JP6596531B2 - Data display system and program - Google Patents

Description

  The present invention relates to a data display system and a program configured by a computer that executes a process for displaying a list of a plurality of time-series data. For example, the present invention relates to a plurality of recording formats provided from different data recording / collection systems. This can be used when displaying a list of time-series data.

  2. Description of the Related Art Conventionally, processing for displaying a list of a plurality of time series data related in terms of content has been performed in various fields. For example, there is known a plant operation training simulation apparatus that displays attributes of simulated data and reference data collected in time series as event data so that they can be compared along the time axis (see Patent Document 1).

  In addition to expressing the recorded contents at the time of delivery in a vertical graph, the graph expression can also be expressed equivalent to paper operation using icons that change according to the fill and implementation contents, and the time series expressed in the graph There is known a partogram display device that displays nursing records and patient schedules by matching the positions of each other (see Patent Document 2).

Japanese Patent Laying-Open No. 2015-230325 (FIG. 2) Japanese Patent Laying-Open No. 2006-57981 (FIG. 8)

  As described above, in a conventional list display of a plurality of time-series data, each time-series data is simply displayed side by side so as to correspond to the time axis using time as a key. For example, in the plant operation training simulation apparatus described in Patent Document 1 described above, the simulation data and the reference data collected in time series are displayed in comparison with each other along the time axis, and each time series data is configured. Since each data is originally recorded at the same time interval, time series data in the same recording format is simply arranged along the time axis with time as a key. In other words, each piece of data constituting each time series data is used as it is and is displayed while being shifted with respect to the time.

  Further, in the partogram display device described in Patent Document 2 described above, different time-series data (nursing records and event data) are displayed in a list using time as a key. The individual data making up the series data are simply used and rearranged for display.

  Therefore, in the conventional list display of a plurality of time series data, the individual data constituting each time series data is simply used as it is, so each time series data is configured on the time axis. When the individual data is in a one-to-many or many-to-one relationship, when viewed from one side, the other side has a plurality of corresponding data or no corresponding data. For this reason, it may not be possible to say that an easy-to-read list display is provided.

  For example, there may be a case in which the display area of one of the time series data has a large number of blanks, or the display of both time series data does not substantially correspond and the list display body is not formed. In the former case, for example, the recording time of one time-series data is 1 minute intervals such as 8:00, 8:01, 8:02, etc., and the recording time of the other time-series data is 8 0:00, 8:10, 8:20,..., And so on, if the interval is 10 minutes, the other time-series data display portion has a large number of blanks. In the latter case, for example, the recording time of one time-series data is an interval of 10 minutes such as 8:00, 8:10, 8:20, 8:30,. If the time is 8:05, 8:15, 8:25, 8:35, and so on, even if the time is recorded at the same time interval of 10 minutes, the time is completely Since there is a shift, the corresponding data is completely lost, and the list display becomes very difficult to see.

  In the conventional list display of a plurality of time series data, when arranging each time series data so as to correspond to the time axis using the time as a key, the reference time is any time series. Since it is a recording time for data, there is currently no concept of displaying a list using a reference time different from any of them.

  However, in the case of the latter case of the inconvenience caused by the conventional list display described above, even if the same time interval of 10 minutes is recorded, the time is shifted, so the list display body is not formed. However, for example, data display at intervals of 5 minutes is performed, for example, 8:00, 8:05, 8:15, 8:20, 8:25, 8:30, 8:35, and so on. (That is, the concept of displaying a list using a reference time different from any of the recording times for a plurality of time-series data to be displayed as a list), and supplementing the display of a portion having no corresponding data If possible, it is possible to realize an easy-to-view list display. Therefore, if various list displays are possible, a user-friendly system can be realized.

  An object of the present invention is to provide a data display system and a program that are easy to see and can display various lists.

  <Basic configuration of the present invention>

The present invention is a data display system configured by a computer that executes processing for displaying a list of a plurality of time-series data,
Multiple time-series data with different recording formats of time-series information to be displayed as a list display unit time-series data consisting of a set of data recorded at regular time intervals, or over the recorded time zone Time-series data storage means for storing time-series data in at least one recording format among time-series data in a start-end format consisting of a set of data having an arbitrary time length in which the content indicated by the data is handled continuously ,
Display definition storage means for storing display definition information including unified unit time for displaying a list of the contents of the plurality of time series data stored in the time series data storage means at a unified time interval;
Using each data constituting the time series data, a generation / shaping process for generating deficient data among the data corresponding to each time set by dividing by the unit time unit stored in the display definition storage means is executed. In addition, when multiple data of each data constituting the time series data is included in the unified unit time, these multiple data are set as candidate data, and the unified unit time is selected from the multiple candidate data. Shaping means for executing selective shaping processing for selecting representative data to be data corresponding to each time set by dividing by,
Output means for executing a list display process for displaying each data constituting the time-series data after the shaping process by the shaping means in a state corresponding to each time set by dividing by a unified unit time. It is characterized by.

  Here, “time-series data in unit time format consisting of a set of data recorded at regular time intervals” means data recorded at a plurality of points (a plurality of times) distributed at equal intervals on the time axis. Is a set of However, there may be a time zone in which data is missing due to an error, power-off, etc., and any recording format that is basically recorded at regular intervals may be used. In addition, when a certain amount of time has passed, the recording time interval is also included in the “certain time interval” here. For example, at night, the time interval is five times as large as the daytime. Even in the case of night time, since it is a constant time interval, it is included in the “constant time interval” here.

  In addition, “the time-series data in the start / end format consisting of a set of data having an arbitrary time length in which the content indicated by the data continues over the recorded time zone” is not a point on the time axis. This means a set of data recorded in a time zone having a length, and each piece of data is treated as if the same contents are continued for all sections from the start of recording to the end of recording.

  Further, since there is “time-series data storage means for storing time-series data in at least one recording format”, the time-series data storage means is [1] time series in two or more unit time formats with different recording time intervals. When storing data, [2] When storing time series data in at least one unit time format and at least one time series data in start / end format, and [3] Time series data in two or more start / end formats May be stored.

  In the above, [1] it differs from the case of storing time series data in two or more unit time formats with different recording time intervals, for example, in units of 1 second, 1 minute, 5 minutes, etc. This means that time-series data recorded at time intervals is stored. In this case, it is not necessary that all the time series data is recorded at different time intervals, and it is only necessary to include time series data recorded at different time intervals. Therefore, for example, a case where three 1-second time-series data, two 1-minute time-series data, and one 5-minute-unit time-series data are stored as a list display target is included.

  In the above, [3] The case of storing time series data of two or more start / end formats is as follows. First, the “start / end format” has four formats for recording the time zone of each data. The recording start time (the time is the time indicated by year / month / day / hour / minute / second, etc.). However, the minimum unit of recording, such as up to hour, minute, second or hundredth of a second, is arbitrary.) And recording with a combination of two types of information of recording time length Format A, format B for recording with a combination of two types of information, recording end time and time length, format C for recording with a combination of two types of information, recording start time and recording end time, and recording start There is a format D in which recording is performed with a combination of three types of information: the time of recording, the time of recording termination, and the time length. Therefore, there are cases where these four formats A, B, C, and D are time-series data of different formats (for example, when time-series data of format A and format B coexist).

  Next, not only when the formats A, B, C, and D are different, but also when the time series data of the format A is 2 or more, when the time series data of the format B is 2 or more, the time series data of the format C is 2 In the above case, the case where the time series data of format D is 2 or more is also included. Even in the case of formats A, B, C, or D, the time length of individual data is arbitrary and varied. There are few cases where the data is separated by the exact time (that is, recorded at the time completely synchronized), and individual data separation times (start time and end time) are different. This is because the recording format is not the same.

  Accordingly, in the present invention, it is assumed that the recording format of the time-series information is different, including such a time difference between the divisions, and the “plurality of time-series data having different recording formats of the time-series information” is listed. It is defined as a state stored in the “time-series data storage unit” as a display target. In the case of “unit time format”, “time series information” is each time recorded at a fixed time interval, and in the case of “start / end format”, the above-described formats A, B, and C are used. , D, and a combination of the recording start time (time), the recording end time (time), and the recording time length. With this time-series information, the time sequence for each data constituting the time-series data The relationship (order) and the position on the time axis can be understood.

  In the “time-series information recording format”, “time-series information” is information that is not related to temporal elements, that is, the number of items or size (number of bytes) of individual data constituting the time-series data. And the data length indicated by the number of bits) are not considered. Therefore, even if time-series data having different numbers of items or the like is used, if the “time-series information recording format” is the same, it does not correspond to “a plurality of time-series data having different time-series information recording formats”. However, as a system configuration, it is only necessary to be able to display a list of “a plurality of time series data having different recording formats of time series information”. Even if the "recording format of information" happens to be the same (in this case, the number of items not related to the time factor may be the same or different), they are not displayed in a list or displayed in a list You can list them instead of not being able to do it. For example, when there are two pieces of time-series data in the unit time format recorded in the same minute unit, it is not a system configuration in which the list display is not performed or cannot be performed. For the two time series data of units, such a display or shaping process may be performed instead of a system configuration in which the shaping process for setting the unified unit time to 5 minutes is not performed or cannot be performed.

  In addition, “insufficient data” in “shaping means” is data that is necessary to display a list in a unified unit time, but means that the data is missing at the stage before processing by the shaping means. is there.

  <Operation and effect of basic configuration>

  In such a data display system of the present invention, when performing a list display of a plurality of time-series data, the shaping unit performs a shaping process using the unified unit time. Compared to the conventional list display in which each piece of data to be configured is simply arranged on the time axis, a list display that is easy to view can be performed.

  For this reason, inconvenience that may occur in the conventional list display can be solved. For example, when time-series data in unit time format of 1-minute units and 10-minute units is displayed as a list, in the conventional list display, there are a large number of blanks in the display portion of time-series data in unit time format of 10-minute units. However, in the present invention, since the deficient data is compensated by the shaping processing by the shaping means, such inconvenience can be solved and an easy-to-view list display can be performed. Therefore, the convenience for the user is improved.

  In addition, the time-series data that has been shaped by the shaping means is in a state in which data for list display is prepared for each time that is set by being divided by a unified unit time. Therefore, for each data (record) corresponding to each time set by dividing by a unit time unit, it is possible to retrieve each data constituting the formatted time series data. It can also be used. In the conventional list display, if there is no corresponding data, a blank list is displayed, so even if the data at that time is taken out, only the data for a part of the time series data of a plurality of time series data is displayed. As a result, the data can not be used effectively, whereas the present invention enables more advanced data utilization.

  Furthermore, since the shaping process using the unified unit time is performed, it is possible to display the list on the basis of the time (time series information) different from any recording format of the plurality of time series data to be displayed as a list, Various list displays are possible. Therefore, for example, the time-series data recorded at intervals of 10 minutes such as 8:00, 8:10, 8:20, 8:30, and the like described above, and 8:05, 8:15, 8:25, 8 : Even when displaying a list of time-series data whose recording times are completely shifted, such as time-series data recorded at intervals of 10 minutes such as 35,. It is possible to avoid the display of the state that is not performed and realize an effective list display. Specifically, if the unified unit time is set to 5 minutes, it is possible to display a list that makes full use of both time series data. In the case of this example, even if the unified unit time is set to 1 minute shorter than 5 minutes or the like, or conversely, the unified unit time is set to 30 minutes longer than 5 minutes, etc., the present invention provides an easy-to-read list display. Is possible.

  Since the shaping unit performs the selective shaping process, it is possible to select a representative data item with appropriate contents from a plurality of candidate data and perform a list display even if the unified unit time is increased. . For this reason, the shaping means not only compensates for the deficient data, but also can cope with a case where a plurality of candidate data are generated by increasing the unified unit time. Therefore, it is possible to freely set the length of the unified unit time, so it is possible to realize various list displays by adjusting the unified unit time, and in this respect as well, the user-friendliness of the system is improved. Is planned.

  Furthermore, since the shaping means is provided, it is possible to display a list of time series data in completely different recording formats such as a unit time format and a start / end format. Therefore, for example, even time series data acquired from different systems of data recording / collection systems can be easily combined. For this reason, by comparing the log data of a plurality of systems, it becomes possible to easily perform relevance analysis of those systems. More specifically, for example, by listing activity information for one day from logs such as “body movement” that is a life log and “PC operation information” that is an operation log, it is possible to analyze work styles and the like. (See FIG. 7 described later). In addition, since the data comparison of the old and new systems with different recording formats can be easily performed, it can be used for system maintenance and the like, thereby achieving the object.

  <Details about generation / shaping process by shaping means: (1) and (2) in FIG. 10 described later, and FIG. 9>

  Further, in the data display system described above, when performing the generation / shaping process, the unit / time format and the start / end format should be configured to perform the generation / shaping process according to each recording format as follows. Is preferred. Depending on the length of the specified unified unit time, both the unit time format and start / end format may require generation and shaping processing. It becomes possible to realize more reliably.

That is, in the data display system described above,
The shaping means is
For time-series data in unit time format, as the generation and shaping process, if the time interval for recording the time-series data in the unit time format is longer than the unified unit time, each time set by dividing by the unified unit time Among these, for data shortage time for which there is no corresponding data in the time series data in the unit time format, at least one of the two data recorded at the last two times sandwiching the data shortage time is used. Process to increase the number of data by generating insufficient data,
For the time series data in the start / end format, as the generation and shaping process, a plurality of times among the times set by dividing by the unified unit time are included in the data constituting the time series data in the start / end format. If it is included in the recording time zone of one data, the content of one data determined to be included is used and has the same content as this content, and is divided and set by a unified unit time. It is desirable that a process for generating data corresponding to a plurality of times among the times is executed.

  Here, by generating at least one of the two data recorded at the last two times sandwiching the data shortage time with respect to the “time series data in the unit time format”, The process of increasing the number of data may be a process of generating insufficient data by duplicating the content of the previous data in the time series order of the two data, and the subsequent data of the two data It may be a process to generate missing data by copying the contents of the data, or if there are multiple missing data, the contents of both the previous data and the subsequent data are duplicated to increase the number of data May be.

  Further, for example, in the case of continuous value items, deficient data may be generated by performing interpolation processing such as linear interpolation using the previous data and the subsequent data. Such interpolation processing, such as linear interpolation, is performed when, for example, the display target item in the same time series data is only one continuous value item, or individual data for a plurality of display target items in the same time series data. This is the case where it is not necessary to maintain the correlation within (in each record), or the case where the actually generated value need not be used for display.

  Furthermore, since “at least one of the two data is used”, data at a time earlier than the previous data and data at a later time than the subsequent data are also used. For example, interpolation processing by a quadratic or higher formula may be performed using three or more data, and the appearance frequency and the like of the preceding data and the subsequent data in the surrounding time zone are obtained and obtained. May be used to determine the distribution ratio when using the previous data, using the subsequent data, or using both data.

  Note that “time series data in the start / end format” is converted from time series data in the start / end format to time series data in the unit time format.

  <Details about the shaping target preparation process by the shaping target preparation means when the display time range extraction process is performed: see FIG. 8 to be described later>

In the data display system described above,
Using a plurality of time-series data stored in the time-series data storage means, comprising a shaping target preparation means for executing processing for preparing data to be subjected to shaping processing by the shaping means,
The display definition storage means
As the display definition information, in addition to the unified unit time, it is configured to store the start date and time and end date and time of the list display target for defining the temporal list target range,
The preparation target preparation means
For time series data in unit time format, among the data constituting the time series data in unit time format, the data recorded in the list target range from the start date to the end date is the data to be displayed as a list Execute the display time range extraction process to extract as
For the start / end format time-series data, it is determined whether each of the start date / time and the end date / time is included in the recording time zone of each data constituting the start / end format time-series data, Each data determined to be included is divided with the start date / time and end date / time as the boundary, and the data within the range to be listed from the start date / time to the end date / time is included in the data obtained by the division. It is desirable that the display time range extraction process for extracting data recorded in the list target range up to the end date and time as list display target data is executed.

  When the shaping target preparation means is provided and the display time range extraction process is executed as described above, the start date and time and the end date and time for determining the temporal listing target range are acquired from the display definition information. As a result, it is possible to display the list only for the necessary data, and since the data is limited to the data required for the list display before the shaping process by the shaping means, the shaping process by the shaping means can be performed smoothly. Can be done.

  <Details about the shaping target preparation process by the shaping target preparation means when the key item reference content range extraction process and the display time range extraction process are performed: see FIG. 8 to be described later>

Furthermore, in the data display system described above,
Using a plurality of time-series data stored in the time-series data storage means, comprising a shaping target preparation means for executing processing for preparing data to be subjected to shaping processing by the shaping means,
The display definition storage means
As display definition information, in addition to a unified unit time, at least one key item selected from items common to each data constituting a plurality of time series data in order to integrate a plurality of time series data, temporal Key item based content range consisting of the start date and time and end date and time of the list display target for determining the list target range, and at least one specified value specified for the key item to determine the specific list target range Designated information is also stored,
The preparation target preparation means
A key item reference content range extraction process for extracting data having key item reference content range designation information from each data constituting a plurality of time-series data is performed.
For time series data in unit time format, within the list target range from the start date to the end date among the data constituting the time series data in the unit time format extracted by the key item standard content range extraction process Execute the display time range extraction process to extract the recorded data as the list display target data,
For start / end format time-series data, the start date / time and end date / time of each of the data constituting the start / end format time-series data extracted by the key item reference content range extraction process Judge whether it is included in the recording time zone, divide the data judged to be included with the start date / time and end date / time as the boundary, and list the data from the start date / time to the end date / time among the data obtained by the division It is desirable that the display time range extraction process is executed to extract data recorded in the list target range from the start date and time to the end date and time, including the data in the range, as list display target data. .

  Here, “at least one key item selected from items common to each data constituting a plurality of time series data in order to integrate a plurality of time series data” means “at least one” There may be one or more “key items”, and in the case where there are a plurality of key items, they may be connected by AND or OR. For example, in the time-series data as shown in FIG. 6 to be described later, “employee code” and “activity state” are selected as key items, and at this time, the designation of employee code = A section manager employee code and activity state = conversation is designated. Then, when connected by AND, only the conversation data of section A is extracted, while when connected by OR, the conversation data of section A, data of activity other than the conversation of section A , Conversation data of employees other than section A is extracted.

  In addition, the “specified value” in the “content range specification information based on the key item consisting of at least one specified value specified for the key item in order to define the content listing target range” does not mean a numerical value, but mainly Possible values of discrete value items (for example, employee code that identifies each employee, personal work / movement / conversation to distinguish the activity state, clear / cloudy / rain / snow etc. to distinguish the weather) However, it may be a value (numerical value) that the continuous value item can take. Furthermore, since it is “at least one specified value”, there may be one or more “specified values” specified for one “key item”. For example, in the time-series data as shown in FIG. 6 to be described later, when the key item is an employee code, if the specified value specified for the key item is the employee code of the section A manager, it is one “specified value”. Only the data of section A is extracted. If the specified values specified for the key item are the employee code of the A section manager and the employee code of the B manager, they are two “specified values”. Extracted.

  In this way, when the shaping target preparation means is provided and the key item reference content range extraction process and the display time range extraction process are executed, the key is used to determine the specific list target range from the display definition information. By obtaining the key item-based content range specification information consisting of at least one specified value specified for the item, and the start date and time and end date and time for determining the temporal listing target range, it is limited to the necessary data Thus, the list display can be performed, and before the shaping process by the shaping unit is performed, the data is limited to data necessary for the list display. Therefore, the shaping process by the shaping unit can be performed smoothly.

  <Configuration for Executing Parallel Display Processing in Same Time Series Data by Output Unit: See FIG. 17 to be described later>

And, as described above, in the case where the shaping target preparation means is provided and the key item reference content range extraction process and the display time range extraction process are executed,
The output means is
When the display definition storage means stores, as the display definition information, key item reference content range specification information consisting of a plurality of specified values specified for the key item, each data including each of the plurality of specified values is stored. Identical to display in a direction orthogonal to the time series direction in a state corresponding to a list display of a plurality of time series data without displaying the data in the same time series as the data constituting the same time series data. It is good also as a structure which performs a parallel display process in time series data.

  Here, “each data including each of a plurality of specified values” means a state in which a plurality of specified values are specified for one key item, and therefore a plurality of specified values are included in one data. For example, when two specified values are specified, it means that there is data including one specified value of the two specified values and data including the other specified value. is there. For example, in the time-series data as shown in FIG. 6 to be described later, if the “key item” is an employee code and the two specified values are the employee code of the A section manager and the employee code of the B deputy manager, the employee code is This means that there is data that is A section manager and data that is employee B's deputy manager code.

  Further, the “time series direction” is a direction in which each time set by dividing by a unit time unit advances on a display screen or a printing surface for outputting a list display.

  In this way, when configured to perform parallel display processing in the same time series data by the output means, each data constituting the same time series data is treated as if each data constituting different time series data, A list can be displayed.

  For example, in the time series data as shown in FIG. 6 to be described later, the “key item” common to the time series data α, β is the employee code, and “a plurality of designated values designated for the key item” is the section manager A. In the case of the employee code and the B-director employee code, the data of the A-section manager and the B-order manager in the time-series direction of the A-section manager data and the B-order manager data constituting the time-series data α Instead of displaying them in a mixed state, they are displayed side by side in a direction orthogonal to the direction of the time series as if they were other time series data. Similarly, the section A data and the B order length data constituting the time series data β are not displayed in a state where the data of the section A data and the B order length data are mixed in the time series direction. Like other time-series data, they are displayed side by side in a direction orthogonal to the time-series direction. Thereby, for example, as shown in FIG. 17 described later, the A section length data of the time series data α, the B order length data of the time series data α, the A section length data of the time series data β, and the time series data β It is possible to realize a display in which the B order length data are arranged in a corresponding state, and each of the four is timed in a time series direction according to each time set by dividing by a unified unit time. It becomes.

  <Details of the selective shaping process for unit time format time series data by the shaping means: (3) in FIG. 10 to be described later, see the upper part of FIG. 11>

In the data display system described above,
The shaping means is
As the selection shaping process, when selecting representative data from a plurality of candidate data included in the unified unit time,
For time-series data in unit time format, if the time interval for recording time-series data in the unit time format is shorter than the unified unit time, obtain the number of occurrences of the same content for each item, and calculate the number of occurrences obtained. A configuration for performing processing for adding candidate data having the respective contents, calculating the total number by adding the number of appearances given for each candidate data, and selecting candidate data having the largest total number It is desirable that

  Here, the “item” in “determining the number of occurrences of the same content for each item” includes not only discrete value items but also continuous value items. In the case of a continuous value item, if the number of cases that a value belonging to the item can take (for example, if a continuous value that fluctuates in the range of 0 to 100% is rounded off or rounded off, 0, 1, 2 ,..., 99, 100, the number of cases is 101.) However, there is usually a tendency to increase compared to the case of discrete value items, and the same content value (that is, the same numerical value) , Hard to appear. However, depending on the rounding method of consecutive values (digit positions where rounding, rounding down, rounding up, etc. are performed), there is a high possibility that the same value will appear. If the possibility of the same content value appearing is extremely low, the number of occurrences of each content is only 1 for the item, and if there are other items, representative data is selected. It will not be a problem.

  In this way, when it is configured to execute the process of selecting candidate data with the largest total number of appearances as the selection shaping process, the total number of occurrences is calculated by adding the number of appearances assigned for each candidate data. Since the candidate data with the largest total number of calculated cases is selected as representative data, in addition to being able to select appropriate candidate data as representative data, the contents of data that actually exist (for each item that actually occurred) It is possible to display a list using a combination of values). For this reason, it is possible to maintain the correlation (the combination of the values of each item) of the values of each item in the individual data (in each record) constituting the time series data.

  More specifically, if there is a plurality of candidate data and the correlation between items in each candidate data can be ignored, the content with a large number of occurrences is obtained for each item, and the contents are collected. It is also conceivable to generate one data for list display. In this case, multiple candidate data are divided into items, and appropriate contents for each item are gathered. Therefore, each candidate data is used partially, and all of the multiple candidate data is used. Sometimes it becomes. However, since the data for list display is a collection, if there is a correlation between items, the correlation is lost, and therefore, the correlation analysis cannot be performed. On the other hand, in the present invention, any one suitable candidate data is selected from a plurality of candidate data, and this is used as representative data. Therefore, the candidate data selected as representative data is the existing data itself. Therefore, since the correlation (combination of values of each item) between items in the data is maintained, correlation analysis or the like can be performed. In addition, since each item in one time series data is usually provided from the same system, there is often a correlation between items unless processing or the like is performed in the time series data providing system. Ignoring this, and performing gathering, for example, contradictory states such as using both cooling and heating at the same time, calling on a mobile phone and calling on a fixed phone at the same time, etc. There is also the possibility of creating. Therefore, in the present invention, it is possible to perform the shaping process so that such a contradiction state does not occur as much as possible.

  <Details of selective shaping process for time series data in start / end format by shaping means (part 1): (4) in FIG. 10 described later, see lower part of FIG. 11>

Furthermore, in the data display system described above,
The shaping means is
As the selection shaping process, when selecting representative data from a plurality of candidate data included in the unified unit time,
For time series data in the start / end format, at least one of the two times of the unified unit time boundary is the recording time zone of the candidate data with the first or last time series in the plurality of candidate data. In this case, the first or last candidate data is divided at at least one of the two times, and the first or last candidate data within the unified unit time obtained by the division is obtained. It is desirable that a process for selecting candidate data having the longest time length among the plurality of candidate data included is executed.

  Here, “a plurality of candidate data including the first or last candidate data within the unified unit time obtained by the division” includes the first candidate data obtained by the division, the last candidate data obtained by the division, If there is candidate data that has not been divided, the first candidate data obtained by division, and candidate data that has not been divided, and the last candidate data obtained by division, and candidates that have not been divided Sometimes there is data. The same applies to the following inventions.

  As described above, when the process for selecting the candidate data having the longest time length is executed as the selection shaping process, it is possible to select appropriate candidate data as representative data, in addition to the data that actually exists. It is possible to display a list by using the contents (combination of the values of the items actually generated). For this reason, it is possible to maintain the correlation (the combination of the values of each item) of the values of each item in the individual data (in each record) constituting the time series data. The maintenance of this correlation is the same reason as described in detail in the case of the selective shaping process for the time series data in the unit time format described above.

  <Details of selective shaping process for time series data in start / end format by shaping means (part 2): see FIG. 12 to be described later, the situation is the same as (4) in FIG. 10 to be described later>

In the data display system described above,
The shaping means is
As the selection shaping process, when selecting representative data from a plurality of candidate data included in the unified unit time,
For time series data in the start / end format, at least one of the two times of the unified unit time boundary is the recording time zone of the candidate data with the first or last time series in the plurality of candidate data. When the intermediate position is, the first or last candidate data is divided at at least one of the two times, and the first or last candidate data within the unified unit time obtained by the division is included. If there is candidate data with the same contents for a plurality of candidate data, the time lengths thereof are summed up into one candidate data, and the candidate data with the longest time length is selected from the candidate data with the respective contents It is desirable to be configured to execute

  Here, if there is candidate data with the same contents, the time lengths are summed up into one candidate data, and the “candidate data with the same contents” in “candidate data with the same contents” is the same in candidate data units (record units) This means that the contents of a plurality of items (display target items other than the time-series information) included in the candidate data all match.

  As described above, when the process for selecting the candidate data having the longest time length is executed as the selection shaping process, it is possible to select appropriate candidate data as representative data, in addition to the data that actually exists. It is possible to display a list by using the contents (combination of the values of the items actually generated). For this reason, it is possible to maintain the correlation (the combination of the values of each item) of the values of each item in the individual data (in each record) constituting the time series data. The maintenance of this correlation is the same reason as described in detail in the case of the selective shaping process for the time series data in the unit time format described above.

  <Details of selective shaping process for time series data in start / end format by shaping means (modified example of the second): see FIG. 15 to be described later, the situation is the same as (4) in FIG. 10 to be described later>

Furthermore, in the data display system described above,
The shaping means is
As the selection shaping process, when selecting representative data from a plurality of candidate data included in the unified unit time,
For time series data in the start / end format, at least one of the two times of the unified unit time boundary is the recording time zone of the candidate data with the first or last time series in the plurality of candidate data. The first or last candidate data is divided at at least one of the two times, and the first or last candidate data within the unified unit time obtained by the division is obtained. For a plurality of candidate data, for each item, the time length of the same content is totaled to determine the appearance time of each content, the obtained appearance time is given to the candidate data having each content, and for each candidate data, The total time may be calculated by adding the given appearance times, and a process of selecting candidate data with the longest calculated total time may be executed.

  As described above, when the selection shaping process is configured to execute the process of selecting the candidate data with the longest total time, in addition to being able to select appropriate candidate data as representative data, It is possible to display a list by using the contents (combination of the values of the items actually generated). For this reason, it is possible to maintain the correlation (the combination of the values of each item) of the values of each item in the individual data (in each record) constituting the time series data. The maintenance of this correlation is the same reason as described in detail in the case of the selective shaping process for the time series data in the unit time format described above.

  <Details of selective shaping process for time series data in start / end format by shaping means (part 3 and part 4): see FIGS. 13 and 14 to be described later>

And in the data display system described above,
The shaping means is
As the selection shaping process, when representative data is selected from a plurality of candidate data included in the unified unit time, if a plurality of candidate data is selected in the first stage selection shaping process, the second As stage selective shaping process,
For discrete value items, by summing the time length of the same content for each item and dividing by the unified unit time, find the appearance rate of each content of a plurality of candidate data for the discrete value item,
For continuous value items, for each item, the probability density function indicating a predetermined or preselected distribution is used to determine the appearance rate of each content of the plurality of candidate data for the discrete value item,
For each of a plurality of candidate data, it is preferable that the appearance rate of each discrete value item and the appearance rate of each continuous value item are totaled, and the process of selecting the candidate data having a high total appearance rate is executed. .

  Here, the “distribution” in the “predetermined or preselected distribution” can be any type of distribution such as a normal distribution, a power distribution, etc., but “predetermined” is used as a system. This means that the distribution is set, and “pre-selected” means that the system has multiple types of distributions that can be used, and the user or system administrator selects the distribution to be used from among them. It means to do.

  In addition, “each discrete value item” or “each continuous value item” in “summing the appearance rate of each discrete value item and the appearance rate of each continuous value item” does not necessarily mean that there are a plurality of discrete value items, There may be one discrete value item. Similarly, it does not necessarily mean that there are a plurality of continuous value items, but there may be one continuous value item. In short, there is a situation where discrete value items and continuous value items are mixed. Show.

  When the first stage and the second stage of selective shaping processing are configured as described above, when the representative data is selected from a plurality of candidate data in the first stage of selective shaping processing, It is possible to cope with a situation where candidate data to be narrowed down to one. In addition, it is possible to cope with a situation where discrete value items and continuous value items coexist. Furthermore, since continuous value items use a distribution, it is possible to realize selective shaping processing according to the characteristics of the items. It becomes.

  <Invention of program>

  The program of the present invention is for causing a computer to function as the data display system described above.

  The above-mentioned program or a part thereof is, for example, a magneto-optical disk (MO), a compact disk (CD), a digital versatile disk (DVD), a flexible disk (FD), a magnetic tape, or a read-only memory (ROM). Recorded on recording media such as electrically erasable and rewritable read only memory (EEPROM), flash memory, random access memory (RAM), hard disk drive (HDD), solid state drive (SSD), flash disk, etc. Can be stored, distributed, etc., and transmitted using a transmission medium such as a wired network such as LAN, MAN, WAN, Internet, Intranet, Extranet, or a wireless communication network, or a combination thereof. You It is possible, also, it can be delivered by placing the carrier. Furthermore, the above program may be a part of another program, or may be recorded on a recording medium together with a separate program.

  As described above, according to the present invention, the shaping process using the unified unit time is executed for a plurality of time-series data, so that there is an effect that it is easy to see and various list displays can be performed.

1 is an overall configuration diagram of a data display system according to an embodiment of the present invention. The figure which shows the specific example of the input data definition information memorize | stored in the time series data storage means of the said embodiment. The figure which shows the specific example of the display item definition information memorize | stored in the display definition memory | storage means of the said embodiment. The figure which shows the specific example of the reference day definition information memorize | stored in the display definition memory | storage means of the said embodiment. The figure which shows the specific example of the representative data selection method definition information memorize | stored in the display definition memory | storage means of the said embodiment. The figure which shows the specific example of the time series data memorize | stored in the time series data storage means of the said embodiment. The figure which shows the specific example of the list display by the output means of the said embodiment. The detailed explanatory view of the shaping object preparation processing by the shaping object preparation means of the embodiment. The detailed explanatory view of the generation shaping processing by the shaping means of the embodiment. Explanatory drawing of the classification of the shaping process by the shaping means of the said embodiment. Explanatory drawing of the detail (the 1) of the selection shaping process by the shaping means of the said embodiment. Explanatory drawing of the detail (the 2) of the selection shaping process by the shaping means of the said embodiment. Explanatory drawing of the detail (the 3) of the selection shaping process by the shaping means of the said embodiment. Explanatory drawing of the detail (the 4) of the selection shaping process by the shaping means of the said embodiment. Explanatory drawing of the detail (the 2nd modification) of the selection shaping process by the shaping means of the said embodiment. The figure of the flowchart which shows the flow of the list display process by the data display system of the said embodiment. Explanatory drawing of the parallel display process in the same time series data by the output means of the said embodiment.

  An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 shows the overall configuration of a data display system 10 of the present embodiment. 2 shows a specific example of the input data definition information stored in the time series data storage means 31, and FIGS. 3 to 5 show various display definition information (displays) stored in the display definition storage means 32. FIG. Specific examples of item definition information, reference date definition information, and representative data selection method definition information are shown. FIG. 6 shows a specific example of time series data stored in the time series data storage means 31, and FIG. 7 shows a specific example of list display by the output means 26 using those time series data. It is shown. Further, FIG. 8 shows a detailed description of the shaping target preparation process by the shaping target preparation unit 24. 9 shows a detailed description of the generation and shaping process by the shaping means 25, FIG. 10 shows a comparison explanation about the type of the shaping process by the shaping means 25, and FIGS. A detailed description of the selective shaping process by the shaping means 25 is shown. FIG. 16 is a flowchart showing the flow of the list display process by the data display system 10, and FIG. 17 shows the description of the parallel display process in the same time series data by the output means 26.

  <Overall system configuration including data display system 10 and its peripheral system>

  In FIG. 1, a data display system 10 is a system that displays a list of a plurality of time-series data, and is constituted by one or a plurality of computers, and includes a time-series data acquisition means 21, an input data definition setting means 22, and the like. Display definition setting means 23, shaping target preparation means 24, shaping means 25, output means 26, time-series data storage means 31, display definition storage means 32, shaping target data storage means 33, and shaped The data storage means 34 is provided.

  Among these, each of the means 21 to 26 includes a central processing unit (CPU) provided in the computer main body constituting the data display system 10, one or more programs that define the operation procedure of the CPU, and It is realized by a working memory such as a main memory or a cache memory. Details of each means 21 to 26 will be described later.

  Each storage unit 31 to 34 is preferably realized by, for example, a hard disk drive (HDD), a solid state drive (SSD), or the like. Other recording media such as flash memory, MO, CD, DVD, FD, magnetic tape, etc. may be employed. The data storage format may be either a database or a file. Details of the storage units 31 to 34 will be described later.

  A plurality of time-series data providing systems 40 and 50 are connected to the data display system 10 via the network 1, and one or a plurality of user terminals 60 operated by a user and a system administrator operate the data display system 10. One or a plurality of administrator terminals 70 are connected.

  Here, the network 1 may be an external network such as the Internet, or an internal network such as an intranet or LAN (for example, a limited network such as an in-house, a group company, a factory, a business office, a university, etc.). Well, or a combination of them, including various forms, whether wired or wireless, or mixed wired and wireless, in short, multiple points (the length of the distance is Anything can be used as long as it can transmit information at a certain speed.

  The time-series data providing systems 40 and 50 record and collect various time-series data used for list display in the data display system 10 and store them, and store the collected time-series data to the data display system 10. It is a system to provide. These time-series data providing systems 40 and 50 may be existing systems, systems that will be developed and installed in the future, or systems having any configuration.

  For example, the time-series data providing system 40 includes a data collection device 41 and collected data storage means 42. The data collection device 41 is, for example, data recorded by a wearable terminal worn by a person, or data recorded by various portable devices such as a smartphone, a tablet terminal, a mobile PC (laptop personal computer), or the like. These are devices that are collected from individual wearable terminals and portable devices, and are transmitted to the collected data storage means 42 via the network 1 periodically or as needed. The collected data storage means 42 is, for example, a database server, and includes functions of a database management system (DBMS) although not shown. The time-series data providing system 40 is configured, for example, such that data transmission is performed from a large number of data collection devices 41 to one or a small number of collected data storage means 42. The data collection devices 41 are scattered in various places. It may be.

  The wearable terminal that records data collected by the data collection device 41 includes, for example, a wristband type worn on the wrist, a glasses type with a computer built in the frame portion of the glasses, and a belt or pocket like a pedometer. There are various types, such as a clip type that mounts with a pinch and measures the number of steps and the amount of activity.

  In the example of FIG. 6 and FIG. 7 to be described later, the data collection device 41 is an employee card worn by a person (employee) (it is a sensor card incorporating an activity amount sensor and is referred to as a name tag type sensor). It is a device that collects sensor data from. When each person is not wearing his / her employee card (for example, from the time of leaving the office until the next morning's work), if the employee card is inserted into the data collection device 41, the sensor data is transferred from the employee card. It is captured. Since the data collection device 41 can insert employee cards of a plurality of employees, it corresponds to a collection box for employee cards. More specifically, the employee card includes, for example, an acceleration sensor that makes it possible to determine whether a person is moving or sitting, and which employees are facing each other. An infrared sensor or the like for enabling discrimination is incorporated. In addition, an infrared beacon transmitter is installed inside each meeting room in the company, and the beacon signal is captured by the infrared sensor on the employee card to determine which meeting room the employee is in. Is possible.

  The time series data providing system 50 may have the same configuration as the time series data providing system 40 described above or a different configuration. This time-series data providing system 50 is described as a representative system for providing time-series data to the data display system 10 and therefore is not limited to one, and may be a plurality of units. The configuration may be the same or different.

  In the example of FIGS. 6 and 7 to be described later, the time-series data providing system 50 collects PC operation data that is a PC (personal computer) operation record by each person (each employee) and transmits the data via the network 1. Provided to the display system 10. For example, if an employee operates his / her PC and uses various applications such as Excel and Word (registered trademarks) while switching, the switching time (start time of use of each application) and operation time (each application) The time length used) is recorded as time-series information, and other information including information on which application was used in a state associated with the time-series information is recorded.

  The user terminal 60 and the administrator terminal 70 are configured by a computer, and include input means such as a mouse and a keyboard and display means such as a liquid crystal display. These terminals 60 and 70 may be not only PCs but also mobile devices such as smartphones and tablet terminals.

  <Detailed Configuration of Data Display System 10>

  <Configuration of time-series data acquisition unit 21 and type / content of time-series data>

  The time-series data acquisition means 21 acquires time-series data from the time-series data providing systems 40, 50, etc., and stores the acquired time-series data in association with time-series data identification information (input data identification information). The process memorize | stored in the means 31 is performed. The acquisition timing is arbitrary. For example, the acquisition timing may be periodically acquired at predetermined time intervals such as one week, one day, one hour, ten minutes, one minute, ten seconds, and the like. It may be acquired irregularly based on an instruction from the system administrator or the like, or may be acquired at a timing transmitted based on the convenience of the time series data providing systems 40 and 50. Therefore, the acquisition timing may be a transmission request at a timing managed by the data display system 10 side, or may be a transmission timing managed by the time-series data providing systems 40 and 50 side. In addition, you may acquire the time series data of another system | strain from each of several time series data provision system 40,50 grade | etc., Or several time series data (time series information from one time series data provision system). The recording format may be the same or different.

  The time series data is roughly classified into two types of time series data having different time series information recording formats, that is, time series data in unit time format and time series data in start / end format. In the present embodiment, the time series data α, α2, α3 in the unit time format is acquired mainly from the time series data providing system 40, and the time series data β, β2, in the start / end format is obtained from the time series data providing system 50. The description is given assuming that β3 is acquired.

  Time-series data in unit time format is a set of data (records) recorded at regular time intervals, such as 1 second, 1 minute, 5 minutes, etc., and is equally spaced on the time axis. This is a set of data recorded at a plurality of points (a plurality of times) dispersed in. The image of the point on the time axis is as described in (1) and (3) in FIG. However, there may be a time period in which data is missing due to an error or power-off, etc., and it is only necessary to record at regular intervals. Also, when the recording time interval is switched after a certain amount of time has passed, it is time-series data in unit time format. For example, when it is nighttime, the time interval is five times as long as the daytime. Since it is recorded at regular time intervals, it is time-series data in unit time format.

  As described above, the time-series data in the unit time format is a set of data (records) recorded at points on the time axis, but the recording format of the time-series information is like that of a point image. However, it is not necessarily limited to data at one point in time (one time) in a true sense that is instantaneously taken in from a data generation source (sensor or the like). For example, the average value, maximum value, minimum value, median value (median) of data having a length of time (for example, continuous value data of 1 second, 5 seconds, 1 minute, etc.) taken from a data generation source (sensor, etc.) ), Data obtained by calculating or extracting a representative value such as a quartile, and the recording time (time-series information) may be such a length (for example, 1 second, 5 seconds). Any one of the time of the start point, the time of the end point, and the time of the intermediate position between them (for example, the time of the middle or the generation time of the representative value such as the maximum value or the minimum value) But you can. Accordingly, the recording time as the time series information is one time point (one time), but the content of the data corresponding to the recording time itself is the processing content of the time series data providing system 40 that is the source of the time series data. It will be due to. Therefore, since the processing contents of the time series data provider are various, the processing contents of the time series data provider and the handling of the unit time format time series data in the data display system 10 need to be linked. Rather, it may be independent.

  The time series data in the start / end format is a set of data (records) having an arbitrary time length in which the content indicated by the data continues over the recorded time zone, not a point on the time axis, It is a set of data recorded in a time zone with a length. The image of the time zone on the time axis is as described in (2) and (4) in FIG. The individual data (individual records) constituting the time series data in the start / end format are handled as the same contents continuing for all the sections from the start of recording to the end of recording. In addition, the whole section here means the whole section in the recording time zone of each piece of data (record), and the whole section over the entire time-series data that is a set of a large number of data (records). It does not mean that.

  In addition, there are four types of recording formats of time-series information for start-end format time-series data, two types of recording start time, recording end time, and recording time length (operating time). It consists of a combination of the above information. That is, a format A for recording with a combination of two types of information of start time and time length (operation time), a format B for recording with a combination of two types of information of end time and time length (operation time), and the start time There is a format C for recording with a combination of two types of information, and an end time, and a format D for recording with a combination of three types of information, a start time, an end time, and a time length (operation time).

  Furthermore, as described above, the time series data in the start / end format is not a point on the time axis, but a set of time zone data on the time axis, but the content of each data (record) is the time. Treat as the same content continues across the belt. However, in the time-series data providing system 50, which is the time-series data provider, the data captured from the data generation source (sensor or the like) is not limited to the case where the same content continues. . As in the case of the time series data in the unit time format described above, the data taken from the data source (sensor, etc.) may fluctuate over the time zone indicated by the time series information, and the fluctuation data Representative values such as mean, maximum, minimum, median (median), and quartile are the data contents of individual data (individual records) that make up time-series data in the start / end format. It may be. Therefore, as in the case of the time series data in the unit time format described above, the true data content meaning the individual data (individual records) constituting the time series data is the time series that provides the time series data. This depends on the processing contents of the data providing system 50. The time length of each piece of data (individual record) is also determined by the processing of the time series data providing system 50. That is, it is the time-series data providing system 50 that creates individual data at continuous time intervals. For example, the content of data taken from a data generation source (sensor or the like) is a predetermined standard. When it is determined that the change has occurred, one piece of data (record) is created at that time, and creation of the next data in the time series is started. Therefore, since the processing contents of the time series data provider vary, the processing contents of the time series data provider and the handling of the time series data in the start / end format in the data display system 10 need to be linked. Rather, it may be independent.

  <Configuration of Input Data Definition Setting Unit 22 and Contents of Input Data Definition Information>

  The input data definition setting means 22 executes a process of accepting a setting input of input data definition information from the administrator terminal 70 by the system administrator. Specifically, a process of receiving the input data definition information transmitted from the administrator terminal 70 via the network 1 and storing the received input data definition information in the time-series data storage unit 31 is executed.

  The input data definition information is information indicating what time-series data is stored in the time-series data storage unit 31, as shown in FIG. Therefore, the user can select which item included in what time-series data is to be displayed as a list by referring to the input data definition information. Also, in the processing by the shaping target preparation unit 24 and the shaping unit 25, by reading this input data definition information, the recording format of the time series data and the type of each item included in the time series data (time series information, discrete value) It is possible to distinguish between items and continuous value items. The specific example of FIG. 2 corresponds to the specific examples of FIGS.

  In FIG. 2, the input data definition information includes input data identification information (time-series data identification information), types of items constituting the input data (time-series data) identified by the identification information, item names (item identification information) ), Information for associating the recording format and the like. For example, the input data α is time-series data in a unit time format in units of one minute, and there are items of date and time as time-series information, and as items other than time-series information, discrete value items include employees Information indicating that there is a code, an activity state, a place, employee 1 and employee 2 and no continuous value item is stored in the time-series data storage means 31. Further, the input data β is time series data in a start / end format of 1 second in the minimum time unit (minimum time unit when time and time are represented as digital data), and the time series information includes date, start time, There is an item of operating time (time length), information that the discrete value item has an employee code, application, and screen title, and the continuous value item has a CPU usage rate is stored in the time-series data storage means 31. Has been.

  <Configuration of Display Definition Setting Unit 23 and Contents of Display Definition Information>

  The display definition setting means 23 executes a process of receiving a setting input of display definition information (display item definition information, reference date definition information, representative data selection method definition information) from the user terminal 60 by the user. Specifically, the display definition setting means 23 reads the input data definition information stored in the time series data storage means 31 in response to a display request from the user, and displays the read input data definition information or the information on the screen. Data is transmitted to the user terminal 60 via the network 1 and display definition information (information created by the user with reference to the input data definition information) transmitted from the user terminal 60 via the network 1 is received. Then, a process of storing the received display definition information in the display definition storage unit 32 is executed. The display definition setting means 23 also transmits reference information (specific information for list display) for the user to create display definition information when transmitting input data definition information or screen display data of the information to the user terminal 60. As information for grasping the image, at least a part of the time series data stored in the time series data storage unit 31 may be transmitted together.

  As shown in FIG. 3, the display definition information includes which items of which time-series data are to be displayed as a list, which items are used as key items, and a plurality of time-series data are integrated. As shown, what kind of specified value is specified for the key item to narrow down the range to be listed (content range), what time of day to specify the time zone to narrow down the list target range (time range), As shown in FIG. 5, the information is defined as to which representative data selection method is used when performing the selective shaping process by the shaping means 25.

  In FIG. 3, the unit definition time is first stored in the display definition storage means 32 as display item definition information of the display definition information. This unified unit time is a unit time indicating a step size of a common time used when displaying a list, such as 1 second unit, 1 minute unit, 5 minute unit, and the like. This is commonly applied to all of the plurality of time series data. In the example of FIG. 3, the unified unit time is 1 minute, and this step size of 1 minute is applied to both the input data α and β. Therefore, in the shaping process by the shaping means 25, for both the input data α and β, data corresponding to each time set by dividing by the unified unit time is prepared as shaped data for list display. Become.

  Next, the display definition storage unit 32 stores key items for integrating input data (time-series data) as display item definition information. This key item is an item that exists in common for all of a plurality of time-series data to be displayed in a list, and in the example of FIG. 3, key item = employee code. The key item need not be set. This is because there may be no items that are common to all of the multiple time-series data, and even in such cases, it is possible to display a list, and it is common to all of the multiple time-series data. This is because even if there is an item to be displayed, it is not necessary to use the item as a key item, and a list display without overlapping items may be performed for a plurality of time-series data.

  Further, the display definition storage means 32 wants to display a list display target among the identification information of a plurality of input data (time series data) to be displayed as list display information and items included in the input data as display item definition information. The item name (item identification information) of the item is stored in association with each other. In the example of FIG. 3, for the input data α, the activity state, location, employee 1 and employee 2 are selected as display target items, and for the input data β, the application and screen title are selected as display target items. Yes. Further, since the employee code is designated as a key item for integration, this employee code is also selected as a display target item for both the input data α and β. When the employee code is not specified as a key item and the employee code is set as the display target item, the employee code is selected as the display target item for each of the input data α and β.

  In FIG. 4, the display definition storage means 32 stores, as reference date definition information in the display definition information, a specified value for a key item, a reference date, a start date and time, and an end date and time in association with each other. ing. Of these, the specified value for the key item is the key item-based content range specification information for narrowing the range to be listed in terms of content, and the reference date, start date and time, and end date and time are to be listed in terms of time. Information for narrowing down the range. This reference date definition information is commonly applied to all of a plurality of time-series data to be displayed as a list. A plurality of reference date definition information can be created.

  In the example of FIG. 4, the first reference date definition information is the employee code of the designated value = “A section manager” for the key item (employee code), the reference date = “8/21”, and the start date = “8”. / 21 7:23 ", end date and time =" 8/22 0:50 ", so in the time range from 7:23 on August 21 to 0:50 on August 22 A list of data is displayed. In this example, it is handled that it is included in the reference date (8/21) until 2:00 of the next day (8/22) of the reference date (8/21). In FIG. 4, if the fourth reference date definition information is set such that the specified value for the key item (employee code) = the employee code of “B Deputy Manager”, a list display based on the reference date definition information In this case, the list display of the data of the B manager is performed and the list of the data of the manager A is not displayed. Therefore, it does not mean that the list display of the data of the manager A and the data of the B manager is mixed. However, although details will be described later, a system configuration capable of displaying a list as shown in FIG. 17 can be adopted.

  In FIG. 5, the display definition storage means 32 associates input data identification information (time-series data identification information) with representative data selection method, continuous value items as representative data selection method definition information in the display definition information. Item name (item identification information) and usage distribution selection information are stored.

  The representative data selection method is a method of selecting representative data when executing the selection shaping process by the shaping means 25, and when selecting representative data from a plurality of candidate data, the first data (record) is selected. Information indicating whether the last data (record) is selected or the data (record) having the longest time length (operation time) is selected.

  The item name and use distribution selection information of the continuous value item is when the candidate data as representative data cannot be narrowed down to one in the first stage process when selecting representative data from a plurality of candidate data In addition, the second-stage processing is executed (see FIGS. 12 to 14), and at this time, information indicating what distribution is used for which continuous value item. For example, for a continuous value item such as a CPU usage rate, selection information such as using a normal distribution is used. The selection of the use distribution is a selection from distributions prepared in advance (for example, a normal distribution, a power distribution, etc.). For example, a normal distribution or the like may be prepared as a default value (default value). Note that the distribution to be used is not selected for each continuous value item, but the same distribution may be used for all continuous value items.

  In the example of FIG. 5, the selection setting of representative data selection method = “last record” is performed for the input data β, but the selection setting is performed for the item name of the continuous value item and the usage distribution selection information. Not. This is because when “first record” or “last record” is selected, there is no situation where candidate data as representative data cannot be narrowed down to one. In addition, the representative data selection method is not set for the input data α. The recording format of the input data α is a unit time format of 1 minute (see FIG. 2), and the unified unit time is 1 minute. Yes (see FIG. 3), both match, and therefore, the selection shaping process by the shaping unit 25 is not executed for the input data α. As shown by the dotted line in FIG. 5, for example, when the selection setting of representative data selection method = “longest time record” is performed, the item name of the continuous value item = “CPU usage rate”, the usage distribution selection Selection setting such as information = “normal distribution (default value)” is performed.

  <Configuration of the shaping target preparation unit 24>

  The shaping target preparation means 24 uses the plurality of time series data (input data α, β) stored in the time series data storage means 31 according to the reference date definition information stored in the display definition storage means 32. The shaping target data (input data α, β) to be shaped by 25 is generated, and the generated shaping target data is associated with the time-series data identification information (input data identification information) in the shaping target data storage unit 33. The process to memorize | store is performed.

  Specifically, as shown in FIG. 8, the shaping target preparation unit 24 selects a key item standard content range from each data (each record) constituting a plurality of time series data (input data α, β). A key item reference content range extraction process for extracting data (record) having specified information is executed. The key item reference content range specifying information is a specified value for the key item, and is included in the reference date definition information stored in the display definition storage means 32. In the example of FIG. 8, the employee code = “section A employee code” is included in the first and second reference date definition information. Therefore, the shaping target preparation unit 24 narrows down the list target range in terms of content by extracting data (record) whose employee code is the employee code of the section A manager.

  Further, as shown in FIG. 8, the shaping target preparation unit 24, for the time series data (input data α) in the unit time format, extracts the time in the unit time format extracted by the key item reference content range extraction process described above. Of the data (each section of A section manager) that make up the series data (input data α), the data (record of section manager A) recorded within the list target range from the start date to the end date is listed display target The display time range extraction process is performed to extract the data as the data. Therefore, the shaping target preparation unit 24 narrows down the list target range in time by extracting data (records) from the start date and time to the end date and time of the reference date.

  In the example of FIG. 8, for the first reference date definition information, reference date = “December 12”, start date = “2018/12/12 22:00”, end date = “2018/12/13 2 0:00 ”, the data (record of section manager A) within this list target range (time range) is extracted. Therefore, in the input data α that is time-series data in unit time format, the data whose time is “2018/12/12 21:59” (A section manager's record) is the list target range (time range). Because it is outside, it is not extracted. In this embodiment, 0: 0 to 1; 59 of the next day (December 13) of the reference date (December 12) is treated as the reference date (December 12), and the next day (December 13). Since the time at 2:00 (Sunday) will be handled on the following day (December 13), the data with the time "2018/12/13 2:00" Since it is outside the list target range (time range), it is not extracted.

  In the example of FIG. 8, for the second reference date definition information, reference date = “December 13”, start date = “2018/12/13 8:00”, end date = “2018/12 / 13 22:00 ", the data (record of section manager A) within this list target range (time range) is extracted. Therefore, in the input data α that is time-series data in the unit time format, the data with the time “2018/12/13 7:59” (record of section manager A) is included in the list target range (time range). Because it is outside, it is not extracted.

  Further, as shown in FIG. 8, the shaping target preparation unit 24 extracts the start date and time and the end date and time from the start / end format time series data (input data β) by the key item reference content range extraction process. In addition, it is determined which data is included in the recording time zone of each data (each record of section manager A) constituting the time series data (input data β) in the start / end format, and is determined to be included Data (section A's record) is divided with the start date and time and the end date and time as boundaries, and the data within the range to be listed from the start date and time to the end date and time (record of section A) is obtained. Including the display time range extraction process that extracts the data recorded in the list target range from the start date to the end date (A section manager's record) as the list display target data. To do. Therefore, the shaping target preparation unit 24 narrows down the list target range in time by extracting data (records) from the start date and time to the end date and time of the reference date.

  In the example of FIG. 8, for the first reference date definition information, reference date = “December 12”, start date = “2018/12/12 22:00”, end date = “2018/12/13 2 First, the start date / time = “2018/12/12 22:00” and the end date / time = “2018/12/13 2:00” are the respective data (A It is determined which data of the section manager's records is included in the recording time zone.

  In the input data β that is time series data in the start / end format, the start date = “2018/12/12 22:00”, the start time = “2018/12/12 21:00”, and the operation time (time length ) = “5:10” is included in the data (section A's record), this data is divided by the start date = “2018/12/12 22:00”, and the start time = “2018 / 12/12 21:56 ”, operation time (time length) =“ 3:40 ”data (section A record), start time =“ 2018/12/12 22:00 ”, operation time (time length) = “1:30” data (section A record) is generated. The total operation time (time length) of the two data (records) obtained by the division is 3: 40 + 1: 30 = 5: 10, which is the operation time (time length) before the division.

  Similarly, in the input data β that is time series data in the start / end format, the end date / time = “2018/12/13 2:00” is the start time = “2018/12/13 1:58”, and the operation time Since (time length) = “3:20” is included in the data (section A's record), this data is divided by the end date = “2018/12/13 2:00”, and the start time = “2018/12/13 1:58”, operation time (time length) = “1:50” data (section A record), start time = “2018/12/13 2:00”, operation time ( Time length) = “1:30” data (section A record) is generated. The total operation time (time length) of the two data (records) obtained by the division is 1: 50 + 1: 30 = 3: 20, so that the operation time (time length) before the division is obtained.

  Then, in the input data β, which is time series data in the start / end format, including the two data obtained by the division (A section manager's record), the start date / time = “2018/12/12 22:00” to the end date / time = Data up to “2018/12/13 2:00” (record of section manager A) is extracted. Therefore, in the input data β that is time series data in the start / end format, data of start time = “2018/12/12 21:56” and operation time (time length) = “3:40” (record of section manager A) Are not extracted because they are outside the list target range (time range). Further, data (record of section manager A) of start time = “2018/12/13 2:00” and operating time (time length) = “1:30” is also outside the list target range (time range). Is not extracted from.

  <Configuration of shaping means 25>

  The shaping unit 25 uses the unified unit time stored in the display definition storage unit 32 to configure each piece of data (input data α, β) of the shaping target time series data (input data α, β) stored in the shaping target data storage unit 33. A shaping process for shaping each record) is executed, and the shaped data obtained by this shaping process is stored in the shaped data storage means 34 in association with time-series data identification information (input data identification information). To do.

  The shaping process by the shaping means 25 includes a generation shaping process and a selective shaping process. As shown in FIG. 10, for each of these two types of shaping processes, time series data in unit time format and start There are different processes depending on the time-series data in the end format.

  The generation and shaping process will be described without distinguishing the unit time format and the start / end format. As shown in (1) and (2) in FIG. 10, the time-series data ( Of the data (records) corresponding to each time set by dividing each unit of data (each record) constituting the input data α, β) by the unified unit time stored in the display definition storage means 32 It is a process which produces | generates the data (record) to perform. The deficient data (record) means deficient data among the data necessary for performing the list display using the unified unit time.

  On the other hand, if the selective shaping process is described without distinguishing the unit time format and the start / end format, as shown in (3) and (4) in FIG. 10, time-series data (input data α , Β) when a plurality of pieces of data (each record) are included, these pieces of data are set as candidate data (candidate records), and a unified unit is selected from the plurality of pieces of candidate data. This is processing for selecting representative data (representative record) to be data corresponding to each time set by dividing by time.

  More specifically, the shaping unit 25 will be described as follows by dividing the unit time format and the start / end format. In the description, generalized items are used instead of the specific items described in FIG. 2 to FIG. 7, so the time-series data (input data α) in unit time format is denoted by the input data. Suppose that α2 and α3 are replaced, and the sign of the start-end format time series data (input data β) is replaced with input data β2 and β3.

  <Configuration of shaping means 25 / generation shaping process in unit time format: (1) in FIG. 10>

  First, the case (1) in FIG. 10 will be described. The case (1) is a generation and shaping process for time-series data (input data α2) in unit time format.

  As shown in (1) in FIG. 9 and FIG. 10, the shaping unit 25 performs time-series data (input data) in the unit time format as a generation shaping process for the time-series data (input data α2) in the unit time format. Corresponding to the time series data (input data α2) in the unit time format among the times set by dividing by the unified unit time when the recording time interval of the data α2) is longer than the unified unit time For the data shortage time when there is no data (record) to be used, at least one of the two data (records) recorded at the last two times sandwiching the data shortage time is used to determine the data shortage (record). A process of increasing the number of data (number of records) by generating is executed.

  In the example of FIG. 9, since the unified unit time = 30 seconds, in the input data α2 which is time-series data in the unit time format, the data (record) at the time “2017/12/12 8:00:30” and Since there is no data (record) at the time “2017/12/12 8:01:30”, these times are data shortage times, and the data (records) corresponding to these times are shortage data. is there.

  Therefore, for the data shortage time “2017/12/12 8:00:30”, the two times before and after this are “2017/12/12 8:00” and “2017/12/12 8: Therefore, the data (record) of “2017/12/12 8:00:30” is generated using the contents of the data (record) recorded at these times. Here, the content (item L = xxxx, item M = mmm) of the data (record) of “2017/12/12 8:00” corresponding to the earlier time of the two previous times sandwiching the data shortage time is shown. It is assumed that data (record) of “2017/12/12 8:00:30” is generated by copying.

  Similarly, for the data shortage time “2017/12/12 8:01:30”, the two times before and after this are “2017/12/12 8:01” and “2017/12/12 8”. : 02 ”, the data (record) of“ 2017/12/12 8:01:30 ”is generated using the contents of the data (record) recorded at these times. Here, the content (item L = yyyy, item M = nnn) of the data (record) of “2017/12/12 8:01” corresponding to the earlier time of the two previous times sandwiching the data shortage time is shown. It is assumed that data (record) of “2017/12/12 8:01:30” is generated by copying.

  Further, the shaping means 25, as the generation and shaping process of the case (1) in FIG. 10 described above, sets the time interval for recording the time series data (input data α2) in the unit time format as the unified unit time. In case of coincidence, that is, in a one-to-one correspondence relationship, all the times set by dividing by a unified unit time correspond to the time-series data (input data α2) in the unit time format. Since there is data (record) and there is no data shortage time, that is, there is no data shortage (record), so there is no need to increase the number of data by duplication or linear interpolation instead. The process of adopting the contents of each data (each record) constituting the time series data (input data α2) as the shaped data as it is is executed.

  <Configuration of shaping means 25 / generation shaping process in case of start / end format: (2) in FIG. 10>

  Next, the case (2) in FIG. 10 will be described. The case (2) is a generation and shaping process for time-series data (input data β2) in the start / end format.

  As shown in (2) in FIG. 9 and FIG. 10, the shaping means 25 sets the start / end format time-series data (input data β2) as a generation shaping process by dividing each unit time by unified unit time. It is included when a plurality of times of the times are included in the recording time zone of one of the data (each record) constituting the time series data (input data β2) in the start / end format. Using the content of one piece of data (one record) determined to be the same, this content has the same content, and corresponds to a plurality of times among the times set by dividing by a unified unit time Execute processing to generate data.

  In the example of FIG. 9, since the unified unit time = 30 seconds, in the input data β2, which is time series data in the start / end format, “2017/12/12 8:00:30” and “2017/12/12 8”. : 01: 00 ”is one data (record / date / start time =“ 2017/12/12 8:00:30 ”, operation time (time length) =“ 00:00 ” ) Is included in the recording time zone. This is because the recording time zone of this data is from 8:00:30 to 8:01:29 as indicated by (2) in FIG. Note that the time of 8:01:30 is not included in the recording time zone.

  Accordingly, the contents (item N = 333) of one data (one record) of date / start time = “2017/12/12 8:00:30” and operation time (time length) = “00:01:00”. ) Is used to generate data (records) corresponding to a total of two times, “2017/12/12 8:00:30” and “2017/12/12 8: 00: 10: 00”.

  In addition, as the generation / shaping process of the case (2) in FIG. 10 described above, the shaping unit 25 determines that one of the times set by dividing by the unified unit time is in the start / end format. When included in the recording time zone of one of the data (each record) constituting the time series data (input data β2), that is, in the case of a one-to-one correspondence, it was determined to be included Using the contents of one piece of data (one record), the data having the same contents as this one and corresponding to one time among the times set by dividing by the unit time unit is generated. Execute the process.

  <Configuration of shaping unit 25 / selective shaping process in unit time format: (3) in FIG. 10>

  Next, the case (3) in FIG. 10 will be described. The case (3) is a selective shaping process for time series data (input data α3) in unit time format.

  As shown in the upper part of FIG. 11 and (3) in FIG. 10, the shaping unit 25 selects representative data (representative record) from among a plurality of candidate data included in the unified unit time as the selection shaping process. In the case of unit time format time-series data (input data α3), when the recording time interval of the unit time format time-series data (input data α3) is shorter than the unified unit time, for each item Obtain the number of occurrences of the same content, assign the obtained number of occurrences to the candidate data (candidate records) having the respective contents, and add the number of occurrences assigned for each candidate data (for each candidate record) to obtain the total number of cases And a process of selecting candidate data (candidate record) having the largest calculated total number as representative data (representative record).

  In the example of FIG. 11, the unified unit time = 10 minutes, and a plurality of candidate data in the time series data (input data α3) in the unit time format from 8:00 to 8:10 (within the unified unit time). Representative data (representative record) is selected from (candidate records). Therefore, the data with the lowest date / time = “2017/12/12 8:11” is excluded from the candidate data (candidate record).

  For item P, the number of occurrences of xxx is one, yyyy is two, zzzz is one, and aaaa is one. Therefore, the number of occurrences is one for candidate data (candidate record) having xxx. Candidate data (candidate record) having the number of occurrences = 2, and candidate data having the zzzzz (candidate record) is given the number of occurrences = 1, and the candidate data having aaa The number of occurrences = 1 is assigned to (candidate record).

  For item Q, the number of occurrences of jjj is one, nnn is one, mmm is two, and kkk is one, so the number of occurrences is one for candidate data (candidate record) having jjj Candidate data (candidate record) with nnn is given 1 occurrence, and candidate data (candidate record) with mmm is given 2 occurrences and candidate has kkk The number of occurrences = 1 is assigned to the data (candidate record).

  For item R, since the number of occurrences of ppp is 3 and rrr is 2, the number of occurrences = 3 is assigned to candidate data (candidate record) having ppp, and candidate data having rrr ( Candidate record) is given 2 appearances.

  When the total number of occurrences is calculated by adding the number of appearances given to each candidate data (each candidate record), that is, in the horizontal direction, the candidate data (date / time = “2017/12/12 8:02”) ( Since the total number of candidate records) is 6, which is the largest, this candidate data (candidate record) is selected as representative data (representative record).

  At this time, if the total number is the same (when there are multiple candidate data arranged in the maximum total number), for example, select the candidate data closer to the top candidate data (candidate record) It is sufficient to make an arrangement such as selecting candidate data closer to the last candidate data (candidate record).

  In addition, the above is a process in the case where the setting is made such that candidate data (candidate record) having the longest operating time (time length) is selected as representative data (representative record). Therefore, if the setting is made such that the top candidate data (candidate record) is selected as representative data (representative record), the candidate data (candidate for date / time = “2017/12/12 8:00”) Record) is selected as representative data (representative record). If the last candidate data (candidate record) is set to be selected as representative data (representative record), the candidate data (candidate for date / time = “2017/12/12 8:06”) Record) is selected as representative data (representative record).

  <Configuration of shaping means 25 / selective shaping process in case of start / end format: (4) in FIG. 10>

  Next, the case (4) in FIG. 10 will be described. The case (4) is a selection shaping process for time-series data (input data β3) in the start / end format.

  As shown in the lower part of FIG. 11 and (4) in FIG. 10, the shaping unit 25 performs representative data (representative record) from among a plurality of candidate data (candidate records) included in the unified unit time as the selection shaping process. ) For the time series data (input data β3) in the start / end format, at least one of the two times at the boundary of the unified unit time is a plurality of candidate data (candidate records). When the order of the time series corresponds to the middle position of the recording time zone of the first or last candidate data (candidate record), the first or last candidate data (candidate record) is Of the multiple candidate data (candidate records) that are divided at at least one of the times and include the first or last candidate data within the unified unit time obtained by the division, There uptime candidate data (candidate record) of (time length), it executes a process of selecting as the representative data (representative records).

  In the example of FIG. 11, the previous time (8 o'clock) of the two times of the boundary of the unified unit time (8 o'clock to 8:10 o'clock) is the first candidate data (candidate record), that is, the date / start Recording time zone of candidate data (candidate record) with time = “2017/12/12 7:55:00” and operating time (time length) = “00:05:00” (7: 59: 00: 00) 8:03:59) (refer to (4) in FIG. 10), this candidate data (candidate record) is divided, and the data obtained by the division within the unified unit time Only data (record), that is, date / start time = “2017/12/12 8:00:00”, operation time (time length) = “00:04:00”, Candidate data (candidate record). The operating time (time length) is shortened from 5:00 to 4:00 due to the division.

  Similarly, the later time (8:10) of the two times at the boundary of the unified unit time (8:00 to 8:10) is the last candidate data (candidate record), that is, the date / start time. = “2017/12/12 8:08:30”, operation time (time length) = “00:04:00” candidate data (candidate record) recording time zone (8: 08: 30-8) : 12: 29) (see (4) in FIG. 10), this candidate data (candidate record) is divided, and the data within the unified unit time among the data obtained by the division (Record), that is, only data (record) with date / start time = “2017/12/12 8:08:30” and operating time (time length) = “00:01:30” are candidates Data (candidate record). The operating time (time length) is shortened from 4:00 to 1:30 due to the division.

  The candidate data with the longest operating time (time length) is selected from a plurality (three in this example) of candidate data (candidate records) including two candidate data (candidate records) obtained by these divisions. (Candidate record) is selected as representative data (representative record). Therefore, in this example, date / start time = “2017/12/12 8:04:00” and operation time (time length) = “00:04:00” in the middle candidate data (candidate record) However, since it is the longest operation time (time length), this is selected as representative data (representative record).

  At this time, when the operation time (time length) becomes the same value (when there are a plurality of candidate data arranged in the longest operation time (time length)), for example, the first candidate data (candidate record) It is only necessary to make an agreement such as selecting candidate data closer to, or selecting candidate data closer to the last candidate data (candidate record).

  In addition, the above is a process in the case where the setting is made such that candidate data (candidate record) having the longest operating time (time length) is selected as representative data (representative record). Therefore, if the setting is made such that the top candidate data (candidate record) is selected as representative data (representative record), date / start time = “2017/12/12 8:00: 00”, operation Candidate data (candidate record) with time (time length) = “00:04:00” is selected as representative data (representative record). If the last candidate data (candidate record) is set as representative data (representative record), date / start time = “2017/12/12 8:08:30”, operation Candidate data (candidate record) with time (time length) = “00:01:30” is selected as representative data (representative record).

  <Configuration of shaping means 25 / selective shaping process in case of start / end format: in the case of FIG. 15 which is a modification of the case (4) in FIG. 10>

  Further, the above-described selective shaping process shown in the lower part of FIG. 11 may be a selective shaping process as shown in FIG.

  In FIG. 15, the shaping unit 25 selects a representative data (representative record) from a plurality of candidate data (candidate records) included in a unified unit time as a selection shaping process, and starts and ends a time series. For data (input data β3), at least one of the two times at the boundary of the unified unit time is the first candidate data with the first or last time-series order among a plurality of candidate data (candidate records). When it corresponds to the intermediate position of the recording time zone (see (4) in FIG. 10), the first or last candidate data (candidate record) is set to at least one of the two times of the unified unit time boundary. For multiple candidate data (candidate records) including the first or last candidate data within the unified unit time obtained by dividing by time, set the same length of time for each item. The appearance time of each content is calculated, the obtained appearance time is given to the candidate data (candidate record) having the respective contents, and the given appearance time is added for each candidate data (each candidate record). Then, a process of calculating the total time and selecting candidate data (candidate record) having the longest calculated total time may be executed.

  In the example of FIG. 15, the process up to the division is the same as that in the lower part of FIG. Subsequent processing, that is, the time length of each content is totaled for each item to determine the appearance time of each content, and the obtained appearance time is assigned to candidate data (candidate record) having each content, For each data (each candidate record), add the given appearance time to calculate the total time, and select the candidate data (candidate record) with the longest calculated total time. is there.

  For item S, the appearance time of xxx is 4:00, yyyy is 4:30, and zzzz is 1:30. Therefore, for the candidate data (candidate record) having xxx, appearance time = 4: 00 Appearance time = 4: 30 is assigned to candidate data (candidate record) having yyyy, and appearance time = 1: 30 is assigned to candidate data (candidate record) having zzz.

  For item T, the appearance time of mmm is 4:00, and the appearance time of nnn is 4: 30 + 1: 30 = 6: 00. Therefore, for the candidate data (candidate record) having mmm, the appearance time = 4:00 is given, and appearance time = 6: 00 is given to the candidate data (candidate record) having nnn.

  For item U, the appearance time of ppp is 4: 00 + 4: 30 = 8: 30, and the appearance time of rrr is 1:30. Therefore, for the candidate data (candidate record) having ppp, the appearance time = 8:30 is assigned, and appearance time = 1: 30 is assigned to the candidate data (candidate record) having rrr.

  When the total time is calculated by adding the given appearance time for each candidate data (each candidate record), that is, in the horizontal direction, date / start time = “2017/12/12 8:04:00”, Since the total time of the candidate data (candidate record) in which the operation time (time length) = “00:04:30” is 19:00 is the longest, the candidate data (candidate record) is represented as representative data (candidate record). Select as a representative record.

  At this time, if the total time becomes the same value (when there are multiple candidate data arranged in the longest total time), for example, the candidate data closer to the first candidate data (candidate record) is selected. It is sufficient to make an arrangement such as selecting candidate data closer to the last candidate data (candidate record).

  <Configuration of shaping means 25 / selective shaping process in case of start / end format: case of FIG. 12 which is another modified example of the case of (4) in FIG. 10>

  Furthermore, instead of the selective shaping process shown in the lower part of FIG. 11, a selective shaping process as shown in FIG. 12 may be performed.

  In FIG. 12, the shaping unit 25 selects a representative data (representative record) from a plurality of candidate data (candidate records) included in the unified unit time as a selection shaping process, and starts and ends a time series. As for data (input data β), at least one of the two times of the boundary of the unified unit time is the time when the time series order of the plurality of candidate data (candidate records) is the first or last candidate data When it corresponds to the intermediate position of the recording time zone (see (4) in FIG. 10), the first or last candidate data (candidate record) is set to at least one of the two times of the unified unit time boundary. A plurality of candidate data (candidate records) including the first or last candidate data within the unified unit time obtained by dividing by time are candidate data (candidate records) having the same contents. If there is a data), the time lengths are summed up into one candidate data (candidate record), and among the candidate data (candidate records) of each content, the candidate data with the longest time length is represented as representative data. You may perform the process selected as (representative record).

  In the example of FIG. 12, the process up to the division is the same as in the case of the lower part of FIG. In this example, the unified unit time is 1 minute, and the representative data is selected within the unified unit time of 7:32 (1 minute from 7:32:00 to 7:32:59). Accordingly, the seventh candidate data (candidate record) with start time = “7:32:58” and operating time (time length) = “0:08” is divided, and two data obtained by the division Among the (records), data (records) within the unified unit time, that is, data (records) with start time = “7:32:58” and operating time (time length) = “0:02” ) Is the new seventh candidate data (candidate record).

  If there is candidate data (candidate record) having the same content for a plurality (seven in this example) of candidate data (candidate record) including the new seventh candidate data (candidate record) obtained by the division, These time lengths are summed up into one candidate data (candidate record). Here, candidate data (candidate record) having the same content means candidate data (candidate record) in which all items have the same content when there are a plurality of display target items.

  In the example of FIG. 12, the fifth and seventh candidate data (candidate records) have the same contents, that is, both of the application = “notepad”, the screen title = “date registration. Since the rate = “60”, these operation times (time lengths) are totaled to create candidate data (candidate record) f having an operation time total of 0: 11 + 0: 02 = 0: 13.

  After that, a plurality (six in this example) of candidate data (candidate records) a, b, c, including candidate data (candidate records) f created by summing up the operation time (time length) as described above. From d, e, and f, candidate data for the longest total operating time (total time length) is selected as representative data (representative record).

  However, in the example of FIG. 12, since the total operating time (total time length) for the two candidate data (candidate records) c and f is the same value at 0:13, the representative data (representative record) ) Cannot be narrowed down to one candidate data (candidate record), so the second stage processing (FIGS. 13 and 14) as shown below is executed. Even when the selective shaping process as shown in the lower part of FIG. 11 described above or the selective shaping process as shown in FIG. 15 is performed, the second-stage process as shown below (FIG. 13, FIG. 14) may be executed.

  <Configuration of shaping means 25 / second stage processing in selective shaping processing: FIGS. 13 and 14>

  In FIG. 13 and FIG. 14, the shaping unit 25 performs the first step when selecting representative data (representative record) from among a plurality of candidate data (candidate records) included in the unified unit time as the selection shaping process. When multiple candidate data (candidate records) are selected in the selection and shaping process, as a second stage of selection and shaping process, for discrete value items, the time length of the same contents is totaled for each item and unified. By dividing by the unit time, the appearance rate of each content of the plurality of candidate data (candidate records) for the discrete value item is obtained, and the continuous value item is predetermined or preselected for each item. An appearance rate of each content of the plurality of candidate data (candidate records) for the discrete value item is obtained by a probability density function indicating the distribution, and a plurality of candidate data (candidate records) is obtained. For each of (b), the process of selecting the candidate data (candidate record) having a high total appearance rate as the representative data (representative record) is performed by summing up the appearance rate of each discrete value item and the occurrence rate of each continuous value item. To do.

  In the example of FIG. 13, the application and the screen title are discrete value items, and the CPU usage rate is a continuous value item. Of these, the appearance rate calculation processing for the CPU usage rate, which is a continuous value item, will be described with reference to FIG.

  In the example of FIG. 13, the unified unit time = 1 minute, and representative data is selected within the unified unit time of 7:32 (1 minute from 7:32:00 to 7:32:59). Therefore, within this unified unit time (1 minute from 7: 32: 0 to 7:32:59), first, for each of the application and the screen title that are discrete value items, the time length of the same content is set for each item. The total operation time (length) is calculated by summing up. Then, by dividing the obtained total operation time (time length) by the unified unit time, a plurality (two in this example) of candidate data (each in this example) for each discrete value item (each of the application and the screen title) ( Candidate records) Appearance rates of the contents of c and f are obtained. In other words, for each discrete value item (each of the application and the screen title), the appearance rate of the content taken by the candidate data (candidate record) c and the appearance rate of the content taken by the candidate data (candidate record) f Ask for.

  For the application, the total operating time (time length) of Explorer is 0.07 + 0.06 = 0.13 (13 seconds), and the total operating time (time length) of Excel (registered trademark) is 0.00. 13 + 0.12 + 0.09 = 0.34 (34 seconds), and the total operation time (time length) of the memo pad is 0.13 (13 seconds). Note that the sum of these total operation times (time lengths) is 13 seconds + 34 seconds + 13 seconds = 1 minute, which is naturally the same as the unified unit time (1 minute), and six candidate data ( The sum of the operation times (time lengths) of the candidate records a, b, c, d, e, and f is naturally the same as the unified unit time (1 minute).

  Therefore, the content (value of the discrete value item) taken by the candidate data (candidate record) c for the application (discrete value item) is Excel, so that the total operation time (time length) of this Excel is 0. 0. If 34 (34 seconds) is divided by 1 minute (60 seconds), which is the unit time, the appearance rate of the content of Excel is 0.34 (34 seconds) ÷ 1 minute (60 seconds) = 57%.

  Similarly, since the content (value of the discrete value item) taken by the candidate data (candidate record) f for the application (discrete value item) is a memo pad, the total operation time (time length) of the memo pad is used. When 0.13 (13 seconds) is divided by 1 minute (60 seconds), which is the unit time, the appearance rate of the contents of the memo pad is 0.13 (13 seconds) ÷ 1 minute (60 seconds) = 22 %.

  As for the screen title, the total operating time (time length) of “10. Medium schedule” is 0.07 + 0.06 = 0.13 (13 seconds), and the total of “business execution.xls” The operation time (time length) is 0.13 + 0.12 = 0.25 (25 seconds), and the total operation time (time length) of “Consistency Check.xls” is 0.09 (9 seconds) Yes, the total operation time (time length) of “Registering date. Txt” is 0.13 (13 seconds). The total sum of these total operation times (time length) is naturally 1 minute.

  Therefore, since the content (value of the discrete value item) taken by the candidate data (candidate record) c for the screen title (discrete value item) is “business execution.xls”, the total of this “business execution.xls”. When 0.25 (25 seconds), which is the operation time (time length), is divided by 1 minute (60 seconds), which is the unified unit time, the appearance rate of the content “business execution.xls” is 0.25 ( 25 seconds) ÷ 1 minute (60 seconds) = 42%.

  Similarly, the content (value of the discrete value item) taken by the candidate data (candidate record) f for the screen title (discrete value item) is “for date registration. Txt”. When the total operating time (time length) of 0.13 (13 seconds) is divided by the unified unit time of 1 minute (60 seconds), the appearance rate of the content “Register date”. 0.13 (13 seconds) ÷ 1 minute (60 seconds) = 22%.

  In FIG. 14, for the CPU usage rate, which is a continuous value item, the appearance rate is calculated using the distribution. Since the occurrence probability for each value cannot be obtained for the continuous value item, a statistical model such as a normal distribution or a power distribution is selected, and it is assumed that the distribution of continuous values follows the statistical model. Here, it is assumed that a normal distribution (default value) is set as the distribution to be used. In this example, the unified unit time = 1 minute and the time unit (minimum time unit of the input data β in the start / end format) = 1 second, so these values, as well as six candidate data (candidate records) a , B, c, d, e, f, and the CPU usage rate value and the operating time (time length), based on the equations shown in FIG. Find the value needed for.

  In the example of FIG. 14, values such as the number of sections≈8, section width = 3, average μ≈61.4, and standard deviation σ≈6.8 are calculated. The number of sections is calculated based on the Sturges formula. Then, the probability density given by the probability density function of the normal distribution shown in FIG. 14 is integrated with the interval width around the CPU usage rate values of the candidate data (candidate records) c and f, and the appearance rate Is calculated.

  Accordingly, since the content (value of the continuous value item) taken by the candidate data (candidate record) c for the CPU usage rate (continuous value item) is 72, the probability density is set with the section width = 3 around 72. When integrated, the appearance rate of the content of CPU usage rate = 72 is calculated as 5%. Similarly, since the content (value of the continuous value item) taken by the candidate data (candidate record) f with respect to the CPU usage rate (continuous value item) is 60, the probability density with section width = 3 centering on 60 Is integrated, the appearance rate of the content of CPU usage rate = 60 is calculated as 17%.

  Further, for each of a plurality of candidate data (candidate records) c and f, the appearance rate of each discrete value item (application, screen title) calculated in FIG. 13 and each continuous value item calculated in FIG. 14 (this example) Then, the appearance rates of only the CPU usage rate are summed, and candidate data (candidate record) having a high total appearance rate is selected as representative data (representative record).

  Accordingly, the total appearance rate for the candidate record c is 57% of the application appearance rate + 42% appearance rate of the screen title + 5% CPU usage rate = 104%. Further, the total appearance rate for the candidate record f is 22% of the application appearance rate + 22% appearance rate of the screen title + 17% CPU usage rate = 61%. Therefore, since the value of the total appearance rate is larger for the candidate record c, the candidate record c is selected as the representative record.

  <When generation shaping processing by the shaping means 25 is not performed>

  An example of list display by the output means 26 is shown at the bottom of FIG. 9, and the state of the shaped input data β obtained by the processing by the shaping means 25 is shown immediately above. In this example, there is a blank for the item N of the input data β2 in the start / end format, but for the time “2017/12/12 8:01:30” among the times set by dividing by the unified unit time, This means that there is no corresponding data in the input data β2 in the start / end format. That is, there is no data in the input data β2 from 2017/12/12 from 8:01:30 to 8:02:00 (less than). Note that less than means that the exact time is not included. In addition, since the immediately preceding data is date / start time = “2017/12/12 8:00:30” and operation time (time length) = “00:01:00”, the list based on the unified unit time Although it seems to include the time “2017/12/12 8:01:30” for display, in this embodiment, the time just “8:01:30” is not included. .

  Therefore, the shaping unit 25 performs a generation shaping process that compensates for the deficient data for the time series data in the unit time format, but the generation shaping process for the time series data in the start / end format until the above-described case. Is not something to do. In addition, even for time series data in unit time format, data may be missing due to errors or power-off, etc. In such cases, generation and shaping processing to compensate for the missing data is performed. This is not intended, but it is insufficient when there is no corresponding data for each time set by dividing into unit time units under the condition that data can be recorded normally at regular time intervals. It is only necessary to perform generation / shaping processing to supplement the data to be processed. For example, in time-series data in unit time format, when data is recorded in units of 1 minute, data is lost for a longer time (for example, 5 minutes, 30 minutes, 1 hour, etc.) If there is a certain time zone, it is not intended to compensate for the lack of data for such a time zone, but in the situation where data is normally recorded in units of 1 minute, According to the unified unit time, for example, the lack of data is compensated at intervals of 5 seconds, 10 seconds, and the like. The time series data in either the unit time format or the start / end format is not left blank, but for example, data for displaying information such as “no data” is processed by the shaping unit 25. May be generated.

  <Configuration of output means 26>

  The output means 26 reads a plurality of time-series data after shaping processing (shaped input data α, β) stored in the shaped data storage means 34, and constitutes the read input data α, β. A list display process is performed in which data (each record) is displayed in a state in which the data (each record) is arranged in a state corresponding to each time set by dividing by a unified unit time and displayed in a comparable manner.

  In this embodiment, the output unit 26 creates screen display data (for example, Web screen data) for list display using a plurality of input data α and β read from the shaped data storage unit 34, The created screen display data is transmitted to the user terminal 70 or, if necessary, the administrator terminal 70 via the network 1.

  The list display process by the output unit 26 is not limited to the transmission process of screen display data (for example, Web screen data) via the network 1. For example, on the terminal side, a file that can be opened by an application such as Word or Excel (registered trademark) instead of a browser (for example, a Web browser) (when the file is opened, a list is displayed on the screen) Or a file that can be printed as a list display), and the created file is sent to the user via the network 1 in response to a download request from the user terminal 70 or the administrator terminal 70. You may transmit to the terminal 70 and the administrator terminal 70, You may transmit such a file by attaching to an email. Further, instead of transmission processing via the network 1, output to a display device directly connected to the data display system 10 or output of screen display to display means (not shown) included in the data display system 10 may be used. Further, it may be a print output to a printing apparatus connected to the data display system 10.

  Specifically, the output unit 26 performs an image list display process as shown in FIG. 7, for example. The list display of FIG. 7 is a comparison display in which unit time format input data α and start / end format input data β as shown in FIG. 6 are integrated (combined). The input data α in the unit time format is sensor data acquired from the time series data providing system 40, and the input data β in the start / end format is PC operation data acquired from the time series data providing system 50. Since the contents of these data have already been described in detail, detailed description is omitted here.

  Further, in the example of FIG. 7, since “integration” is performed in which the key item is the employee code and the specified value for the key item is the employee code of the section manager A, there is a display “section manager” as the title of the list display. However, there is no column for employee codes, and therefore, employee codes, which are key items, are not displayed in a list in a state corresponding to each time set by dividing by a unified unit time. This is because even if the employee code column is displayed, all of them become section A managers. For this reason, in the case of “integration” in which key items are designated, it is not a simple “combination” in which all the display target items are displayed in a list. If no key item is specified, it is a simple “join”. Even when a key item is designated, the key item field may be added to the list display.

  In the example of FIG. 7, the content of the item “activity state” (the value of the discrete value item) includes movement, personal work, conversation, and the like, and these are displayed in different colors (however, illustration of the color classification is omitted). ing). Further, the contents of the item “application” (values of the discrete value items) include mail, browser, office, screen saver, and the like, and these are also displayed in different colors (however, illustration of the color classification is omitted). The state shown in the figure is an example in which the key item is the employee code, and the specified value for the employee code (key item) is the employee code of the manager A. Only the data of the manager A is extracted and displayed in a list. is there. By this list display, for example, at 7:34 on August 21, 2017, Section A is talking to Deputy Director B and C in the meeting room 1, or on August 21, 2017. At 7:26, section A can use the Excel (registered trademark) to grasp information such as personal work related to the file “Resource Consistency Check.xls”.

  <Configuration of each storage unit 31-34>

  The time-series data storage unit 31 stores time-series data identification information (input data identification information) such as input data α, β (see FIG. 6) which are time-series data in various recording formats acquired by the time-series data acquisition unit 21. ) And the input data definition information (see FIG. 2) set by the input data definition setting means 22 is stored.

  Here, the time-series data identification information is the file name or folder name when the input data α, β, etc. are saved in a file. In the case of database identification information that is stored in the same database, it is, for example, a table name or the like. In short, it is only necessary to identify which time-series data the stored data is. The same applies to the following storage means 33 and 34.

  The display definition storage unit 32 stores the display definition information (display item definition information, reference date definition information, representative data selection method definition information: see FIGS. 3 to 5) set by the display definition setting unit 23. .

  The shaping target data storage means 33 stores the input data α, β, etc., which are time series data after processing by the shaping target preparation means 24, in association with time series data identification information (input data identification information).

  The shaped data storage means 34 stores the input data α, β, etc., which are time series data after the shaping processing by the shaping means 25, in association with time series data identification information (input data identification information).

  <Overall Flow of List Display Processing by Data Display System 10>

  In this embodiment, a list display process for a plurality of time-series data is performed by the data display system 10 as follows.

  In FIG. 16, the time series data acquisition means 21 obtains input data α, β (see FIG. 6) etc., which are time series data of various recording formats, from the time series data providing systems 40, 50, etc. via the network 1. Obtained and stored in the time-series data storage means 31 (step S1).

  Further, the system administrator operates the administrator terminal 70 to create input data definition information (see FIG. 2), and transmits the created input data definition information to the data display system 10 via the network 1. In the data display system 10, the input data definition setting means 22 receives the input data definition information transmitted from the administrator terminal 70 and stores it in the time series data storage means 31 (step S1). In the input data definition information setting process, when the system administrator knows in advance information on the internal structure of the input data α, β, etc., the input data α, β, etc. by the time series data acquisition means 21 It may be performed before the acquisition process.

  Next, the user operates the user terminal 60 to display display definition information (display item definition information, reference date definition information, representative data selection method definition information: see FIGS. 3 to 5) for performing a desired list display. The creation request is transmitted to the data display system 10 via the network 1. In the data display system 10, the input definition information (see FIG. 2) stored in the time series data storage unit 31 is read by the display definition setting unit 23 and transmitted to the user terminal 60 via the network 1. The user refers to the received input data definition information (see FIG. 2) and creates display definition information (see FIGS. 3 to 5). Then, the display definition setting unit 23 receives display definition information transmitted from the user terminal 60 via the network 1 by the user, and stores it in the display definition storage unit 32 (step S2).

  Subsequently, the display definition information (display item definition information, reference date definition information, representative data selection method definition information: see FIGS. 3 to 5) is read from the display definition storage means 32 by the shaping target preparation means 24, and this display definition is read. From the information, information of a plurality of time-series data that is a list display target (information about which time-series data is used) is acquired, and the input data α, β (see FIG. 6) and the like are read (step S3). If the display definition information received from the user terminal 60 in step 2 described above remains in the work memory such as the main memory, reading from the display definition storage unit 32 may not be performed. Alternatively, in this case, the display definition information storage area in the work memory such as the main memory may be considered as the display definition storage means 32.

  Then, the input data α, β read from the time-series data storage unit 31 by the shaping target preparation unit 24 based on the reference date definition information (content range and time range designation information: see FIG. 4) in the display definition information. (See FIG. 6) or the like is used to perform preparation processing (key item reference content range extraction processing and display time range extraction processing) of target data for shaping processing, and extract data obtained by this preparation processing is subjected to shaping processing. The data is stored in the data storage means 33 (step S3).

  Thereafter, the shaping unit 25 reads the target data of the shaping process from the shaping target data storage unit 33 and also displays the display definition information (display item definition information, reference date definition information, representative data selection method definition information: 3 to 5), and performs shaping processing (generation shaping processing, selection shaping processing) according to the unified unit time included in the display item definition information in the display definition information. Then, it is stored in the shaped data storage means 34 (step S4). As described above, when the display definition information remains in the work memory such as the main memory, it is not necessary to read.

  Subsequently, the output means 26 reads the shaped data for a plurality of time-series data (input data α, β, etc.) stored in the shaped data storage means 34, and outputs data for list display (for example, , Screen display data such as Web screen data) is created and transmitted to the user terminal 60 via the network 1 (step S5).

  <Effect of this embodiment>

  According to this embodiment, there are the following effects. That is, the data display system 10 performs the shaping process using the unified unit time by the shaping means 25 when performing the list display of the plurality of time series data. Compared to a conventional list display in which data is simply arranged on the time axis, a list display that is easy to view can be performed.

  For this reason, inconveniences that may occur in the conventional list display can be solved. For example, when time-series data in unit time format of 1-minute units and 10-minute units is displayed as a list, in the conventional list display, there are a large number of blanks in the display portion of time-series data in unit time format of 10-minute units. However, in the data display system 10, since the deficient data is compensated by the shaping process by the shaping means 25, such inconvenience can be solved and an easy-to-view list display can be realized. Therefore, the convenience for the user can be improved.

  In addition, the time-series data that has been shaped by the shaping unit 25 is in a state in which data for list display is prepared for each time set by dividing by a unit time unit. Therefore, for each data (record) corresponding to each time set by dividing by a unit time unit, it is possible to retrieve each data constituting the formatted time series data. Can also be used. In the conventional list display, if there is no corresponding data, a blank list is displayed, so even if the data at that time is taken out (even if only one record is extracted), it is one of a plurality of time-series data. However, the data display system 10 can realize more advanced data utilization, whereas the data display system 10 cannot effectively utilize the data. For example, time-series data α4 in unit time format in units of 1 minute is recorded at 8:20, 8:21, 8:22, and so on. Time-series data α5 in unit time format in units of 5 minutes is If the time series data α6 recorded at 8:20, 8:25,... And recorded in a unit time format of 10 minutes is recorded at 8:20, 8:30,... In the display, when only the data of 8:22 is taken out, there is only data α4, but in the data display system 10, there is data in any of α4, α5 and α6.

  Furthermore, since the shaping process using the unified unit time is performed, it is possible to display the list on the basis of the time (time series information) different from any recording format of the plurality of time series data to be displayed as a list, Various list displays can be realized. For this reason, for example, time-series data recorded at intervals of 10 minutes such as 8:00, 8:10, 8:20, 8:30,..., 8:05, 8:15, 8:25, 8:35 Even when displaying a list of time-series data whose recording times are completely deviated, such as time-series data recorded at 10 minute intervals such as...,. It is possible to avoid the display of the state (a state where data is displayed in a zigzag manner), and it is possible to realize an effective list display. Specifically, if the unified unit time is set to 5 minutes, it is possible to display a list that makes full use of both time series data. In the case of this example, even if the unified unit time is 1 minute shorter than 5 minutes or the like, or conversely, the unified unit time is 30 minutes longer than 5 minutes, etc. Display can be realized.

  Since the shaping unit 25 performs the selective shaping process (see (3) and (4) in FIG. 10), even if the unified unit time is increased, an appropriate content is selected from a plurality of candidate data. List data can be displayed by selecting representative data. For this reason, the shaping means 25 not only compensates for the deficient data, but can also cope with a case where a plurality of candidate data are generated by increasing the unified unit time. Accordingly, the unit time can be freely set. Therefore, by adjusting the unit time, various list displays can be realized, and the user-friendliness of the system can be improved. .

  Furthermore, since the shaping means 25 is provided, it is possible to display a list of time series data in completely different recording formats such as a unit time format and a start / end format. Therefore, for example, it is obtained from data recording / collection systems of different systems (for example, the case of the time series data providing system 40 of the input data α and the time series data providing system 50 of the input data β shown in FIG. 7). Even time series data can be easily combined. For this reason, by comparing the log data of a plurality of systems, it is possible to easily perform relevance analysis of these systems. More specifically, as shown in the example of FIG. 7, by listing the daily activity information from logs such as “body movement” that is a life log and “PC operation information” that is an operation log, Can analyze work styles. In addition, data comparison between the old and new systems whose recording format has been changed can be easily performed, and can be used for system maintenance.

  Further, since the data display system 10 includes the shaping target preparation unit 24, by obtaining the start date and time and the end date and time (see FIG. 4) for determining the temporal listing target range from the display definition information. The list display can be performed only on necessary data, and before the shaping process by the shaping unit 25, the data is limited to the data necessary for the list display. Therefore, the shaping process by the shaping unit 25 is efficiently performed. be able to.

  Further, since the data display system 10 includes the shaping target preparation unit 24, a specified value (for example, a key item) specified for a key item in order to determine a content listing target range from the display definition information. By obtaining key item-based content range designation information (see FIG. 4) consisting of the employee code of section A designated for the employee code), the list display limited to necessary data and the efficiency by the shaping means 25 Can be realized.

  Further, the shaping unit 25 calculates the total number by adding the given number of appearances for each candidate data, as shown in the upper part of FIG. 11, as the selective shaping process for the time series data α3 in the unit time format. Since the candidate data with the largest calculated total number is selected as representative data, appropriate candidate data can be selected as representative data, and the contents of data that actually exists (actually occurred A list display can be performed using a combination of values of each item. For this reason, the correlation (the combination of the values of each item) of the values taken by each item can be maintained in the individual data (in each record) constituting the time series data α3. Analysis can be performed. Since the maintenance of the correlation between items has already been described in detail, detailed description thereof is omitted here.

  Further, the shaping means 25 performs the process of selecting the candidate data having the longest time length as shown in the lower part of FIG. 11 as the selection shaping process for the time series data β3 in the start / end format. Even in this case, in addition to being able to select appropriate candidate data as representative data, it is possible to display a list using the actual data contents (combination of the values of each item that actually occurred). . For this reason, as in the case of the unit time format, the correlation (the combination of the values of each item) of the values of each item is maintained in the individual data (in each record) constituting the time series data β3. Therefore, correlation analysis or the like can be performed.

  When the shaping unit 25 performs the selective shaping process for the time-series data in the start / end format, the process shown in FIGS. 12 and 15 may be performed. However, even when these processes are performed, the process shown in FIG. Similar to the processing shown in the lower part, correlation between items (combination of values of each item) can be maintained, so that correlation analysis or the like can be performed.

  Further, since the shaping means 25 is configured to perform the first and second stage selective shaping processes (see FIGS. 13 and 14), candidate data to be representative data in the first stage selective shaping process. It is possible to appropriately cope with the situation where it is not possible to narrow down to one. Further, it is possible to deal with a situation where discrete value items and continuous value items are mixed. Furthermore, since the distribution is used for continuous value items, it is possible to realize selective shaping processing according to the characteristics of the items.

  <Deformation form>

  Note that the present invention is not limited to the above-described embodiment, and modifications and the like within a scope where the object of the present invention can be achieved are included in the present invention.

  For example, the data display system 10 of the above embodiment is configured to acquire a plurality of time-series data and output a list display via the network 1, but may be a stand-alone configuration.

  Further, as shown in FIG. 4, the data display system 10 according to the embodiment is configured to display a list by setting a reference date and its start date / time and end date / time. The list display may be continued as long as the display request is continued without setting the range. In this case, for example, the time series data acquisition means 21 continuously acquires a plurality of time series data from the time series data providing systems 40, 50, etc. via the network 1, and the shaping means 25 performs real time or abbreviation. A shaping process can be performed in real time, and the output unit 26 can output a list display in real time or substantially real time.

  Further, the data display system 10 of the above embodiment is configured to include the shaping target preparation unit 24 and the shaping unit 25, and once the shaping target data prepared by the shaping target preparation unit 24 is processed, the shaping target data storage unit 33, and the shaping unit 25 reads the shaping target data from the shaping target data storage unit 33 and performs the shaping process. However, the functions of the shaping target preparation unit 24 and the shaping unit 25 are integrated and shaped. The target data storage unit 33 may not be provided. In the case of such a configuration, the shaping target data storage unit 33 may be regarded as a working memory such as a main memory.

  Further, the output means 26 may be configured to execute parallel display processing within the same time series data as shown in FIG. In order to perform the parallel display processing within the same time series data, the display definition storage means 32 stores, as display definition information, key item-based content range designation information including a plurality of designated values designated for key items. To be. In the case of performing the list display as in the example of FIG. 7 of the embodiment, one specified value (for example, the employee code of the section manager) may be designated for the key item (employee code), but the example of FIG. Then, a plurality of (for example, two) designated values are designated. In parallel display processing within the same time series data, each data including each of a plurality of specified values is displayed side by side in the time series direction (for example, the vertical direction of the screen) as each data constituting the same time series data. Without executing the processing, a process of arranging and displaying in a direction orthogonal to the direction of the time series (for example, the horizontal direction of the screen) is executed in a state corresponding to the list display of the plurality of time series data.

  In the example of FIG. 17, the employee code of the section manager A and the employee code of the subordinate manager B are specified as a plurality of specified values for the key item (employee code). Thereby, the A section length data of the time series data α, the B order length data of the time series data α, the A section length data of the time series data β, and the B order length data of the time series data β correspond to each other. The four are arranged in a state, and each of the four is displayed in a time series direction (vertical direction in FIG. 17) according to each time set by dividing by a unit time. In addition, as shown in FIG. 17, you may enable it to switch the order of the horizontal direction about each display object item.

  When the output means 26 is configured to perform parallel display processing within the same time-series data in this way, each data constituting the same time-series data (in the example of FIG. 17, data of section A and B-order length) Data) can be handled and displayed as a list as if they were data constituting different time-series data. For this reason, various list displays can be realized and user satisfaction can be enhanced.

  As described above, the data display system and program of the present invention are suitable for use when, for example, displaying a list of a plurality of time-series data having different recording formats provided from a separate data recording / collecting system. ing.

DESCRIPTION OF SYMBOLS 10 Data display system 25 Formatting means 26 Output means 31 Time series data storage means 32 Display definition storage means

Claims (11)

A data display system configured by a computer that executes processing for displaying a list of a plurality of time-series data,
As a plurality of time-series data having different recording formats of time-series information to be displayed in a list, time-series data in a unit time format consisting of a set of data recorded at regular time intervals, or in a recorded time zone Time-series data storage means for storing time-series data in at least one recording format among time-series data in a start-end format composed of a set of data having an arbitrary time length that is handled so that the content indicated by the data continues When,
Display definition storage means for storing display definition information including unified unit time for displaying a list of the contents of the plurality of time series data stored in the time series data storage means at a unified time interval;
Generating / shaping process for generating deficient data among data corresponding to each time set by dividing by the unified unit time stored in the display definition storage unit using each data constituting the time series data And when a plurality of pieces of data constituting the time series data are included within the unified unit time, the plurality of pieces of data are set as candidate data, and a plurality of pieces of the candidate data are A shaping means for executing a selection shaping process for selecting representative data for data corresponding to each time set by dividing by the unified unit time,
Output means for executing a list display process for displaying each data constituting the time series data after the shaping process by the shaping means in a state corresponding to each time set by dividing by the unified unit time. A data display system characterized by that. The shaping means is
For the time-series data in the unit time format, as the generation and shaping process, when the time interval of recording the time-series data in the unit time format is longer than the unified unit time, set by dividing by the unified unit time Of the time series data in the unit time format for which there is no corresponding data, at least one of the two data recorded at the last two times sandwiching the data shortage time Using the data of the above, executing the process of increasing the number of data by generating the insufficient data,
For the time series data in the start / end format, as the generation / shaping process, a plurality of times among the times set by dividing by the unified unit time constitute each data constituting the time series data in the start / end format Is included in the recording time zone of one of the data, the content of the data determined to be included is used, and the content is the same as this content, and is divided by the unified unit time. 2. The data display system according to claim 1, wherein a process for generating data corresponding to the plurality of times among the set times is executed. Using a plurality of the time-series data stored in the time-series data storage means, comprising a shaping target preparation means for executing processing for preparing data to be subjected to shaping processing by the shaping means,
The display definition storage means includes
As the display definition information, in addition to the unified unit time, it is configured to store a start date / time and an end date / time of a list display target for defining a temporal list target range,
The shaping target preparation means is
As for the time series data in the unit time format, among the data constituting the time series data in the unit time format, the data recorded in the list target range from the start date to the end date is displayed as a list. Execute the display time range extraction process to extract as target data,
For the time series data in the start / end format, the start date / time and the end date / time are included in the recording time zone of each of the data constituting the time series data in the start / end format. Each of the data determined to be included is divided with the start date and time and the end date and time as a boundary, and the data within the list target range from the start date to the end date of the data obtained by the division is determined. It is configured to execute a display time range extraction process that extracts data recorded within a list target range from the start date and time to the end date and time, including data, as list display target data. The data display system according to claim 1 or 2. Using a plurality of the time-series data stored in the time-series data storage means, comprising shaping target preparation means for executing processing for preparing data to be subjected to shaping processing by the shaping means,
The display definition storage means includes
As the display definition information, in addition to the unified unit time, at least one key item selected from items common to each data constituting the plurality of time series data in order to integrate the plurality of time series data A key including a start date and time and an end date and time of a list display target for defining a temporal listing target range, and at least one designated value designated for the key item to define a content listing target range It is also configured to store item range content range specification information,
The shaping target preparation means is
While performing a key item reference content range extraction process for extracting data having the content range designation information of the key item reference from each data constituting a plurality of the time series data,
As for the time series data in the unit time format, among the data constituting the time series data in the unit time format extracted by the key item reference content range extraction process, listing from the start date to the end date Execute the display time range extraction process to extract the data recorded in the target range as the list display target data,
For the time series data in the start / end format, each of the start date / time and the end date / time is included in the data constituting the start / end format time series data extracted by the key item reference content range extraction process. It is determined which data is included in the recording time zone, the data determined to be included is divided with the start date and time and the end date and time as a boundary, and the start date and time among the data obtained by the division is determined. Display time range extraction processing for extracting data recorded in the list target range from the start date and time to the end date and time, including data in the list target range until the end date and time, as list display target data The data display system according to claim 1, wherein the data display system is configured to be executed. The output means includes
When the display definition storage means stores, as the display definition information, the key item reference content range designation information composed of a plurality of designated values designated for the key item, each of the plurality of designated values is stored. Are orthogonal to the direction of the time series in a state corresponding to the list display of the plurality of time series data without being displayed side by side in the direction of the time series as each data constituting the same time series data. 5. The data display system according to claim 4, wherein parallel display processing within the same time-series data that is displayed side by side in a direction is executed. The shaping means is
As the selection shaping process, when selecting the representative data from the plurality of candidate data included in the unified unit time,
For the time series data in the unit time format, when the time interval of recording the time series data in the unit time format is shorter than the unified unit time, the number of occurrences of the same content is obtained for each item, and the found appearance The number of cases is given to the candidate data having the respective contents, and the number of appearances given is added for each candidate data to calculate the total number of cases, and the candidate data having the largest calculated total number is selected. It is set as the structure which performs a process. The data display system in any one of Claims 1-5 characterized by the above-mentioned. The shaping means is
As the selection shaping process, when selecting the representative data from the plurality of candidate data included in the unified unit time,
For the time series data in the start / end format, at least one of the two times at the boundary of the unified unit time is the candidate data whose time series order is the first or last among the plurality of candidate data. The first or last candidate data is divided at at least one of the two times, and the first unit time in the unified unit time obtained by the division is obtained. Or it is set as the structure which performs the process which selects the said candidate data of the longest time length among the said some candidate data containing the said candidate data. The one in any one of Claims 1-6 characterized by the above-mentioned. The data display system described. The shaping means is
As the selection shaping process, when selecting the representative data from the plurality of candidate data included in the unified unit time,
For the time series data in the start / end format, at least one of the two times at the boundary of the unified unit time is the candidate data whose time series order is the first or last among the plurality of candidate data. The first or last candidate data is divided at at least one of the two times, and the first unit time in the unified unit time obtained by the division is obtained. Or, if there is the candidate data having the same content for a plurality of candidate data including the last candidate data, the time lengths thereof are summed up into one candidate data, and among the candidate data of each content The data display system according to any one of claims 1 to 6, wherein a process of selecting the candidate data having the longest time length is executed. Mu. The shaping means is
As the selection shaping process, when selecting the representative data from the plurality of candidate data included in the unified unit time,
For the time series data in the start / end format, at least one of the two times at the boundary of the unified unit time is the candidate data whose time series order is the first or last among the plurality of candidate data. The first or last candidate data is divided at at least one of the two times, and the first unit time in the unified unit time obtained by the division is obtained. Alternatively, for a plurality of candidate data including the last candidate data, the time lengths of the same contents are totaled for each item to obtain the appearance time of each content, and the obtained appearance time is the candidate data having the respective contents. For each candidate data, calculate the total time by adding the given appearance times, and select the candidate data with the longest calculated total time. Data display system according to claim 1, characterized in that it is configured to. The shaping means is
When the representative data is selected from the plurality of candidate data included in the unified unit time as the selective shaping process, when the plurality of candidate data are selected in the first-stage selective shaping process In the second stage selective shaping process,
For discrete value items, by summing the time length of the same content for each item and dividing by the unified unit time, obtain the appearance rate of each content of the plurality of candidate data for the discrete value item,
For continuous value items, for each item, a probability density function indicating a predetermined or preselected distribution is used to determine the appearance rate of each of the plurality of candidate data for the discrete value item,
For each of the plurality of candidate data, the appearance rate of each discrete value item and the appearance rate of each continuous value item are summed, and the process of selecting the candidate data having a high total appearance rate is executed. The data display system according to claim 7, wherein the data display system is a data display system.   The program for functioning a computer as a data display system in any one of Claims 1-10.

Images