JPS5846738B2 - Time series processing method for table format data - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a time-series processing method for tabular data, and in particular, a data storage unit that stores aggregated data in a table format, and a system that processes, edits, analyzes and outputs stored data. The present invention relates to a processing method that enables time-series processing and non-time-series processing to be operated in the same manner in a data processing system that has a data processing unit and whose data processing target is time-series processing of aggregated data.

Much data handled in the business field is in table format (two-dimensional or three-dimensional array), such as product sales by office.

However, conventional table format data processing lacks support for data with time elements (time series data), and it is necessary to consider time mapping and time series-specific editing report processing each time.

For example, if product sales by sales office are prepared on MT (magnetic tape) on a monthly basis, a comparison of sales by sales area for a specific month (non-chronological processing) and sales of a specific sales office If you want to create a transition (time series processing), you need to prepare MT (magnetic tape) that stores monthly sales figures, and create a program to edit them for each purpose before processing. Ta.

In addition, in the conventional time series processing method, the target data was stored in a one-dimensional array (vector) and was a dedicated system that only performed time series-specific processing. cannot be stored.

(b) Processing data cannot be grouped and handled.

Due to these drawbacks, it was difficult to use it in the business field.

The present invention solves the above problems, allows time-series processing and non-time-series processing to be performed in the same manner on data stored in table format, and allows time-series-specific processing to be easily summarized. The present invention aims to realize a processing method that can be manipulated, and for this purpose, the present invention has a data storage unit that stores aggregated data in a table format, and a data processing unit that processes, edits, analyzes and outputs the stored data. In a data processing system whose data processing target is time-series processing of aggregated data, a time-series attribute is defined on one axis of the table stored in the data storage unit, and a time series attribute is defined in one axis of the table stored in the data storage unit. A time-series processing unit that executes time-series-specific processing and a non-time-series processing unit that executes processes unrelated to time series are provided separately, and the time-series processing unit that executes processing that is unique to time series and the non-time series processing unit that executes processing that is not related to time series is provided. The present invention is characterized in that sequential/non-time-series processing is sorted, and either the above-mentioned time-series processing section or non-time-series processing section is activated.

Hereinafter, the present invention will be explained with reference to the drawings.

Figure 1 is an example of non-time series table format data, which is referred to by the table name A, and Figure 2 is an example of time series table format data, which is referred to by the table name B. .

Furthermore, in the case of time series data, time series attribute information as shown in FIG. 3 is given to each table.

The data processing system according to the embodiment of the present invention processes the various table-format data as described above using the configuration shown in FIG.

In FIG. 4, 1 is an input device into which the contents of data processing and a table name indicating the processing target are input, and 2 is an input information interpreter which interprets the input information to determine the processing contents and target data. 3 is a processing data reference unit that retrieves the target data from the table format data storage file, 4 is a table format data storage file that stores table format data regardless of whether it is in time series. , 5 is a section for distributing processing contents and starts a processing section that requests processing depending on the processing contents, 6
7 is a non-time series processing input section into which specific data processing contents for non-time series processing of the non-time series processing section 10 is input; 7 is a time series processing input section; 7 is a time series processing input section;
8 is an arithmetic processing input section where specific data processing contents for time series processing are input, 9, 11, and 14 are attribute checks. 10 is a non-time series processing unit, which checks the existence and type of time series attributes and the data attributes required for processing preset in each processing unit, and specifically performs cross-tab analysis. 12 is a time series processing unit, which specifically performs time series-specific processing such as seasonal adjustment, and 13 and 15 are error processing units, which handle data attributes required by each processing unit. 16 is a calculation content distribution unit that calls the calculation unit that requests calculation processing depending on the calculation content. 17 is a time-series synchronization calculation unit that determines the time correspondence. 18 is a non-time series calculation unit that performs calculation processing on corresponding elements according to the data structure; 19 is an output device that outputs the processing contents; be.

Note that although the target data is the same in the attribute checking mechanisms 9, 11, and 14, the actual checking contents are different because the processing contents are different.

An example of operation will be explained below.

First, if it is desired to output the sales amount for each item in the Tokyo area in the table shown in FIG. 1, information instructing to that effect is inputted from the input device 1.

After interpreting the instruction, the input information interpreting unit 2 activates the processed data reference unit 3.

Based on the input instructions, the processed data reference unit 3
Required data is read from the table-format data storage file 4, and the processing contents are sent to the section 5.

The processing content distribution unit 5 analyzes the instruction content regarding the processing, and in this case, since the processing is related to non-time-series data, the non-time-series processing input unit 6 is activated to read data from the table-format data storage file 4. Send the data.

Thereafter, required processing is executed in the non-time series processing section 6, the attribute checking mechanism 9, and the non-time series processing section 10, and the processing results are output to the output device 19.

Next, for example, if you want to output the annual sales of televisions in the table shown in FIG. , the time series processing input section 7 is activated, and the data read out from the table format data storage file by the processed data reference section 6 is sent out.

Thereafter, necessary processing is executed in the time series processing input section 7 attribute check mechanism 11 and the time series processing section 12, and the processing results are output to the output device 19.

Next, for example, when creating a new time series table by adding the contents of two types of time series tables as shown in FIG. When input, table format data storage file 4
The data read out is processed through the processing route of the processing content allocation section 5, the calculation processing input section 8, the attribute checking mechanism 14, the calculation content allocation section 16, and the time series synchronization calculation section 17.

The processing results are then output to the output device 19 in the same manner as above.

As explained above, according to the present invention, one axis of table-format data has a time series attribute, and processing specific to time series (seasonal adjustment, year-on-year comparison, quarterly, half-yearly, monthly data,
(year editing, etc.), report output, and calculations that associate time with each table or part of the table. Furthermore, calculations with table-format data that does not have time-series attributes should be performed by the processing system. Since these are performed separately, it is possible to perform data processing on a table-type data structure, which is the easiest to understand, without having to be aware of time-series/non-time-series processing.

For example, if you edit the monthly MT of product sales by sales office into table format data in a three-dimensional array (three axes: sales office, product, and month), you can specify the direction (axis) of the data to be extracted. By simply doing this, you can separate time-series/non-time-series processing.

Therefore, according to the present invention, it is possible to bring about an excellent effect in that even end users in the business field who do not have knowledge of equivalence can easily use the system.

[Brief explanation of drawings]

Figure 1 is an example of non-time series table format data, Figure 2 is an example of time series table format data, Figure 3 is an example of time series attribute information, and Figure 4 is data processing of an embodiment according to the present invention. The system configuration diagram, FIG. 5, is an example of time series table calculation. In FIG. 4, 1 is an input device, 4 is a table format data storage file, 5 is a processing content distribution section, 6 is a non-time series processing input section, 7 is a time series processing input section, and 8 is an arithmetic processing input section. , 10 is a non-time series processing unit, 12 is a time series processing unit, 1
7 is a time-series synchronous calculation section, 18 is a non-time-series calculation section, and 19 is an output device.

Claims (1)

[Claims] 1. It has a data storage unit that stores aggregated data in a table format, and a data processing unit that stores, edits, analyzes and outputs the stored data, and the data processing target is time-series processing of aggregated data. In the data processing system, one axis of the table stored in the data storage unit defines and has a time series attribute, and the data processing unit includes a time series processing unit that executes processing specific to the time series. A non-time-series processing unit that executes processing unrelated to the above is provided separately, and the time-series processing unit A time-series processing method for table-format data, characterized in that one of the following and a non-time-series processing unit is activated.