes connected to the node so that the value of the loss function indicating the difference between the input data to the data generation model and the output data output from the data generation model becomes 0. As a result, a data generation model with parameters set to output the integrated data is generated for the input of the integrated data.

[0044] The transmission unit 133 transmits the data generation model generated by the generation unit 132 to at least one of the first device 2 and the second device 3. Note that the data generation model before being transmitted by the transmission unit 133 is also referred to as a pre - transmission model.

[0045] The first device 2 and the second device 3 have a reception unit and receive the data generation model transmitted by the information processing device 1. Further, the first device 2 and the second device 3 have an update unit, input at least one of the first data and the second data as integrated data to the data generation model, and update the parameters of the data generation model.

[0046] For example, the first device 2 and the second device 3 set the value of the setting parameter of the data generation model to a value that outputs pseudo - integrated data. The first device 2 and the second device 3 can input either the first data or the second data to the data generation model and obtain the pseudo - integrated data output from the data generation model. As a result, the first operator and the second operator can input the first data or the second data they hold to the data generation model to obtain pseudo - integrated data that complements the data they do not hold, so that analysis can be performed based on the data they do not hold.

[0047] Further, when the first data is input, the data generation model updates the common parameters and the first specific parameters by means of a loss function, and when the second data is input, the data generation model updates the common parameters and the second specific parameters by means of a loss function. Here, the number of updates is not limited to one. Further, in order to prevent overfitting, regularization may be performed to limit the degree of change of the common parameters when updating the values of the common parameters. The data generation model with updated parameters on the side of the first operator or the second operator is also referred to as an updated model. The first device 2 and the second device 3 transmit the updated model to the information processing device 1, for example, at regular intervals.

[0048] The receiving unit 131 receives an updated model, which is a data generation model with parameters updated by means of a loss function, from at least one of the first device 2 and the second device 3. The updating unit 134 updates the pre-transmission model by updating the parameters provided in the pre-transmission model generated by the generating unit 132 based on the parameters included in the updated model received by the receiving unit 131.

[0049] The updating unit 134 compares the parameters included in the updated model and the pre-transmission model, and identifies the updated parameters. The updating unit 134 updates the updated parameters among the parameters included in the pre-transmission model to the updated parameters included in the updated model.

[0050] For example, when the common parameters and the first specific parameters are updated in the updated model, the updating unit 134 updates the values of the common parameters and the first specific parameters included in the pre-transmission model to the values of the common parameters and the first specific parameters included in the updated model. Further, when the common parameters and the second specific parameters are updated in the updated model, the updating unit 134 updates the values of the common parameters and the second specific parameters included in the pre-transmission model to the values of the common parameters and the second specific parameters included in the updated model.

[0051] Here, when the updated models are received from the first device 2 and the second device 3, the values of the common parameters are updated in both of the two received updated models. In this case, the update unit 134 updates the value of the common parameter included in the pre-transmission model based on the value of the common parameter included in the updated model received from the first device 2 and the value of the common parameter included in the updated model received from the second device 3. The update unit 134 updates, for example, the value of the common parameter included in the pre-transmission model to the average value of the value of the common parameter included in the updated model received from the first device 2 and the value of the common parameter included in the updated model received from the second device 3.

[0052] Here, the update unit 134 may update the value of the common parameter included in the pre-transmission model to a statistical value such as the median of the value of the common parameter included in the updated model received from the first device 2 and the value of the common parameter included in the updated model received from the second device 3. Further, in order to prevent overfitting, when updating the value of the common parameter included in the pre-transmission model, the update unit 134 may perform regularization to limit the degree of change of the common parameter, or prepare a plurality of learning environments, update the value of the common parameter included in the pre-transmission model in each of the plurality of learning environments, and aggregate the updated values of the common parameter. By doing so, the information processing apparatus 1 can reflect the learning results of the data generation model performed in the first device 2 and the second device 3 in the pre-transmission model.

[0053] Note that the receiving unit 131 may obtain information indicating the number of times of use as information indicating the usage status of the data generation model from each of the first device 2 and the second device 3. Then, the updating unit 134 may update the value of the common parameter included in the pre-transmission model based on the usage status of the data generation model in each of the first device 2 and the second device 3, the value of the common parameter included in the updated model received from the first device 2, and the value of the common parameter included in the updated model received from the second device 3. Based on the number of times of use of the data generation model in each of the first device 2 and the second device 3, each of the value of the common parameter included in the updated model received from the first device 2 and the value of the common parameter included in the updated model received from the second device 3 may be weighted, and the average value of the two weighted common parameters may be used as the value of the common parameter included in the pre-transmission model.

[0054] Further, the updating unit 134 updates the parameters of the pre-transmission model based on the parameters included in the updated model received from at least one of the first device 2 and the second device 3, but is not limited thereto. For example, the updating unit 134 may update the parameters of the pre-transmission model based on the pseudo-integrated data.

[0055] In this case, the receiving unit 131 receives pseudo-integrated data from at least one of the first device 2 and the second device 3. Here, the receiving unit 131 may receive pseudo-integrated data with noise added thereto at a predetermined probability from at least one of the first device 2 and the second device 3. By doing so, the information processing apparatus 1 can collect pseudo-integrated data subjected to anonymization processing, and can suppress the leakage of user information that has not been subjected to anonymization processing during the collection process.

[0056] The updating unit 134 updates the data generation model (pre-transmission model) by using the pseudo-integrated data received by the receiving unit 131 to perform learning of the data generation model and updating the parameters so that the pseudo-integrated data is output for the input of the pseudo-integrated data.

[0057] Further, the update unit 134 may update the parameters of the pre - transmission model based on intermediate data for updating parameters generated by the data generation model in the process of generating the pseudo - integrated data, for example, gradient information indicating the gradient of the parameters. In this case, the reception unit 131 receives, from at least one of the first device 2 and the second device 3, intermediate data for updating parameters generated by the data generation model in the process of generating the pseudo - integrated data. The intermediate data is, for example, a data set output in a predetermined layer included in the intermediate layer of the data generation model and input to the layer next to the predetermined layer. Here, the reception unit 131 may receive intermediate data with noise added thereto from at least one of the first device 2 and the second device 3 with a predetermined probability. In this case, it is assumed that the possible values that each of the plurality of data included in the intermediate data can take are specified in advance, and the plurality of data included in the intermediate data are replaced with other possible data with a predetermined probability.

[0058] The update unit 134 inputs the intermediate data received by the reception unit 131 into the layer to which the intermediate data is input among the plurality of layers constituting the intermediate layer, thereby performing learning of the data generation model (pre - transmission model). Thereby, the update unit 134 updates the parameters based on the intermediate data so that integrated data corresponding to the intermediate data is output, thereby updating the data generation model.

[0059] Further, the receiving unit 131 may receive the first pseudo-integrated data and the latent variable obtained in the process of outputting the first pseudo-integrated data from at least one of the first device 2 and the second device 3. Then, the transmitting unit 133 may transmit the latent variable received by the receiving unit 131 to the other device different from the source of the latent variable among the first device 2 and the second device 3. In this case, the other device causes the data generation model that it has received in advance to output the second pseudo-integrated data using the latent variable, and transmits the second pseudo-integrated data to the information processing device 1. Here, regularization may be added to the learning of the parameters in the other device. The receiving unit 131 of the information processing device 1 receives the second pseudo-integrated data from the other device.

[0060] The updating unit 134 updates the pre-transmission model by updating the parameters so that the first pseudo-integrated data is output from the pre-transmission model based on the comparison result between the first pseudo-integrated data and the second pseudo-integrated data received by the receiving unit 131. For example, the updating unit 134 compares the first pseudo-integrated data and the second pseudo-integrated data received by the receiving unit 131, and updates the parameters so that the data generation model generated by the generation unit 132, that is, the first pseudo-integrated data is output from the pre-transmission model on the condition that the degree of coincidence exceeds a predetermined threshold value, thereby updating the pre-transmission model.

[0061] When the degree of coincidence between the first pseudo-integrated data and the second pseudo-integrated data exceeds a predetermined threshold, there is a high probability that the parameters of the data generation model distributed to the operator who output the first pseudo-integrated data have been updated. In contrast, when there is a high probability that the change in the parameters of the data generation model distributed to the operator side is large, the information processing apparatus 1 can reflect the parameters in the pre-transmission model parameters. Note that the update unit 134 updates the parameters so that the first pseudo-integrated data is output from the pre-transmission model on the condition that the degree of coincidence exceeds a predetermined threshold, but is not limited thereto. The update unit 134 may unconditionally update the parameters so that the first pseudo-integrated data is output from the pre-transmission model in response to the reception unit 131 receiving the first pseudo-integrated data.

[0062] The transmission unit 133 transmits the data generation model updated by the update unit 134 to at least one of the first device 2 and the second device 3. Thereby, the first operator can obtain pseudo-integrated data with improved accuracy using the data generation model in which the parameters updated based on the second data are reflected in the second operator. Similarly, the second operator can obtain pseudo-integrated data with improved accuracy using the data generation model in which the parameters updated based on the first data are reflected in the first operator.

[0063] [Charging Associated with Use of Data Generation Model] Subsequently, the charging associated with the use of the data generation model will be described. The receiving unit 131 receives usage status information indicating the usage status of the data generation model from at least one of the first device 2 and the second device 3. For example, the usage status information is information associating the operator identification information indicating the operator who used it with the number of times the data generation model was used. Also, the number of times the data model was used is the number of times of inputting the first data or the second data, which is a part of the integrated data, as input data and obtaining the pseudo-integrated data from the data generation model. Note that there are a learning phase of the data generation model and a prediction phase of obtaining pseudo-integrated data using the data generation model, and the number of times the data model was used may be the number of times of obtaining pseudo-integrated data in the prediction phase.

[0064] The charging unit 135 charges the operator who used the data generation model based on the usage status information received by the receiving unit 131. For example, the charging unit 135 charges the operator indicated by the operator identification information indicated by the usage status information so that the higher the number of times the data generation model indicated by the usage status information is used, the higher the charging amount. By doing so, the aggregator operating the information processing apparatus 1 can obtain the consideration for the usage of the data generation model by the operator.

[0065] [Operation Sequence] Subsequently, the processing flow of the information processing apparatus 1 will be described. FIG. 4 is a sequence diagram showing the processing flow until the information processing apparatus 1 generates and updates the data generation model.

[0066] First, the receiving unit 131 receives a plurality of first records associating the data ID with the first data from the first device 2 (S1), and receives a plurality of second records associating the data ID with the second data from the second device 3 (S2).

[0067] Subsequently, the generation unit 132 generates integrated data by concatenating the first record and the second record using the data IDs included in the received first record and second record as keys (S3). Subsequently, the generation unit 132 generates a data generation model by training the data generation model using the generated integrated data as teacher data (S4). Subsequently, the transmission unit 133 transmits the data generation model generated by the generation unit 132 to the first device 2 and the second device 3 (S5, S6).

[0068] The first device 2 inputs the first data as input data to the data generation model received from the information processing device 1, and obtains pseudo-integrated data by causing the data generation model to output the pseudo-integrated data (S7). When the first data is input as input data, the data generation model updates the common parameters and the first specific parameters. The first device 2 transmits the data generation model in which the common parameters and the first specific parameters are updated, that is, the updated model, to the information processing device 1 (S8).

[0069] Similarly, the second device 3 inputs the second data as input data to the data generation model received from the information processing device 1, and obtains pseudo-integrated data by causing the data generation model to output the pseudo-integrated data (S9). When the second data is input as input data, the data generation model updates the common parameters and the second specific parameters. The second device 3 transmits the data generation model in which the common parameters and the second specific parameters are updated, that is, the updated model, to the information processing device 1 (S10). The reception unit 131 of the information processing device 1 receives the updated model from the first device 2 and the second device 3.

[0070] The update unit 134 updates the parameters of the data generation model generated in S4 based on the parameters included in the updated models received from the first device 2 and the second device 3 (S11). Thereby, when only one of the first data and the second data is input as the input data, the information processing apparatus 1 can reflect the learning result based on the input data in the data generation model generated in S4.

[0071] Subsequently, the transmission unit 133 transmits the data generation model updated by the update unit 134 to the first device 2 and the second device 3 (S12, S13). Thereafter, in the first device 2 and the second device 3, acquisition of pseudo-integrated data using the data generation model is performed.

[0072] [Effect by the information processing apparatus 1] As described above, the information processing apparatus 1 according to the present embodiment is a data generation model having a plurality of latent variables determined based on input data and used for determining output data. When receiving an input of either the first data acquired from the first operator or the second data acquired from the second operator, by generating values that the latent variable corresponding to the uninput data, which is the data that was not input among the first data or the second data, can take, a pseudo-integrated data, which is integrated data associating the input first data or second data with pseudo-data corresponding to the uninput data, is output. Then, the information processing apparatus 1 transmits the generated data generation model to at least one of the first device 2 used by the first operator and the second device 3 used by the second operator. By doing so, in the first device 2 and the second device 3, it is possible to predict what content of data will be combined with the data that was not provided to the operator for data combination using the data generation model.

[0073] Note that the present invention can contribute to Goal 9, "Build the infrastructure for industry and innovation," of the Sustainable Development Goals (SDGs) led by the United Nations.

[0074] As described above, the present invention has been described using embodiments. However, the technical scope of the present invention is not limited to the scope described in the above embodiments, and various modifications and changes are possible within the scope of the gist. For example, all or part of the device can be configured by functionally or physically dispersing and integrating it in an arbitrary unit. Also, new embodiments resulting from an arbitrary combination of a plurality of embodiments are included in the embodiments of the present invention. The effects of the new embodiments resulting from the combination have the effects of the original embodiments combined.

Description of Reference Numerals

[0075] 1 Information processing device 2 First device 3 Second device 11 Communication unit 12 Storage unit 13 Control unit 131 Receiver 132 Generator 133 Transmitter 134 Updater 135 Charging unit S Information processing system

Claims

1. A data generation model having a plurality of latent variables determined based on input data and used for determining output data, which outputs pseudo-integrated data that is integrated data associating the input first data or second data with pseudo-data corresponding to uninput data, which is the data that was not input, among the first data or the second data, by generating values that the latent variables corresponding to the uninput data can take when receiving either the first data obtained from a first operator or the second data obtained from a second operator; a generation unit that generates the data generation model; A transmission unit that transmits the data generation model generated by the generation unit to at least one of a first device used by the first operator and a second device used by the second operator; An information processing apparatus having the above.

2. The generation unit generates the data generation model that is set to output the integrated data with respect to the input of integrated data that is data associating the first data and the second data, and that can update parameters by a loss function set to output the input integrated data in response to at least one of the first data and the second data being input as the integrated data, The information processing apparatus according to Claim 1.

3. A reception unit that receives, from the first operator, data identification information for identifying the first data and the second data, data identification information for associating the first data and the second data, and the first data, and receives, from the second operator, the data identification information and the second data; The generation unit generates integrated data that is data associating the first data and the second data based on the data identification information, and generates the data generation model in which the parameters are set by the loss function set to output the generated integrated data with respect to the input of the integrated data. The information processing apparatus according to claim 2.

4. The receiving unit receives the first data or the second data with noise added thereto at a predetermined probability from at least one of the first device used by the first operator and the second device used by the second operator. The information processing apparatus according to claim 3.

5. A receiving unit that receives an updated model, which is a data generation model in which the parameter is updated by the loss function, from at least one of the first device and the second device; An updating unit that updates the data generation model by updating the parameter provided in the data generation model generated by the generation unit based on the parameter included in the updated model; and The transmitting unit transmits the data generation model updated by the updating unit to at least one of the first device and the second device. The information processing apparatus according to claim 2.

6. The parameter includes a first unique parameter that is not updated based on the second data but is updated based on the first data, and a second unique parameter that is not updated based on the first data but is updated based on the second data. A generation unit that generates a data generation model capable of updating the first unique parameter when the first data is input and capable of updating the second unique parameter when the second data is input. The information processing apparatus according to claim 2.

7. A receiving unit that receives the pseudo-integrated data from at least one of the first device and the second device; An updating unit that updates the data generation model by updating the parameter so that the pseudo-integrated data is output in response to the input of the pseudo-integrated data received by the receiving unit; and The transmitting unit transmits the data generation model updated by the updating unit to at least one of the first device and the second device. The information processing apparatus according to claim 2.

8. The receiving unit receives the pseudo-integrated data with noise added thereto from at least one of the first device and the second device with a predetermined probability. The information processing apparatus according to claim 7.

9. A receiving unit that receives intermediate data for updating the parameters, which is generated by the data generation model in the process of generating the pseudo-integrated data, from at least one of the first device and the second device; An updating unit that updates the parameters based on the intermediate data received by the receiving unit; and The transmitting unit transmits the data generation model updated by the updating unit to at least one of the first device and the second device. The information processing apparatus according to claim 2.

10. The receiving unit receives the intermediate data with noise added thereto from at least one of the first device and the second device with a predetermined probability. The information processing apparatus according to claim 9.

11. A receiving unit that receives first pseudo-integrated data and the latent variable obtained in the process of outputting the first pseudo-integrated data from at least one of the first device and the second device; A transmitting unit that transmits the latent variable received by the receiving unit to the other device different from the source of the latent variable among the first device and the second device; and The receiving unit receives second pseudo-integrated data output from the data generation model using the latent variable transmitted by the transmitting unit from the other device. Based on the comparison result between the first pseudo-integration data and the second pseudo-integration data received by the receiving unit, the parameter of the data generation model is updated so that the first pseudo-integration data is output from the data generation model generated by the generation unit, thereby having an update unit that updates the data generation model. The transmitting unit transmits the data generation model updated by the updating unit to at least one of the first device and the second device. The information processing apparatus according to claim 2.

12. A receiving unit that receives usage status information indicating the usage status of the data generation model from at least one of the first device and the second device. A charging unit that charges a business operator who uses the data generation model based on the usage status information received by the receiving unit. Having The information processing apparatus according to claim 1.

13. Executed by a computer A data generation model having a plurality of latent variables determined based on input data and used for determining output data. When receiving an input of either the first data obtained from the first business operator or the second data obtained from the second business operator, by generating values that the latent variables corresponding to the uninput data, which is the data that was not input among the first data or the second data, can take, a step of generating a pseudo-integration data, which is integration data associating the input first data or second data with pseudo-data corresponding to the uninput data, and outputting the pseudo-integration data. A step of transmitting the generated data generation model to at least one of the first device used by the first business operator and the second device used by the second business operator. An information processing method having

14. An information processing system having a first device used by a first operator, a second device used by a second operator, and an information processing device communicably connected to the first device and the second device, wherein the information processing device, is a data generation model having a plurality of latent variables determined based on input data and used for determining output data, and when receiving either the first data acquired from the first device or the second data acquired from the second device, generates values that the latent variables corresponding to the uninput data, which is the data that was not input among the first data or the second data, can take, thereby outputting pseudo-integrated data, which is integrated data associating the input first data or second data with pseudo data corresponding to the uninput data; a generation unit that generates the data generation model; a transmission unit that transmits the data generation model generated by the generation unit to at least one of the first device and the second device; and has, wherein at least one of the first device and the second device, has a reception unit that receives the data generation model; and an acquisition unit that inputs at least one of the first data and the second data as the integrated data to the data generation model and acquires the pseudo-integrated data output from the data generation model; and has, an information processing system.

Images