Data analysis device

Abstract

Even users without specialized data processing knowledge or skills will be able to freely calculate and compare values ​​from previous periods, such as business metrics. [Solution] The data analysis device 1 sets up a relationship between the calendar table and the data table based on the dates included in the calendar table and the dates included in the data table. The data analysis device 1 sets the time-series attribute of the aggregation unit from among the time-series attributes included in the calendar table, and sets the attribute to be aggregated from among the attributes included in the data table. For each aggregation period, the data analysis device 1 aggregates the attribute values ​​of the attribute to be aggregated to calculate the aggregated value, and also aggregates the attribute values ​​of the attribute to be aggregated in the comparison period to calculate a comparison value corresponding to the aggregated value. For each time-series attribute of the aggregation unit, the data analysis device 1 displays the aggregated value and the comparison value corresponding to the aggregated value on the monitor 3.

Description

Technical Field

[0001] The present disclosure relates to a data analysis apparatus for analyzing various data.

Background Art

[0002] For example, Patent Document 1 discloses an apparatus that hierarchizes various attributes of products and performs matrix analysis on product data using this attribute hierarchy.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] By the way, for example, sales indicators, books, business intelligence, etc. are important business indicators. In the process of analyzing such business indicators using various tools, calculating and visualizing the values obtained by comparing the indicators with the values of the previous month or the same period of the previous year is of great value for many organizations and companies in conducting sound economic activities.

[0005] Conventionally, when performing analysis operations such as extracting and quoting a specific period from the original data to calculate the value of the previous period ratio, most methods assume the description of calculation processing using SQL (Structured Query Language) or DSL (Domain Specific Language), and there is a problem that it is limited in use to users with specialized data processing knowledge and skills. Also, even in a tool that can be simply analyzed, it can only compare pre-defined periods, and there is a problem that users cannot compare their own periods.

[0006] This disclosure is made in light of the above points, and its purpose is to enable users who do not possess specialized data processing knowledge or skills to freely calculate and compare, for example, the values ​​of business indicators from the previous period. [Means for solving the problem]

[0007] To achieve the above objective, one aspect of this disclosure may be based on a data analysis device that analyzes various types of data. The data analysis device includes a storage unit that stores a calendar table having multiple dates and time-series attributes corresponding to each date; a data input unit that accepts input of a data table having multiple attributes including dates; a relationship setting unit that sets a relationship between the calendar table and the data table based on the dates included in the calendar table stored in the storage unit and the dates included in the data table input by the data input unit; an aggregation setting unit that sets time-series attributes of aggregation units from among the time-series attributes included in the calendar table stored in the storage unit, and sets attributes to be aggregated from among the attributes included in the data table input by the data input unit; and the aggregation units set by the aggregation setting unit The system includes: an aggregation period defining unit that defines the aggregation period based on the time-series attributes; an identifier assigning unit that assigns an identifier to the aggregation period defined by the aggregation period defining unit to determine the time-series order; a comparison period identification unit that identifies a comparison period for comparing the aggregated values ​​of the aggregation period based on the identifier assigned by the identifier assigning unit; a calculation unit that calculates an aggregated value by aggregating the attribute values ​​of the aggregated attributes for each aggregation period defined by the aggregation period defining unit, and also calculates a comparison value corresponding to the aggregated value by aggregating the attribute values ​​of the aggregated attributes in the comparison period identified by the comparison period identification unit; and a display control unit that displays the aggregated value and the comparison value corresponding to the aggregated value on the display unit for each time-series attribute of the aggregation unit.

[0008] In this configuration, the aggregation period is defined based on the time-series attributes of the aggregation unit, and for each defined aggregation period, the attribute values ​​of the aggregated attributes are aggregated to calculate the aggregated value. Then, the attribute values ​​of the aggregated attributes in the comparison period, which are used to compare the aggregated value of the aggregation period, are aggregated to calculate the comparison value corresponding to the aggregated value. The calculated aggregated value and the comparison value corresponding to that aggregated value are displayed in the display section for each time-series attribute of the aggregation unit, so that, for example, business indicators can be calculated and compared with values ​​from the previous period or the same period of the previous year without performing calculation processing using SQL or DSL. [Effects of the Invention]

[0009] As explained above, even users without specialized data processing knowledge or skills can freely calculate and compare previous period values, such as business metrics. [Brief explanation of the drawing]

[0010] [Figure 1] Figure 1 is a diagram showing a schematic configuration of a data analysis device according to an embodiment of the present invention. [Figure 2] Figure 2 is a block diagram of the data analysis system. [Figure 3] Figure 3 is a flowchart showing an example of a data analysis process using a data analysis device. [Figure 4] Figure 4 shows an example of a data table that forms the basis of business indicators. [Figure 5] Figure 5 shows an example of a calendar table. [Figure 6] Figure 6 shows the user interface for registering setting items. [Figure 7] Figure 7 shows the user interface for comparison settings when comparing with the previous period. [Figure 8] Figure 8 shows the user interface for setting comparisons when comparing with the same period of the previous year. [Figure 9] Figure 9 shows the user interface for setting up aggregation. [Figure 10]Figure 10 is a diagram for explaining the concept of the aggregation process in the case of the comparison with the previous period. [Figure 11] Figure 11 is a diagram for explaining the generation process of the intermediate table in the case of the comparison with the previous period. [Figure 12] Figure 12 is a diagram for explaining the aggregation operation in the case of the comparison with the previous period. [Figure 13] Figure 13 is a diagram for explaining the comparison calculation process in the case of the comparison with the previous period. [Figure 14] Figure fourteen is a diagram for explaining another example of the comparison calculation process in the case of the comparison with the previous period. [Figure 15] Figure 15 is a diagram for explaining yet another example of the comparison calculation process in the case of the comparison with the previous period. [Figure 16] Figure 16 is a diagram showing the user interface when the period setting is automatically set in the case of the comparison with the previous period. [Figure 17] Figure 17 is a diagram showing an output example of the comparison value in the case of the comparison with the previous period. [Figure 18] Figure 18 is a diagram showing an example in which the rows of missing values are deleted in the case of the comparison with the previous period. [Figure 19] Figure 19 is a diagram for explaining the concept of the aggregation process in the case of the comparison with the same period of the previous year. [Figure 20] Figure 20 is a diagram for explaining the generation process of the intermediate table in the case of the comparison with the same period of the previous year. [Figure 21] Figure 21 is a diagram showing an output example of the comparison value in the case of the comparison with the same period of the previous year. [Figure 22] Figure 22 is a diagram for explaining the comparison calculation process in the case of the comparison with the same period of the previous year. [Figure 23] Figure 23 is a diagram for explaining another example of the comparison calculation process in the case of the comparison with the same period of the previous year. [Figure 24] Figure 24 is a diagram for explaining yet another example of the comparison calculation process in the case of the comparison with the same period of the previous year. [Figure 25] Figure 25 is a diagram showing the user interface when the period setting is automatically set in the case of the comparison with the same period of the previous year. [Figure 26] Figure 26 is a diagram showing an output example of the comparison value in the case of the comparison with the same period of the previous year. [Figure 27] Figure 27 shows an example where rows with missing values ​​have been removed in the case of year-on-year comparison. [Figure 28] Figure 28 is a simplified diagram illustrating the overall data analysis process. [Figure 29] Figure 29 illustrates how the display changes when attribute values ​​are grouped or ungrouped. [Modes for carrying out the invention]

[0011] Embodiments of the present invention will be described in detail below with reference to the drawings. The following description of preferred embodiments is essentially illustrative and is not intended to limit the present invention, its applications, or its uses.

[0012] Figure 1 is a schematic diagram showing the configuration of a data analysis device 1 according to an embodiment of the present invention, and Figure 2 is a block diagram of the data analysis device 1. The data analysis device 1 is a device capable of analyzing business indicators such as sales indicators, ledgers, and business intelligence, but it can also analyze various types of data other than business indicators. In other words, the types of data that can be analyzed by the data analysis device 1 are not limited to business indicators, but can analyze a variety of data. In the process of analyzing business indicators, it is necessary to calculate and visualize values ​​by comparing those indicators with values ​​from the previous month or the same period of the previous year (not limited to the previous year). With the data analysis device 1 according to this embodiment, users (analysts) can easily perform comparisons between their own periods without having to write SQL or DSL. In other words, even users without specialized data processing knowledge or skills can calculate and compare the values ​​of any business indicator, etc., from the previous period or the same period of the previous year.

[0013] First, let's explain the overview of Data Analysis Device 1. When a user creates an analysis application, it performs a join process on a table (hereinafter referred to as the calendar table) for centrally managing date and time-related information that is automatically added when the user creates an analysis application, or date and time-related information that the user defines themselves, and on transaction data containing time-series information owned by the user, using the condition that the data is equivalent between columns that can be displayed as dates in the YYYY-MM-DD format, which is part of the ISO 8601 standard.

[0014] Using the data analysis device 1, when setting up comparisons between specific periods, users can complete the settings with just clicks or simple input actions via the graphical user interface. Furthermore, users can freely select the periods to compare from those included in the calendar table. Additionally, when referencing data prior to a specified period from the registered settings, it is possible to reference the data based on its chronological order. Moreover, the data analysis device 1 can also provide an option to dynamically determine the period to be referenced depending on the state of the graphical user interface.

[0015] (Overall configuration of data analysis device 1) As shown in Figures 1 and 2, the data analysis device 1 comprises a main unit 2, a monitor (display unit) 3, a keyboard 4, and a mouse 5. The monitor 3, keyboard 4, and mouse 5 are connected to the main unit 2. The main unit 2 and the monitor 3 may be integrated, or a part of the main unit 2 or a part of the functions performed by the main unit 2 may be built into the monitor 3. The monitor 3 does not have to be a component of the data analysis device 1, and the monitor 3 can be connected and used with a data analysis device 1 that does not include a monitor 3.

[0016] The data analysis device 1 incorporates a communication module (not shown) and is configured to communicate with the outside world. This allows for data downloads from external servers, for example, via an internet connection. The communication method may be wireless or wired. The keyboard 4 and mouse 5 are examples of operating devices for the data analysis device 1, as well as examples of input devices for entering various types of information and examples of selection devices for performing selection operations. In addition to, or instead of, the keyboard 4 and mouse 5, touch panel input devices, voice input devices, pen-type input devices, etc., can also be used.

[0017] For example, a data analysis device 1 can be created by installing a program capable of executing the control and processing described later on a general-purpose personal computer. Alternatively, the data analysis device 1 can be configured with dedicated hardware on which the program is installed. For example, the program may be directly installed on the user's personal computer, allowing the personal computer to be used as the data analysis device 1. Alternatively, the program may be installed on a server to build the data analysis device 1, allowing each user to access it via the network from their personal computer's browser. Alternatively, the data analysis device 1 may be located on the cloud, allowing each user to access it from their personal computer's browser. Furthermore, some of the control and processing described later may be executed on the user's personal computer, while the rest can be executed on another person's personal computer or on the cloud. In other words, it is not necessary for all of the control and processing performed by the data analysis device 1 to be performed on the same personal computer, and any system that achieves similar effects is the data analysis device 1. Similarly, in the data analysis process using the data analysis device shown as an example in Figure 3, it is not necessary for all steps SA1 to SA8 to be performed on the same personal computer.

[0018] (Monitor 3 configuration) The monitor 3 shown in Figure 1 consists of, for example, an organic EL display or a liquid crystal display, and can be called a display unit on its own, or the monitor 3 and the display control unit 3a shown in Figure 2 can be called a display unit together. The display control unit 3a may be built into the monitor 3 or built into the main unit 2 of the device. The display control unit 3a includes a display DSP for displaying images on the monitor 3. The display control unit 3a may include video memory such as VRAM for temporarily storing image data when displaying an image. Based on display commands sent from the CPU 11a of the main control unit 11 (described later), the display control unit 3a transmits control signals to the monitor 3 to display a predetermined image. For example, it transmits control signals to display various graphical user interfaces, icons, and user operations using the keyboard 4 and mouse 5 on the monitor 3, as well as various graphical user interfaces as described later. It is also possible to display a pointer that can be operated with the mouse 5 on the monitor 3. The monitor 3 can also be a touch-operated panel type monitor, which allows the monitor 3 to have input functions for various information, operation functions for the data analysis device 1, and selection operation functions.

[0019] (Overall configuration of device body 2) The device body 2 shown in Figure 1 comprises a control unit 10 and a storage unit 30. The storage unit 30 is composed of a hard disk drive, a solid-state drive (SSD), etc. The storage unit 30 is connected to the control unit 10 and controlled by the control unit 10, and can store various types of data and also read the stored data. Part or all of the storage unit 30 may reside in the cloud, or it may reside in a facility separate from the device body 2. Part of the storage unit 30 may also be part of a core system. In other words, the device body 2 and the storage unit 30 may be installed in different locations.

[0020] The memory unit 30 stores a calendar table containing multiple dates and multiple time-series attributes, each corresponding to at least one year attribute. The calendar table stored in the memory unit 30 is configured so that there are no duplicates or missing dates. The time-series attributes of the calendar table may include, for example, year, half-year, quarter, month, etc., but it is not necessary to include all of these time-series attributes; it is sufficient to include any time-series attributes.

[0021] If the time series attributes include year, half-year, quarter, month, etc., the memory unit 30 will store a calendar table as time series attributes, which for example includes a year attribute (first level attribute), a half-year attribute (second level attribute), a quarter attribute (third level attribute), and a month attribute (fourth level attribute). The first level is the highest level, and the fourth level is the lowest level. Therefore, the calendar table stored in the memory unit 30 includes a first time series attribute and a second time series attribute, which is a subordinate attribute of the first time series attribute. For example, if the year attribute is the first time series attribute, then the half-year attribute becomes the second time series attribute. Furthermore, the quarter attribute can also be called the third time series attribute, which is a subordinate attribute of the second time series attribute, and the month attribute can also be called the fourth time series attribute, which is a subordinate attribute of the third time series attribute.

[0022] Furthermore, the storage unit 30 also stores a data table that has multiple attributes, including the date. The part where the data table is stored and the part where the calendar table is stored may be different.

[0023] (Control unit 10) Although not specifically illustrated, the control unit 10 can consist of an MPU, system LSI, DSP, and dedicated hardware. The control unit 10 incorporates various functions, which may be implemented by logic circuits or by executing software.

[0024] As shown in Figure 2, the control unit 10 includes a main control unit 11 connected to bus B. The main control unit 11 performs numerical calculations, arithmetic processing, and various information processing based on various programs, as well as controlling each part of the hardware. The main control unit 11 includes a CPU 11a that functions as a central processing unit, a work memory 11b such as RAM that functions as a work area when the main control unit 11 executes various programs, and a program memory 11c such as ROM, flash ROM, or EEPROM that stores startup programs, initialization programs, etc.

[0025] The control unit 10 further comprises a data input unit 12, a relationship setting unit 13, an aggregation setting unit 14, an aggregation target period definition unit 15, an identifier assignment unit 16, a comparison target period specification unit 17, a calculation unit 18, and a switching unit 19. In this specification, each part is described separately as the main control unit 11, data input unit 12, relationship setting unit 13, aggregation setting unit 14, aggregation target period definition unit 15, identifier assignment unit 16, comparison target period specification unit 17, calculation unit 18, and switching unit 19, but some or all of these may be made up of the same hardware, or they may be made up of different hardware. For example, the main control unit 11, data input unit 12, relationship setting unit 13, aggregation setting unit 14, aggregation target period definition unit 15, identifier assignment unit 16, comparison target period specification unit 17, calculation unit 18, and switching unit 19 may be configured in a single computer, or the main control unit 11, data input unit 12, relationship setting unit 13, aggregation setting unit 14, aggregation target period definition unit 15, identifier assignment unit 16, comparison target period specification unit 17, calculation unit 18, and switching unit 19 may be configured in multiple computers.

[0026] One piece of hardware in the control unit 10 may be configured to execute multiple types of processing from among the main control unit 11, data input unit 12, relationship setting unit 13, aggregation setting unit 14, aggregation target period definition unit 15, identifier assignment unit 16, comparison target period specification unit 17, calculation unit 18, and switching unit 19, or it may be configured to execute a single processing by dividing it into smaller parts and coordinating multiple parts. Each of the above pieces of hardware is connected via an electrical communication path (wiring) such as bus B shown in Figure 2, enabling bidirectional or unidirectional communication as needed.

[0027] The following is an overview of each section, followed by a more detailed explanation of each section using specific examples.

[0028] The data input unit 12 is the part that accepts input of a data table having multiple attributes, including a date. For example, it can accept input of a data table by reading a data table stored in the storage unit 30.

[0029] The aggregation setting unit 13 is the part that can accept the setting of arbitrary attributes for the aggregation unit. The aggregation setting unit 13 sets the time series attributes of the aggregation unit to include at least the year attribute from among the multiple time series attributes included in the calendar table stored in the storage unit 30, and sets the attributes to be aggregated from among the multiple attributes included in the data table entered in the data input unit 12.

[0030] The aggregation period definition section 15 is the part that defines the aggregation period based on the time-series attributes of the aggregation unit set in the aggregation setting section 14.

[0031] The identifier assignment unit 16 is the part that assigns identifiers to determine the chronological order for the aggregation period defined in the aggregation period definition unit 15.

[0032] The comparison period identification unit 17 is the part that identifies the comparison period for comparing the aggregated values ​​of the aggregation period, based on the identifier assigned by the identifier assignment unit 16. If the difference period for comparison is entered, for example by a user, the comparison period identification unit 17 accepts the input of the difference period and identifies the target period for calculating the comparison value based on the accepted difference period and the identifier assigned by the identifier assignment unit 16.

[0033] The comparison period identification unit 17 can also accept input of the time series attributes to be compared, the difference period, and the common time series attributes. In this case, the identifier assignment unit 16 can assign an identifier that determines the time series order for each attribute value included in the time series attributes to be compared, based on the time series attributes to be compared, the time series attributes of the aggregated result table, and the calendar table set as the calendar master.

[0034] Furthermore, the comparison period identification unit 17 can identify candidate comparison periods for which to calculate the comparison value based on the difference period and the identifier assigned by the identifier assignment unit 16, and from among the identified candidate comparison periods, it can also identify a comparison period for which to calculate the comparison value based on common time series attributes.

[0035] The calculation unit 18 calculates aggregated values ​​by aggregating the attribute values ​​of the attribute to be aggregated for each aggregation period defined by the aggregation period definition unit 15, and also calculates comparison values ​​corresponding to the aggregated values ​​by aggregating the attribute values ​​of the attribute to be aggregated in the comparison period identified by the comparison period identification unit 17. For example, the calculation unit 18 calculates aggregated values ​​for each aggregation period based on the data table received by the data input unit 12, the time-series attributes included in the calendar table stored in the storage unit 30, and the time-series attributes of the aggregation unit set in the aggregation setting unit 13. In this case, the calculation unit 18 creates an aggregation result table that includes the time-series attributes and the aggregated values. When the calculation unit 18 has created an aggregation result table, the identifier assignment unit 16 can identify the time-series attributes included in the aggregation result table based on the aggregation result table created by the calculation unit 18 and the calendar table stored in the storage unit 30, and assign identifiers that determine the time-series order for each aggregation period defined by the aggregation period definition unit 15 based on the attribute values ​​included in the time-series attributes.

[0036] The calculation unit 18 calculates a comparison value between the aggregated value of the comparison period identified by the comparison period identification unit 17 and the aggregated value of the aggregation period. Once the calculation unit 18 calculates the comparison value corresponding to the aggregated value, the display control unit 3a displays the aggregated value and the comparison value corresponding to that aggregated value on the monitor 3 for each time-series attribute of the aggregation unit.

[0037] The comparison period identification unit 17 identifies candidate comparison periods for which to calculate the comparison value based on the difference period and the identifier assigned by the identifier assignment unit 16, and from among the identified candidate comparison periods, it identifies the comparison period for which to calculate the comparison value based on common time series attributes. In this case, the calculation unit 18 calculates a comparison value between the aggregated value of the comparison period identified by the comparison period identification unit 17 and the aggregated value of the aggregation period.

[0038] The calculation unit 18 can also calculate aggregate values ​​for each arbitrary attribute for each aggregation period based on the time-series attributes included in the calendar table and the time-series attributes and arbitrary attributes of the aggregation unit set in the aggregation setting unit 14, and create an aggregation result table that includes time-series attributes, arbitrary attributes, and aggregate values.

[0039] The switching unit 19 is the part that switches the time series attribute of the aggregation unit between the first time series attribute and the second time series attribute when the calendar table contains a first time series attribute and a second time series attribute which is a subordinate attribute of the first time series attribute.

[0040] For example, if the aggregation setting unit 14 sets a first time series attribute as the first aggregation unit from among multiple time series attributes included in the calendar table, and sets a second time series attribute, which is a subordinate attribute of the first time series attribute, as the second aggregation unit, the identifier assignment unit 16 identifies the first time series attribute included in the aggregation result table based on the aggregation result table and the calendar table set as the calendar master, assigns a first identifier that determines the time series order based on the attribute values ​​included in the first time series attribute, and further identifies the first and second time series attributes included in the aggregation result table based on the aggregation result table and the calendar table set as the calendar master, assigns a second identifier that determines the time series order based on the attribute values ​​included in the first and second time series attributes. In this case, the comparison period identification unit 17 identifies a first comparison period for which the comparison value is calculated based on the difference period and the first identifier assigned by the identifier assignment unit when the attribute of the aggregation unit is switched to the first time series attribute by the switching unit 19. On the other hand, when the attribute of the aggregation unit is switched to the second time series attribute by the switching unit 19, the comparison period for which the comparison value is calculated based on the difference period and the second identifier assigned by the identifier assignment unit.

[0041] Then, if the attribute of the aggregation unit is switched to the first time-series attribute by the switching unit 19, the calculation unit 18 calculates a first comparison value, which is a comparison value between the aggregated value of the first comparison period and the aggregated value of the aggregation period. On the other hand, if the attribute of the aggregation unit is switched to the second time-series attribute by the switching unit 19, the calculation unit 18 calculates a second comparison value, which is a comparison value between the aggregated value of the second comparison period and the aggregated value of the aggregation period.

[0042] Furthermore, the identifier assignment unit 16 can also assign identifiers that determine the time-series order for each attribute value included in the first time-series attribute, based on the first time-series attribute, the time-series attribute of the aggregation result table, and the calendar table set as the calendar master. In this case, the comparison target period identification unit 17, when the attribute of the aggregation unit is switched to the first time-series attribute by the switching unit 19, identifies a first comparison target period for calculating the comparison value based on the difference period and the identifier assigned by the identifier assignment unit. On the other hand, when the attribute of the aggregation unit is switched to the second time-series attribute by the switching unit 19, it identifies a second comparison target period for calculating the comparison value based on the difference period, the identifier assigned by the identifier assignment unit, and the second time-series attribute.

[0043] Then, if the attribute of the aggregation unit is switched to the first time-series attribute by the switching unit 19, the calculation unit 18 calculates a first comparison value, which is a comparison value between the aggregated value of the first comparison period and the aggregated value of the aggregation period. If the attribute of the aggregation unit is switched to the second time-series attribute by the switching unit 19, the calculation unit 18 calculates a second comparison value, which is a comparison value between the aggregated value of the second comparison period and the aggregated value of the aggregation period.

[0044] When the first and second comparison values ​​are calculated by the calculation unit 18, the display control unit 3a displays the switching unit 19 on the data analysis screen, and the switching unit 19 follows the switching of attributes of the aggregation unit and displays the first and second comparison values ​​on the monitor 3.

[0045] The display mode of the display control unit 3a is not limited to one mode. For example, it can also display each attribute value included in the first time series attribute and the attribute values ​​included in the second time series attribute grouped together on the monitor 3. In this case, the switching unit 19 can switch the attribute of the aggregation unit by opening and closing the grouping of the attribute values ​​included in the first time series attribute. The display control unit 3a then switches to the first time series attribute as the time series attribute of the aggregation unit if the grouping of the attribute values ​​included in the first time series attribute is closed, and switches to the second time series attribute as the time series attribute of the aggregation unit if the grouping of the attribute values ​​included in the first time series attribute is open.

[0046] Next, the processing flow performed by each part of the control unit 10 will be explained based on the flowchart shown in Figure 3. In step SA1 after the start, table data (calendar table and data table) is read from the storage unit 30 and table data input is accepted. Furthermore, in step SA1, aggregation settings are obtained from the GUI (also called "graphical user interface" or "UI") displayed on the monitor 3.

[0047] In step SA2, an identifier is created from the aggregation settings and calendar table obtained in step SA1 to identify the unit to which sequential numbers will be assigned. In step SA3, an intermediate table with sequential numbers assigned is created based on the calendar table and the identifier.

[0048] In step SA4, the raw data and the created intermediate table are joined based on the aggregation settings obtained in step SA1 to create the aggregation results. In step SA5, the row that each row of the aggregation results should reference is determined based on the serial number and the period-on-period setting. In step SA6, a comparison calculation is performed based on each row of the table data, the row to be referenced, and the period-on-period setting. In step SA7, the calculation results are added to the application table. In step SA8, various analyses are performed based on the calculation results obtained in step SA7.

[0049] Next, we will explain in detail the details of each process shown in Figure 3, using specific data examples. In step SA1, the data input unit 12 accepts input from a data table that has multiple attributes, such as dates. In this example, we will explain the case where two types of data provided from the base system are read. The first data is a data table (e.g., sales history data) that forms the basis of the visualized business indicators, as shown in Figure 4, and this data table always contains one time-series column. The second data is a calendar table in which actual time-series information is registered as master data, as shown in Figure 5, and can be represented as a date in YYYY-MM-DD format, which is part of the ISO 8601 standard. The calendar table has one column that can be considered as master data of actual dates, consisting of data that is unique and has no missing values. Furthermore, in addition to the column that can be considered as master data of actual dates, the calendar table has columns of data as attribute values ​​(time-series attributes) that can divide the data at a specific period granularity. For example, "month" and "year". However, the attributes do not need to be standard time-series attributes, and users can freely edit and add time-series attributes at any time, including adding attributes such as "quarterly" and "semi-annually."

[0050] Furthermore, the relation setting unit 13 sets up a relation between the calendar table and the data table based on the dates included in the calendar table shown in Figure 5 and the dates included in the data table shown in Figure 4. For example, it also sets up a 1:N (N⊆1 or greater natural number) automatic join between columns that can be considered as master data for actual dates among the multiple columns that make up the calendar table and columns that can similarly represent dates among the multiple columns that make up the data table (columns that can be considered equivalent).

[0051] In steps SA2 and beyond, an application is built on the base system to aggregate and visualize table data based on the data acquired in step SA1. For example, the display control unit 3a generates a graphical user interface 100 for registering setting items as shown in Figure 6 and displays it on the monitor 3. The user operates the keyboard 4 and mouse 5 on the graphical user interface 100 to define how to calculate the data created in step SA1 using aggregate functions such as sum and average, and registers these defined values ​​as setting items. In addition to aggregate functions, conditions for data extraction and arithmetic operations can also be defined.

[0052] After defining calculations using aggregate functions and arithmetic operations, values ​​with comparison settings are registered. When registering values ​​with comparison settings, the display control unit 3a generates a comparison setting user interface 110 and displays it on the monitor 3, as shown in Figures 7 and 8. The comparison setting user interface 110 is provided with a target setting area 111 for setting the values ​​to be compared and a period setting area 112 for setting the comparison period. The values ​​to be compared can be set by operating the target setting area 111 with a mouse 5 or the like. The values ​​that can be set as comparison targets are values ​​that you want to compare over specific periods, and in this example, you select the values ​​that have been registered as setting items.

[0053] The comparison period can be set by operating the period setting area 112 with the mouse 5 or similar. The comparison period is a specific period to be compared, and in this example, there are two options: period-over-period and year-over-year. Figure 7 shows the case where period-over-period is selected. Period-over-period is not limited to one period prior; it is also possible to compare with multiple periods prior, such as two or three periods prior. This period-over-period is the period to be compared. The periods that can be set as period-over-period are the column names of the registered data, such as "year" and "month," and these can be selected by the user as options. Furthermore, as will be described later, it is also possible to provide an option to automatically select the period to be compared, so that the data analysis device 1 can automatically select the period without the user having to perform a period selection operation.

[0054] The comparison setting user interface 110 shown in Figure 7 includes a comparison method setting area 113. In the comparison method setting area 113, it is possible to select the comparison method when comparing specific periods. For example, the comparison method setting area 113 includes options such as "Difference" to calculate the difference between the values ​​of the previous period and the current period, "Increase / Decrease Rate" to calculate a value that is (value of current period / value of previous period)-1, and "Value Only" to output the value of the previous period. When the user performs an operation to select the desired option with the mouse 5 or the like, the corresponding comparison method is set.

[0055] The comparison setting user interface 110 shown in Figure 7 includes a difference setting area 114 for setting the difference in the comparison period. In the difference setting area 114, the user sets how far apart the comparison periods should be using an integer value of 1 or greater. This is the setting of the difference period, and it is set by the user entering a numerical value into the difference setting area 114.

[0056] Figure 8 shows the case where year-on-year comparison is selected. In this specification, we refer to it as year-on-year comparison, and for convenience, we display it as "previous year," but it is not limited to "year." Year-on-year comparison is a setting item for comparing with values ​​from the same period in the past, and the following two options are selected as the comparison period. Specifically, the comparison period can be set using the period setting area 115 provided in the comparison setting user interface 110 shown in Figure 8. The first option is the past period to compare. The period that can be set at this time is the column name of the registered data, such as "year" or "month," and these can be selected by the user as options (time series attribute of the comparison target). The second option is the period to reference, and the period that can be set at this time can be the column name of the registered data. There is also an option to automatically select this comparison period (common time series attribute). The comparison setting user interface 110 shown in Figure 7 is also provided with a comparison method setting area 113 and a difference setting area 114.

[0057] In this embodiment, it is also possible to set the location to be aggregated. Specifically, the user can set which location will be used as the aggregation period on the graphical user interface displayed on monitor 3. For example, in the case of a table application with two-dimensional data (matrix data), the user can select "the bottom position of the row settings".

[0058] If you want to add more than one value, simply repeat the comparison settings operation described above. For example, it is possible to add two or more period-over-period or year-over-year comparisons to a single application simultaneously. Furthermore, even if you create an application with two or more period-over-period values, the aggregation process described later can be executed simultaneously.

[0059] Furthermore, if you want to display multiple values ​​for either the previous period-over-period or the same period last year, that is, multiple values ​​of the same type, you can create columns for each aggregation target (e.g., number of units sold, sales amount) in the intermediate table, allowing you to specify the comparison target within a single intermediate table. For example, if there is a mix of previous period-over-period and same period last year-over-period data, an intermediate table with multiple identifiers will be created. The first identifier can be used for month-over-month comparisons, the second for year-over-year comparisons, and the third for quarter-over-quarter comparisons, for example.

[0060] Next, the details of the process by which the aggregation setting unit 13 sets the attributes to be aggregated will be described. When setting the attributes to be aggregated, the display control unit 3a displays the aggregation setting user interface 120 shown in Figure 9 on the monitor 3. The aggregation setting user interface 120 is provided with a column setting area 121 and a row setting area 122. The column setting area 121 and the row setting area 122 are used to register time-series data in accordance with the comparison period set as described above. At this time, other dimensions may be added.

[0061] After the attributes to be aggregated are set, the data for analysis is read based on these settings. First, the data is constructed based on the calculation of values ​​other than those registered in the comparison settings above and the dimension settings (year / quarter / month, etc.) registered in the application. Figure 10 explains the concept of aggregation processing when comparing with past periods, and shows a calendar table, sales history (an example of a data table), and a first intermediate table created based on the calendar table and sales history. The calendar table and sales history can be joined using the "date" column, which has the same value. The first intermediate table is an aggregation result table and is a table that includes the "year," "half-year," and "month" columns of the calendar table, and the "flag" and "sales" columns of the sales history. When joining the calendar table and sales history, aggregation processing is performed based on the set dimensions, and the first intermediate table is created by performing aggregation processing.

[0062] Subsequently, as shown in Figure 11, from the output data, only the columns containing the actual time-series information registered as master data are selected, an aggregation operation is performed on those columns, the data is sorted in ascending order by the column that can be considered as the master data for the actual date, and a sequential number assigned in ascending order is added as a column to prepare the reconstructed data. This sequential number is arranged in chronological order by the above process. The intermediate table created here is created one-to-one with the created values, and the value of the previous period with different settings refers to a different intermediate table. In this process, only the columns included in the calendar table are selected from the set dimensions, and duplicate data is automatically deleted. Furthermore, a second intermediate table is obtained by calculating and selecting the minimum value of the date column.

[0063] For the second intermediate table, sequential numbers are assigned in ascending order based on the minimum value of the date column, and these numbers are added to the column. Finally, the minimum value of the date column is removed to obtain the third intermediate table.

[0064] As shown in Figure 12, the aggregation operations for the first and third intermediate tables are performed based on the data contained in the calendar table. For example, even if there is no data for "May" and "June," the system identifies the target time-series attribute from the calendar table and performs the aggregation operation on the data in that calendar table to ensure that there are no missing months.

[0065] After the aggregation operation is performed, a comparison calculation process is carried out based on the set comparison settings, as shown in Figure 13. In the comparison calculation process, for a given value, the value in the row that satisfies the predetermined conditions is determined to be the reference value. In this example, the order of the periods is determined by the set serial number. The row to be referenced is determined using this serial number and the difference between the set periods. For example, if the difference between the set periods is "1" and the serial number is "21", the row of the previous period to be referenced will be "20".

[0066] For example, the "flag" column shown in Figure 14 represents a dimension other than the value and is not included in the calendar table. In cases where there is a column not included in the calendar table, the system will refer to items with the same value in that column.

[0067] Furthermore, the "Month" column shown in Figure 14 represents a dimension other than the value, but it is included in the calendar table. In this case, for example, if "Use the rightmost position as the aggregation target" is set, the time series column to be aggregated will be the "Month" column. If the column to be aggregated determined here is the same period as the set period-to-previous-period comparison or comparison period, the aggregation will be performed from there; otherwise, the process of filling in missing values ​​will be executed (see Figure 15). Note that, as shown in Figure 16, if an option for automatic selection is selected in the period setting area 112, the aggregation will be performed as is with the determined aggregation target.

[0068] Values ​​that satisfy the set conditions are calculated according to the set calculation method and output as comparison values, as shown in Figure 17. Rows that do not exist in the data are deleted. For example, as shown in Figure 18, rows where the value referenced by the added period-over-period setting is a missing value are deleted.

[0069] Figure 19 illustrates the concept of aggregation processing when comparing with the same period in the past. It shows a calendar table, a sales history (an example of a data table), and a first intermediate table created based on the calendar table and the sales history. The calendar table and the sales history are joined using the "Date" column, which has the same value.

[0070] First, calculations are performed for values ​​other than those registered based on the comparison settings. Then, when joining the calendar table and the sales history, aggregation processing is performed based on the set dimensions, and by performing aggregation processing, the first intermediate table is created.

[0071] Subsequently, as shown in Figure 20, only the columns included in the past period to be compared (time-series attributes to be compared) are selected from the output data, an aggregation operation is performed on those columns, the data is sorted in ascending order by the column that can be considered as the master data for the actual date, and a new data set is prepared by adding a sequential number assigned in ascending order as a column. This sequential number is arranged in chronological order by the above process. The intermediate table created here is created one-to-one with the created values, and the value of the previous period with different settings refers to a different intermediate table. In this process, only the columns included in the calendar table are selected from the set dimensions, and duplicate data is automatically deleted. Furthermore, a second intermediate table is obtained by calculating and selecting the minimum value of the date column.

[0072] For the second intermediate table, sequential numbers are assigned in ascending order based on the minimum value of the date column, and these numbers are added to the column. Finally, the minimum value of the date column is removed to obtain the third intermediate table.

[0073] As shown in Figure 21, the aggregation operation of the first intermediate table and the third intermediate table is performed using an outer join, with the join condition using only the columns included in the calendar table, which contains actual time-series information registered as master data, and values ​​other than those registered by the comparison settings.

[0074] After the aggregation operation is performed, a comparison calculation process is carried out based on the set comparison settings, as shown in Figure 22. In the comparison calculation process, for a given value, the value in the row that satisfies the predetermined conditions is determined to be the reference value. In this example, the order of the periods is determined by the set serial number. The row to be referenced is determined using this serial number and the difference between the set periods. For example, if the difference between the set periods is "1" and the serial number is "21", the row of the previous period to be referenced will be "20".

[0075] The "flag" column shown in Figure 23 represents a dimension other than the value and is not included in the calendar table. In cases where there is a column not included in the calendar table, the reference target will be the one with the same value in that column.

[0076] Furthermore, the "Month" column shown in Figure 23 represents a dimension other than the value, but it is the column set as the past period to be compared. If a column is set as the past period to be compared, only items with the same value in that column will be referenced. If there is no past period to compare, it will be excluded from the aggregation.

[0077] When referencing values, the columns to be aggregated are those not included in the calendar data representing dimensions other than values, and those other than the columns representing values ​​other than dimensions, based on the setting of the location to be aggregated. For example, if it is set to "aggregate the rightmost position", the time series column to be aggregated will be the "Month" column. If the column to be aggregated determined here is the same as the reference period, the aggregation will be performed; otherwise, the process of filling in missing values ​​will be executed (see Figure 24). Note that, as shown in Figure 25, if the option to automatically select is selected in the period setting area 115, the aggregation will be performed as is with the determined aggregation target.

[0078] Values ​​that satisfy the set conditions are calculated according to the set calculation method and output as comparison values, as shown in Figure 26. Rows that do not exist in the data are deleted. For example, as shown in Figure 27, rows where the value referenced by the added period-over-period setting is a missing value are deleted.

[0079] Figure 28 is a simplified diagram illustrating the overall data analysis process performed using the data analysis device 1. As shown in this figure, the relationship setting unit 13 sets a relationship between the calendar table and the data table. The aggregation setting unit 14 performs processing on the calendar table and sets the time series attribute of the aggregation unit from among the time series attributes contained in the calendar table. The aggregation target period definition unit 15 defines the aggregation target period based on the time series attribute of the aggregation unit, and furthermore, the identifier assignment unit 16 assigns a sequential number as an identifier that determines the time series order for the aggregation target period. By performing these processes, a calendar table with sequential numbers assigned and duplicates removed is obtained.

[0080] Furthermore, the aggregation setting unit 14 performs aggregation processing on the data table and sets the attributes to be aggregated from among the attributes included in the data table. By joining the data table with the attributes to be aggregated set with the calendar table, an aggregation result table is obtained. If the previous period value setting 1, previous period value setting 2, previous period value setting N, etc. have been set, the calculation unit 18 calculates the aggregated value for each aggregation target period, aggregates the attribute values ​​of the attributes to be aggregated in the comparison target period, and calculates the comparison value corresponding to the aggregated value. In this way, a final result that allows for comparison between periods is obtained.

[0081] Figure 29 illustrates the switching of the display due to the opening and closing operation of attribute value grouping. This Figure 29 shows the graphical user interface 150 generated by the display control unit 3a and displayed on the monitor 3, and corresponds to the data analysis screen. The graphical user interface 150 is provided with an annual display area 151, a semi-annual display area 152, a quarterly display area 153, and a monthly display area 154. The annual display area 151, semi-annual display area 152, quarterly display area 153, and monthly display area 154 are included in the switching unit 19 and are displayed on the graphical user interface 150 by the display control unit 3a. In Figure 29, "H1" is the first half of the year, and "H2" is the second half of the year. Also, in Figure 29, "Q1" is the first quarter, "Q2" is the second quarter, "Q3" is the third quarter, and "Q4" is the fourth quarter. Also, in Figure 29, "2020", "2021", and "2022" are the fiscal years. Furthermore, in Figure 29, "4" represents April, "5" represents May, and "6" represents June.

[0082] When the user closes the monthly view area 154 using the mouse (5, etc.), the display switches from monthly to quarterly. When the user closes the quarterly view area 153, the display switches from quarterly to semi-annual. When the user closes the semi-annual view area 152, the display switches from semi-annual to yearly.

[0083] Conversely, opening the yearly display area 151 switches the display from yearly to semi-annual. Opening the semi-annual display area 152 switches the display from semi-annual to quarterly. Opening the quarterly display area 153 switches the display from quarterly to monthly. In this way, the switching unit 19 follows the switching of the attributes of the aggregation unit and allows the monitor 3 to display the first comparison value, which is the comparison value between the aggregated value of the first comparison period and the aggregated value of the aggregation target period, and the second comparison value, which is the comparison value between the aggregated value of the second comparison period and the aggregated value of the aggregation target period.

[0084] Year, quarter, semi-annual, and month are time-series attributes, respectively. The display control unit 3a can group each attribute value included in, for example, a semi-annual (an example of the first time-series attribute) and the attribute values ​​included in a quarter (an example of the second time-series attribute) and display them on the monitor 3. The switching unit 19 can switch the attribute of the aggregation unit by opening and closing the grouping of the attribute values ​​included in a semi-annual. If the grouping of the attribute values ​​included in a semi-annual is closed, the display control unit 3a switches to the semi-annual display format as the time-series attribute of the aggregation unit, and if the grouping of the attribute values ​​included in a semi-annual is open, it switches to the quarter as the time-series attribute of the aggregation unit. The same applies to years, months, etc.

[0085] The final results obtained as described above can also be visualized in the application. Visualization means displaying the final results on monitor 3 as a pie chart, bar graph, line graph, etc.

[0086] The embodiments described above are merely illustrative in all respects and should not be interpreted restrictively. Furthermore, any modifications or changes that fall within the equivalent scope of the claims are all within the scope of the present invention. [Industrial applicability]

[0087] As explained above, the data analysis device related to this disclosure can be used to analyze various types of data. [Explanation of symbols]

[0088] 1. Data analysis device 3. Monitor (display unit) 3a Display Control Unit 12. Data Input Section 13. Relationship Setting Section 14 Aggregation Settings Section 15. Section defining the period for aggregation. 16 Identifier Assignment Unit 17. Identifying the Comparison Period 18 Calculation Section 19 Switching section 30 Storage section

Claims

1. In a data analysis device that analyzes data, A storage unit that stores a calendar table having multiple dates and corresponding time-series attributes for each date, A data input unit that accepts input from a data table having multiple attributes including a date, A relationship setting unit sets a relationship between the calendar table and the data table based on the dates included in the calendar table stored in the storage unit and the dates included in the data table entered in the data input unit. The aggregation setting unit sets the time-series attributes of the aggregation unit from among the time-series attributes included in the calendar table stored in the storage unit, and sets the attributes to be aggregated from among the attributes included in the data table entered in the data input unit. A defined aggregation period unit defines the aggregation period based on the time-series attributes of the aggregation unit set in the aggregation setting unit, An identifier assignment unit assigns identifiers that determine the chronological order to the aggregation period defined in the aggregation period definition unit, A comparison period identification unit identifies a comparison period for comparing aggregated values ​​of the aggregation period based on the identifier assigned by the identifier assignment unit, A calculation unit calculates an aggregated value by aggregating the attribute values ​​of the attribute to be aggregated for each aggregation period defined in the aggregation period definition unit, and also aggregates the attribute values ​​of the attribute to be aggregated during the comparison period specified in the comparison period specification unit, and calculates a comparison value corresponding to the aggregated value. A display control unit that displays the aggregated value and the corresponding comparison value on the display unit for each time-series attribute of the aggregation unit, A data analysis device equipped with the following features.

2. In the data analysis apparatus according to claim 1, The storage unit stores a calendar table having multiple time-series attributes, including at least a year attribute. The aggregation setting unit sets the time series attribute of the aggregation unit to include at least the year attribute from among a plurality of time series attributes included in the calendar table stored in the storage unit, and sets the attribute to be aggregated from among a plurality of attributes included in the data table entered in the data input unit, in the data analysis device.

3. In the data analysis apparatus according to claim 2, The storage unit stores a calendar table as time-series attributes, which includes a year attribute as the first level attribute, a semi-annual attribute as the second level attribute, a quarter attribute as the third level attribute, and a month attribute as the fourth level attribute.

4. In the data analysis apparatus according to claim 3, The aforementioned storage unit is a data analysis device that stores a calendar table configured so that there are no duplicates or missing entries for each date.

5. In the data analysis apparatus according to claim 1, The calculation unit calculates aggregated values ​​for each aggregation period based on the data table received by the data input unit, the time-series attributes included in the calendar table stored in the storage unit, and the time-series attributes of the aggregation unit set by the aggregation setting unit, and creates an aggregation result table that includes the time-series attributes and aggregated values.

6. In the data analysis apparatus according to claim 5, The identifier assignment unit identifies the time-series attributes included in the aggregated result table based on the aggregated result table created by the calculation unit and the calendar table stored in the storage unit, and assigns identifiers that determine the time-series order for each aggregation target period defined by the aggregation target period definition unit based on the attribute values ​​included in the time-series attributes. The comparison period identification unit receives input of a difference period and identifies the period for which the comparison value is calculated based on the received difference period and the identifier assigned by the identifier assignment unit. The calculation unit is a data analysis device that calculates a comparison value between the aggregated value of the comparison period identified by the comparison period identification unit and the aggregated value of the aggregation period.

7. In the data analysis apparatus according to claim 1, The comparison period identification unit accepts input of the time series attributes of the comparison target, the difference period, and the common time series attributes, The identifier assignment unit is a data analysis device that assigns identifiers to each attribute value included in the time series attribute of the time series attribute to be compared, based on the time series attribute of the time series attribute of the time series attribute of the comparison target, the time series attribute of the aggregated result table, and the calendar table, thereby determining the time series order.

8. In the data analysis apparatus according to claim 7, The comparison period identification unit identifies candidate comparison periods for which a comparison value is calculated based on the difference period and the identifier assigned by the identifier assignment unit, and from among the identified candidate comparison periods, it identifies a comparison period for which a comparison value is calculated based on common time series attributes. The calculation unit is a data analysis device that calculates a comparison value between the aggregated value of the comparison period identified by the comparison period identification unit and the aggregated value of the aggregation period.

9. In the data analysis apparatus according to claim 1, The aggregation setting unit accepts the setting of arbitrary attributes for the aggregation unit, The calculation unit calculates aggregate values ​​for each arbitrary attribute for each aggregation period based on the time-series attributes included in the calendar table and the time-series attributes and arbitrary attributes of the aggregation unit set by the aggregation setting unit, and creates an aggregation result table that includes the time-series attributes, arbitrary attributes and aggregate values, as a data analysis device.

10. In the data analysis apparatus according to claim 2, The calendar table stored in the memory unit includes a first time series attribute and a second time series attribute which is a subordinate attribute of the first time series attribute. A data analysis device further comprising a switching unit that switches the time series attribute of the aggregation unit using the first time series attribute and the second time series attribute.

11. In the data analysis apparatus according to claim 10, The aggregation setting unit is, From among the multiple time series attributes included in the calendar table stored in the memory unit, a first time series attribute is set as the first aggregation unit, and a second time series attribute, which is a subordinate attribute of the first time series attribute, is set as the second aggregation unit. The identifier assigning unit, Based on the aggregated results table and the calendar table, a first time-series attribute included in the aggregated results table is identified, and a first identifier is assigned to determine the time-series order based on the attribute values ​​included in the first time-series attribute. Based on the aggregated results table and the calendar table, the first and second time-series attributes included in the aggregated results table are identified, and a second identifier is assigned to determine the time-series order based on the attribute values ​​included in the first and second time-series attributes. The aforementioned comparison period identification unit is: When the attribute of the aggregation unit is switched to the first time-series attribute by the switching unit, a first comparison period is identified for calculating the comparison value based on the difference period and the first identifier assigned by the identifier assignment unit. When the attribute of the aggregation unit is switched to the second time-series attribute by the switching unit, a second comparison period is identified for calculating the comparison value based on the difference period and the second identifier assigned by the identifier assignment unit. The calculation unit described above, A data analysis device that, when the attribute of the aggregation unit is switched to a first time-series attribute by the switching unit, calculates a first comparison value which is a comparison value between the aggregated value of the first comparison period and the aggregated value of the aggregation period, while when the attribute of the aggregation unit is switched to a second time-series attribute, calculates a second comparison value which is a comparison value between the aggregated value of the second comparison period and the aggregated value of the aggregation period.

12. In the data analysis apparatus according to claim 10, The aggregation setting unit is, From among the multiple time series attributes included in the calendar table stored in the memory unit, a first time series attribute is set as the first aggregation unit, and a second time series attribute, which is a subordinate attribute of the first time series attribute, is set as the second aggregation unit. The identifier assigning unit, Based on the first time series attribute, the time series attribute of the aggregated result table, and the calendar table, an identifier is assigned to each attribute value included in the first time series attribute to determine the time series order. The aforementioned comparison period identification unit is: When the attribute of the aggregation unit is switched to the first time-series attribute by the switching unit, a first comparison period is identified for calculating the comparison value based on the difference period and the identifier assigned by the identifier assignment unit. When the attribute of the aggregation unit is switched to the second time series attribute by the switching unit, a second comparison period is identified for calculating the comparison value based on the difference period, the identifier assigned by the identifier assignment unit, and the second time series attribute. The calculation unit described above, When the attribute of the aggregation unit is switched to the first time-series attribute by the switching unit, a first comparison value is calculated, which is the comparison value between the aggregated value of the first comparison period and the aggregated value of the aggregation period. A data analysis device that, when the attribute of the aggregation unit is switched to a second time-series attribute by the switching unit, calculates a second comparison value which is a comparison value between the aggregated value of the second comparison period and the aggregated value of the aggregation period.

13. In the data analysis apparatus according to claim 12, The display control unit displays the switching unit on the data analysis screen, and the display unit displays the first comparison value and the second comparison value in accordance with the switching of attributes of the aggregation unit by the switching unit, in a data analysis device.

14. In the data analysis apparatus according to claim 13, The display control unit groups each of the attribute values ​​included in the first time series attribute and the attribute values ​​included in the second time series attribute and displays them on the display unit. The switching unit switches the attributes of the aggregation unit by opening and closing the grouping of attribute values ​​included in the first time-series attribute. The display control unit switches to the first time series attribute as the time series attribute of the aggregation unit if the grouping of attribute values ​​included in the first time series attribute is closed, and switches to the second time series attribute as the time series attribute of the aggregation unit if the grouping of attribute values ​​included in the first time series attribute is open.

Images