# US 11,308,103 B2 · Data analyzing device and data analyzing method > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,308,103 B2 - Document ID: US11308103B2 - Status: Granted - Kind: B2 - Application number: 16/507,063 - Filed: 2019-07-10 - Published: 2020-08-13 - Granted: 2022-04-19 - Earliest claimed: 2019-02-13 - Assignees: Keyence Corporation - Applicants: Keyence Corporation - Inventors: Taiga Nomi; Yoshinori Nakano - Web page: https://patentgrep.com/patent/US11308103B2 ## CPC - G06F 16/2465: {Query processing support for facilitating data mining operations in structured databases} - G06F 16/9032: Query formulation - G06N 5/02: Knowledge representation; Symbolic representation - G06N 5/01: Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound - G06N 7/01: Probabilistic graphical models, e.g. probabilistic networks - G06N 20/00: Machine learning - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract To make it possible to automatically generate an objective variable so that even those who are unskilled in data processing can freely conduct an analysis. It is recognized whether a type of an attribute value of each attribute of analysis target data is a numeric type indicating that it is a numeric value or a category type indicating that it is a category. At least one attribute that is a numeric type is set as an attribute of an aggregation target, at least one attribute that is a category type is set as an attribute of an aggregation unit, and for each category value of the attribute of the aggregation unit, a numeric value of the attribute of the aggregation target is aggregated based on a previously prepared aggregation function to generate a new objective variable candidate. ## Claims ### Claim 1 A data analyzing device which receives an input of tables of analysis target data including a plurality of attributes, defines a relation between the tables by specifying one attribute respectively for the tables, receives a selection of an objective variable being predicted, and generates a prediction model for predicting the objective variable using at least one attribute of the tables, the device comprising a processor in communication with a memory, the processor being configured to execute instructions stored in the memory that cause the processor to: recognize whether a type of an attribute value of each attribute of the analysis target data is a numeric type indicating that it is a numeric value or a category type indicating that it is a category, set at least one attribute of an aggregation unit that is used to define the relation between the tables, the attribute of the aggregation unit being recognized as the category type, set at least one attribute of an aggregation target, the attribute of the aggregation target being recognized as the numeric type aggregate, for each category value of the attribute of the aggregation unit, a numeric value of the attribute of the aggregation target based on a previously prepared aggregation function to generate a plurality of new objective variable candidates, the plurality of new objective variable candidates not being included in the attributes of the tables and being generated by using the attribute of the aggregation unit that is used to define the relation between the tables, display the plurality of new objective variable candidates, receive a selection of one new objective variable by a user from among the plurality of displayed new objective variable candidates, and generate a new prediction model for predicting the selected new objective variable. ### Claim 2 (depends on 1) The data analyzing device according to claim 1, wherein the processor is further caused to display an original attribute of the analysis target data and the plurality of new objective variable candidates, and receive a selection of an objective variable by a user from the original attribute and the plurality of new objective variable candidates that are displayed. ### Claim 3 (depends on 2) The data analyzing device according to claim 2, wherein the processor is further caused to display a value obtained by converting an attribute value of an original attribute of the analysis target data into a flag, and receive a selection of an objective variable by a user from the original attribute, the plurality of new objective variable candidates, and the value converted into the flag that are displayed. ### Claim 4 (depends on 1) The data analyzing device according to claim 1, wherein the processor is further caused to set at least one attribute that is a numeric type recognized as an attribute of an aggregation target, set at least one attribute that is a numeric type recognized as an attribute of an aggregation unit, and aggregate, for each numeric value of the attribute of the aggregation unit, a numeric value of the attribute of the aggregation target based on a previously prepared aggregation function to generate a new objective variable candidate. ### Claim 5 (depends on 1) The data analyzing device according to claim 1, wherein the processor is further caused to automatically generate a predetermined period as an aggregation condition when the analysis target data includes a date type attribute. ### Claim 6 (depends on 1) The data analyzing device according to claim 1, wherein the processor is further caused to calculate a score representing usefulness of an objective variable candidate, and display an objective variable candidate having a score calculated by the processor equal to or greater than a predetermined value. ### Claim 7 (depends on 6) The data analyzing device according to claim 6, wherein the processor is further caused to calculate a score based on the number of lines of generated objective variables. ### Claim 8 (depends on 6) The data analyzing device according to claim 6, wherein the processor is further caused to calculate a score based on a variance of the generated objective variables. ### Claim 9 (depends on 1) The data analyzing device according to claim 1, wherein the processor is further caused to delete, after the generation of the plurality of new objective variable candidates, an objective variable candidate that meets a predetermined condition. ### Claim 10 (depends on 1) The data analyzing device according to claim 1, wherein the processor is further caused to automatically generate a new feature value by applying a predetermined function to attributes excluding at least a part of attribute values of attributes of the aggregation target used when generating the new selected objective variable. ### Claim 11 (depends on 1) The data analyzing device according to claim 1, wherein the processor is further caused to automatically generate a new feature value by applying a predetermined function to the original attribute or a combination of a plurality of attributes included in the analysis target data, and exclude from the generated feature values the feature value including information of at least a part of attribute values of attributes of the aggregation target used for the selected new objective variable to generate a final feature value. ### Claim 12 (depends on 1) The data analyzing device according to claim 1, wherein the processor is further caused to receive an input of keyword information from a user, and display an objective variable candidate related to the keyword information input by the user. ### Claim 13 (depends on 1) The data analyzing device according to claim 1, wherein the processor is further caused to receive a selection input from a user regarding an attribute of an aggregation target or an aggregation unit, and display an objective variable candidate generated by using an attribute input by the user as an aggregation target or an aggregation unit.