WO2019239964A1

WO2019239964A1 - Information processing device, control method therefor, program, storage medium, and system

Info

Publication number: WO2019239964A1
Application number: PCT/JP2019/022188
Authority: WO
Inventors: 雄一日塔
Original assignee: キヤノン株式会社
Priority date: 2018-06-15
Filing date: 2019-06-04
Publication date: 2019-12-19
Also published as: JP2019219758A; JP7129235B2

Abstract

The present invention: acquires time-series data consisting of a series of data pieces acquired per a first unit-time; divides the time-series data into partial time-series data per a second unit-time, which is longer than the first unit-time; identifies, as a retrieval subject, one of the data pieces acquired per the first unit-time, from the time-series data; determines a retrieval range for retrieving the retrieval subject in the time-series data, on the basis of the temporal relative position of the identified data piece in the partial time-series data of the second unit-time, and also on the basis of the relative positions of the data pieces acquired per the first unit-time in the partial time-series data of another second unit-time; and retrieves a data group similar to the determined data group, from data pieces included in the determined retrieval range.

Description

Information processing apparatus, control method thereof, program, storage medium, and system

The present invention relates to a technique for managing time series data.

In recent years, data management systems that can store and retrieve time-series data have been proposed, and users can refer to information obtained when a similar phenomenon occurred in the past when a certain phenomenon occurs. can do.

In Patent Document 1, given from a certain period in the past from the present, based on a plurality of time series data obtained in real time such as fluctuations in stock prices of multiple brands and sensor values obtained from a plurality of sensors. A technique for searching for the most similar partial time-series data pairs is proposed.

Japanese Patent No. 5711171

In the fields of crop cultivation and livestock production, a certain phenomenon is often repeated at a predetermined cycle such as every year or every few years, and a record over a plurality of cycles becomes a huge amount of data. On the other hand, since growth delays and weather fluctuations are natural phenomena, the same phenomenon does not always occur at the same timing in each cycle. In Japanese Patent Laid-Open No. 2004-260260, a time lag that occurs between time series data of a plurality of series is taken into account, but it is possible to efficiently search for data relating to a phenomenon that is repeated approximately at a predetermined cycle in the same series. That was not taken into account.

An object of the present invention is to make it possible to efficiently search data related to a phenomenon repeated approximately at a predetermined cycle from accumulated time series data.

In order to solve the above-described problems, an information processing apparatus according to the present invention includes an acquisition unit that acquires time-series data in which data acquired every first unit time continues, and the time-series data is converted into the first unit. A management unit that divides the data into partial time-series data for each second unit time longer than the time and holds the data in a holding unit; and any one of the data acquired for each first unit time among the time-series data (A) a reference position in the partial time-series data for each second unit time of the data acquired for each first unit time specified as the search target; And (b) each of the data acquired for each first unit time included in the time series data in the partial time series data for each second unit time. From the reference position A determination means for determining a search range for searching for data similar to the search target in the time series data based on the relative position, and the data determined in the determined search range Search means for searching for data similar to the data.

According to the present invention, it is possible to efficiently search data relating to a phenomenon that is repeated at a roughly predetermined cycle from the accumulated time-series data.

The figure showing an example of composition of an information processing system, and composition of hardware The figure showing an example of composition of an information processing system, and composition of hardware The figure showing an example of composition of an information processing system, and composition of hardware The figure showing an example of functional composition of an information processing system Sequence diagram showing an example of processing of each component of information processing system Table representing an example of data processed by information processing system Table representing an example of data processed by information processing system Table representing an example of data processed by information processing system Sequence diagram showing an example of processing of each component of information processing system The figure showing an example of the user interface displayed in a client apparatus A flowchart showing an example of processing executed in the server device Table representing an example of data processed by information processing system Table representing an example of data processed by information processing system Table representing an example of data processed by information processing system Table representing an example of data processed by information processing system A flowchart showing an example of processing executed in the server device The figure showing an example of the user interface displayed in a client apparatus The figure showing an example of functional composition of an information processing system Sequence diagram showing an example of processing of each component of information processing system The figure showing an example of the user interface displayed in a client apparatus The figure showing an example of the user interface displayed in a client apparatus Table representing an example of data processed by information processing system Table representing an example of data processed by information processing system Table representing an example of data processed by information processing system

Hereinafter, embodiments according to the present invention will be described in detail with reference to the drawings. Note that the configurations described in the following embodiments are representative examples, and the scope of the present invention is not necessarily limited to these specific configurations.

<First Embodiment>
In this embodiment, an information processing system that collects and manages time-series data related to cultivation of agricultural products is shown as an example. In particular, an example of a system capable of searching a data group obtained in at least one day designated as a target (search query), or a similar data group, from past time-series data accumulated regarding cultivation of agricultural products Will be explained. In the cultivation of agricultural crops, the weather conditions, crop growth conditions, cultivation work records, etc. are digitized and managed. The phenomenon recorded in the cultivation of crops and the record of the work performed in accordance with the phenomenon are characterized by being repeated at roughly fixed intervals. For example, in response to the temperature rising in early spring every year, the soil is prepared and the appearance of buds is confirmed. Every summer, the number of leaves that grow and the temperature are observed, and the number of leaves that should be reduced for ventilation is broken. However, similar data is not always available on the same day of the month. For example, the temperature may rise early in spring, or there may be a lot of rain before summer and the leaves may be delayed. Therefore, when a certain phenomenon is confirmed, checking the data obtained on the same day just one year ago is not always helpful, so conventionally required information is obtained from a huge amount of time-series data. It took time and effort to find it. Therefore, in the first embodiment, partial time series data similar to data obtained in at least one day when time series data is specified as a search query is efficiently identified from the accumulated time series data. And present it to the user.

<System configuration>
FIG. 1A is a diagram illustrating a configuration example of an information processing system according to the first embodiment. In FIG. 1, the information processing system includes a server device 10 and a client device 20. The server apparatus 10 and the client apparatus 20 are connected so as to be capable of bidirectional communication by a known communication technique such as a communication network such as the Internet or a cellular phone network, a wireless LAN (Local Area Network), or short-range wireless communication. . The server device 10 is an information processing device that functions as a cloud on a network, for example. The server apparatus 10 accumulates time series data input via the client apparatus 20. Furthermore, a user's instruction input via the client device 20 is acquired, and desired partial time-series data is retrieved from the accumulated time-series data and output. The client device 20 is, for example, a tablet PC that can be carried and operated by a user. In the present embodiment, the client device 20 receives time series data related to cultivation of crops input by the user and an instruction for searching for data related to past cultivation conditions, transmits the received data to the server device 10, and outputs from the server device 10. The search results are displayed.

<Hardware configuration>
FIG. 1B is a hardware configuration of the server apparatus 10. The server device 10 has a role as a management device that manages data related to cultivation of agricultural products in the system of the present embodiment. The CPU 101 is a central processing unit, performs calculations for various processes, logical determinations, and the like, and controls each component connected to the system bus 108. The server device 10 is equipped with a memory including a program memory and a data memory. A RAM (Random Access Memory) 102 is a data memory, and includes a work area for the program of the CPU 101, a data save area during error processing, a load area for the control program, and the like. A ROM (Read-Only Memory) 103 is a program memory, and stores a program for control by the CPU including various processing procedures to be described later. The program memory may be realized by loading a program into the RAM 102 from an external storage device or the like connected to the server device 10. The communication interface (I / F) 104 performs two-way communication by wire or wireless with other information processing devices, communication devices, external storage devices, and the like by a known communication technique. In the present embodiment, data can be transmitted to and received from the client device via the Internet. The communication with other devices includes indirect communication through a relay device in addition to direct communication.

The display device 105 is, for example, a CRT display, a liquid crystal display, or the like, and is a device that outputs an image on a display screen. The display device 105 may be an external device connected to the server device 10 by wire or wirelessly. The input device 106 is a device for capturing user operation information input to the server device 10 via an input device such as a touch panel, a keyboard, or a mouse. However, in the case of the server device 10 of the present embodiment, the display device 105 and the input device 106 may be omitted by causing the client device to perform functions related to display and input. The HDD 107 is a hard disk for storing a plurality of electronic data and programs according to the present embodiment. An external storage device may be used as the same role. Here, the external storage device can be realized by, for example, a medium (recording medium) and an external storage drive for realizing access to the medium. As such media, for example, a flexible disk (FD), a CD-ROM, a DVD, a USB memory, an MO, a flash memory, and the like are known. The external storage device may be a server device connected via a network.

FIG. 1C shows a hardware configuration of an information processing apparatus that functions as the client apparatus 20. The client device 20 has a role as a terminal device that inputs or displays data related to cultivation of crops in the system of the present embodiment. Similar to the server device, the CPU 111 uses the RAM 112 as a work memory, executes an OS and a program stored in the ROM 113 and the HDD 117, performs various processing operations and logical determinations, and configures each component connected to the system bus 118. Control. The communication interface (I / F) 114 performs wired or wireless bi-directional communication with other information processing apparatuses, communication devices, external storage devices, and the like by a known communication technique. In the present embodiment, data can be transmitted to and received from the server device via the Internet. The communication with other devices includes indirect communication through a relay device in addition to direct communication.

The display device 115 is a device that performs display output under the control of the CPU 111. The input device 116 is a device for capturing user operation information input to the client device 20 under the control of the CPU 111. In the present embodiment, the client apparatus 20 includes a touch panel display as the display device 115 and the input device 116. When handling image information as time-series data, the input device 116 may be provided with an imaging device or an input interface for capturing image information from an external camera device. Further, various sensor devices may be provided, and detection information thereof may be input as time series data. The HDD 117 is a hard disk for storing a plurality of electronic data and programs according to the present embodiment. However, it may be connectable to an external storage device via a network or various interfaces.

<Functional configuration>
FIG. 2 is an example of a block diagram illustrating a software configuration of the entire information processing system according to the present embodiment. First, a functional configuration of an information processing apparatus that functions as the server apparatus 10 is represented. The server device 10 includes an acquisition unit 201, a management unit 202, a specification unit 204, a determination unit 205, a communication unit 206, and a search unit 207. Each of these functional units is realized by the CPU 101 developing a program stored in the ROM 103 in the RAM 102 and executing processing described later. In addition, for example, when configuring hardware as an alternative to software processing using the CPU 101, it is only necessary to configure arithmetic units and circuits corresponding to the processing of each functional unit described here. The server device 10 further includes a holding unit 203 that is a functional unit of the HDD 107 or the RAM 102. Hereinafter, each element will be described.

The acquisition unit 201 acquires time-series data input in the client device 20 via the communication unit 206 described later. In the present embodiment, the time-series data includes information representing at least one of temperature, weather, humidity, growth state, image, and record of work performed by the operator, which is collected every day regarding the cultivation of the crop, in the first unit. A group of data collected every hour. However, the time series data is acquired not only directly from the client device 20, but also, for example, by reading out data stored in an external storage device or acquiring information distributed from an external service. Good. Note that the first unit time is a predetermined unit such as one day or one week.

The management unit 202 divides the acquired time series data into at least the partial time series data of the second unit time longer than the first unit time and holds the divided time series data in the holding unit. In this embodiment, each data group acquired during the first unit time is handled as partial time series data by associating an identifier indicating the partial time series data of the second unit time to which it belongs. Furthermore, in this embodiment, an identifier indicating a relative position in the partial time-series data of the second unit time is associated with each data group acquired in the first unit time. The second unit time longer than at least the first unit time is a predetermined period such as one month, half year, one year, or three years, and is equivalent to one cycle in which the contents represented by the time series data are roughly repeated. It means length. For example, in the cultivation of a single crop (when a single crop is cultivated once a year in a certain field), the crop is cultivated in accordance with the four seasons, the harvest is reached, and the soil is rested. The work for saving is repeated. In this case, the time series data obtained in one year is processed as partial time series data of the second unit time. Hereinafter, in the present embodiment, a predetermined period in which time series data repeats roughly is referred to as “season”. Each data block obtained by dividing the time series data into the first unit time is referred to as “classification”. The management unit 202 assigns to each section of the time-series data an identifier that uniquely indicates the season to which the section belongs and an identifier that indicates the temporal relative position of the section in the season.

The holding unit 203 can identify the relative position in the partial time-series data of the second unit time of each data group included in the time-series data and acquired in the first unit time. As described above, the time series data is retained. In the case of the present embodiment, the identifier assigned by the management unit 202 is held in association with each data group constituting the time series data. Further, search history information of search results by the search unit 207 is held.

The identifying unit 204 identifies any of the categories included in the held time-series data received by the client device 20 as the target category. The target category is a category having a parameter used as a search query in the search process of the search unit 207 described later. The user of the client device 20 operates the client device 20 and designates the target category in order to search the data series similar to the data group included in the target category from the accumulated time series data.

Based on the relative time position of the category identified as the target in the season to which the category belongs and the relative position of the category belonging to the other season, The search range to be searched is determined. In the present embodiment, the search range is narrowed down based on the identifier of each section assigned to the management unit 202. The communication unit 206 transmits / receives data to / from the communication unit 209 of the client device 20 described later via the NETIF 104 and the network. In the case of the present embodiment, at least time-series data is received and the result of the search process is transmitted. The retrieval unit 207 retrieves a data group similar to the data group included in the section identified as the target from the time series data included in the determined search range. In the present embodiment, as a search result by the search unit 207, a section including a data group similar to the specified section is explicitly shown, and further, the contents of the data group included in the section can be browsed. The search result is provided to the client device 20.

Next, the functional configuration of the information processing apparatus that functions as the client apparatus 20 is represented. The server device 10 includes an input unit 208, a communication unit 209, a reception unit 210, and a display control unit 211. Each of these functional units is realized by the CPU 111 developing a program stored in the ROM 113 in the RAM 112 and executing processing to be described later. In addition, for example, when configuring hardware as an alternative to software processing using the CPU 111, it is only necessary to configure arithmetic units and circuits corresponding to the processing of each functional unit described here. Hereinafter, each element will be described.

The input unit 208 inputs information input via the input device 116 to the system as time series data. For example, time-series data in which text information input from a keyboard displayed on a touch panel display or image information acquired from an imaging device is associated with time information acquired from the information is obtained. The time information may be obtained using a clock provided in the client device 20 or manually input by the user.

The communication unit 209 transmits / receives data to / from the communication unit 206 of the server apparatus 10 via the NETIF 110 and the network. In the present embodiment, at least time-series data is transmitted and the search processing result is received. The accepting unit 210 accepts an input specifying a target category in order to request the server device 10 to execute a process of searching for a desired data block from time-series data. In the case of the present embodiment, an operation for the user interface displayed on the display device 115 is accepted. The display control unit 211 generates and outputs an image to be displayed on the display device 115. In the present embodiment, an image representing a user interface screen for designating a target category and an image for presenting a search result received from the server device 10 to the user are displayed.

Note that the functional configuration shown in FIG. 2 is the minimum configuration necessary for realizing the present embodiment, and each device may further have a functional unit. For example, the server device may include a display control unit for performing display output on the display device 105.

<Process flow>
In the present embodiment, the flow of processing executed by each element of the information processing system will be described with reference to the sequence diagram of FIG. In the following, the present embodiment will be described by taking as an example a case where time series data obtained by crop cultivation that repeats the one-year period from April to March of the following year is processed with the first unit time as one day. .

Hereinafter, each process (step) will be described with S added to the head of the reference numerals. In step S 301, the input unit 208 of the client device 20 inputs time-series data to the system in response to a user operating a touch panel or an imaging device or connecting an external device. In S 302, the input time series data is transmitted to the communication unit 206 of the server device 10 by the communication unit 209. In step S 303, the acquisition unit 201 of the server apparatus 10 acquires time series data received by the communication unit 206. The input data 401 in FIG. 4A illustrates a part of the acquired time series data. The input data 401 has values for the attributes of the data date and time 402, the data type 403, and the data details 404. As the value of the data detail 404, a value according to a unit or dimension corresponding to the data type 403 is stored.

In S304, the management unit 202 generates partial time series data obtained by dividing the acquired time series data for each second unit time. In this embodiment, the data group which has the same identifier produces | generates as partial time series data by matching the identifier which shows the season which belongs to each data contained in time series data uniquely. FIG. 4B is a season definition table 405 that the server apparatus 10 of the present embodiment holds in advance as reference information. Here, one year from April to March of the following year is defined as one season, and an identifier 406 that uniquely represents the season is associated with the start date 407 and end date 408 of the season. For example, the identifier representing the season from April 2017 to March 2018 is a numerical value “6”. In S304, the management unit 202 identifies the season to which each data belongs by referring to the data date and time 402 and the season start date and time 407 and the end date and time 408 of each data, and the identifier 406 representing the identified season is determined. Associate a value with the data.

In S305, the management unit 202 assigns an identifier indicating a relative position in the partial time series data to each data. In the present embodiment, based on the date / time information of each data, an index indicating a temporal relative position within the season s (s is any value of the identifier 406) of the data is specified. In the present embodiment, the following formula 1 is used to calculate an index x representing the relative position of each data within the season s. Here, Ti is the data date and time 402 associated with each data, Tb is the start date and time of the season s to which each data belongs, and Td is a predetermined first unit time. For example, the time unit Td = 1 day.

In this embodiment, the index x determined in this way means what day each data is obtained after the start of each season, which is a one-year period. Represents a relative position.

In S306, the management unit 202 holds time series data in which each data is associated with an identifier indicating a season and a relative position in the holding unit 203. FIG. 4C is an example of time-series data held in the holding unit 203, and the season 409 assigned by the management unit 202 and the value of the index 410 are associated with the input data 401.

FIG. 5 is a sequence diagram showing an example of processing executed by each component of the system according to the first embodiment at the time of searching for a category similar to the target category from the time series data. In step S 501, the reception unit 210 of the client device 20 receives an input for designating a target category that is a query for processing for searching for similar categories. In this embodiment, the user operates the user interface screen displayed on the display device 115 to specify the category to be a query.

FIG. 6 is an example of a user interface screen in the present embodiment. In FIG. 6, at least a part of the sections included in the season selected by the season selection button 601 is displayed in the section display area 602. In the present embodiment, since the first unit time for dividing the time series data into the divisions is one day, the date corresponding to each division is displayed side by side in the division display area 602 so that it can be easily identified. The user can designate one classification corresponding to one data block in the time-series data by selecting one of the displayed dates. Further, in the present embodiment, in response to selection of any date, the selected date is displayed so as to be identifiable as shown in an area 603 in FIG. 6, and a date of one day or more can be selected. A range adjustment slider 604 is displayed. The displayed range 605 is designated by operating the displayed slider 604. In the form shown in FIG. 6, the range for the previous three days is designated from 2017/4/5 to 2017/4/3 with 2017/4/6 as a reference.

Here, the influence range 605 will be described. For example, in the cultivation of agricultural products, there is a case where a phenomenon that “a leaf has changed color” occurs and it is desired to refer to a record when such a phenomenon has occurred in the past. Even if the phenomenon of “discolored leaves” was recorded for the first time on a certain day, the phenomenon occurred in connection with the phenomenon of “rainy” and “low temperature” a few days ago. there is a possibility. Therefore, in the present embodiment, the range adjustment slider 604 is used to instruct a search for a similar data group based on the contents of the data group in an arbitrary period before that in addition to a certain target day. Is possible. If such a search is performed, even if there are various possible causes of the phenomenon that “the leaves have changed color”, for example, it is influenced by the case where the temperature is low and the temperature is high. It is possible to provide search results that distinguish between cases.

Returning to the description of the sequence in FIG. 5, in S 502, the communication unit 209 transmits information related to the specified target category to the communication unit 206 of the server device 10. In step S503, the determination unit 205 determines a search range based on a comparison of relative positions in the partial time series data. A specific example of the processing in S503 will be described later with reference to the flowchart of FIG. Due to the processing of S503, partial time-series data in which the temporal relative position in each season is close to the relative position of the target section from the stored time-series data becomes the search range.

In S504, the search unit 207 searches for a data group similar to the target from the determined search range. A specific example of processing in S504 will be described later with reference to the flowchart of FIG. Through the process of S504, one or more sections of the stored time-series data are specified as search results. In this embodiment, one division is specified from one season, and it is not specified in the similar division from a season in which no division similar to the target division exists.

In S505, the search unit 207 updates the search history information held in the holding unit 203. The search history information is information in which the season and index of the category specified as the target category are associated with the season and index of the category specified as the search result, and is used in the processing of S503 described in detail later. The In S506, the search unit 207 outputs the search result. In step S 507, the search result is transmitted to the communication unit 209 of the client device 20 by the communication unit 206. Note that the search result output in S506 may be displayed on the display device 105 of the server apparatus 10 or the like.

In S508, the display control unit 211 causes the search unit 207 to display image information including information indicating a data group output as a search result on the display screen. FIG. 10 is an example of a screen that displays search results displayed on the client device 20. The elements already shown in FIG. 6 are denoted by the same reference numerals, and detailed description thereof is omitted. In the client device 20 according to the present embodiment, when a search result is shown, one division display area 1001 is displayed for each season in which similar divisions are specified, along with the division display area 602. Furthermore, as indicated by a region 1002, the date corresponding to the category specified as the similar category is expressed so as to be identifiable. By checking the area 1002, the user can know on which date the data group output as the search result is the acquired data group. Further, in the category display area 1001, not only one category specified as a search result but also one or more categories 1003 in which the specified category and the index are continuous are displayed. By arranging the section display area 602 including a plurality of sections and the section display area 1001 side by side, it is easy to compare not only the target section and its similar sections but also the periods having a width. Furthermore, in this embodiment, when any section (date) of the section display area 1001 is instructed by a user operation (for example, a tap operation on the touch panel), each data group corresponding to the section is displayed in the data display area. A list is displayed to 1004. As a result, the user can easily specify and browse a data group similar to the data group obtained in the specified date and several days up to that date from the accumulated time-series data.

<Example of processing for determining search range>
FIG. 7 is a flowchart illustrating an example of processing for determining a search range, which is executed by the server device 10 in S503. In step S 701, the determination unit 205 sets parameters based on the information indicating the target category acquired from the client device 20 in step S 502. The parameters set here are the season s, the index x, and the number of divisions n representing the influence range. For example, in the case of the conditions shown in FIG. 6, parameters such as season s = 6, index x = 5, and the number of divisions n = 3 representing the influence range are set.

In step S 702, the determination unit 205 assigns a value to the variable i so that the following processing in steps S 703 to S 708 is performed on the identifier ranges of all seasons of the time-series data held in the holding unit 203. . For example, referring to the definition table 405 shown in FIG. 4B, it can be seen that the time series data includes data corresponding to seasons 1-6. Therefore, iterative processing is performed while giving a value of 1 to 6 to i. However, seasons that include the target category are excluded. That is, when the variable i becomes the same as the season s set in S701, the following process does not have to be executed. In step S703, the determination unit 205 assigns a value to the variable j so that the following processing in steps S704 to S705 is performed in the range of 1 to n.

In S704, the determination unit 205 refers to the search history information held in the holding unit 203. Then, the category belonging to the partial time-series data of season i is specified from the category retrieved in the past as the category similar to the category of index xj in season s set in S701. FIG. 8A is a table 801 representing an example of search history information. The table 801 stores the season 802 and index 803 of the target segment specified as the query and the values of the season 804 and index 805 of the segment specified as the search result in association with each other. In S704, the value of index 805 is specified when the value of season 802 is season s, the value of index 803 is index xj, and the value of season 804 is i. However, if the corresponding data is not held in the search history information, an empty value is returned.

In S705, the determination unit 205 identifies the segment of the range of the index R (j) as the search range (candidate including a similar segment) in the season i, starting from the segment of the index identified in S704, and stores it in the RAM 102. To do. Specific Here, R (j) is a function for calculating an increase / decrease range of the index, and is represented by the following Expression 2, for example.

The table 806 in FIG. 8B is an example of data held in the RAM 102 as a result of S705. In the table 806, the season 807 and the index group 808 are associated with each other. Here, paying attention to the row 810, the processing of S703 to S705 in a cycle in which i = 5 is given as a season variable will be specifically shown. First, when j = 1, referring to the row 809 corresponding to the season s = 6, the index x = 5-1 = 4, and i = 5 in the table 801 representing the search history information, the index 805 of the search result is 8. Further, since the increase / decrease range is R (1) = [0, 2] when j = 1, the search range at season i = 5 has an increase / decrease range of R (1) = [0, 2] at index 8. The specified index is specified as a range of 8, 9, 10. Further, the process is repeated for j = 2 and 3, but information corresponding to the operation history information is not stored. Therefore, the search range specified for the season of i = 5 is the range of

indexes

8, 9, and 10. As described above, when processing is performed while changing the variable i and the variable j and the indexes are selected so as not to overlap, the result of the table 806 is obtained. In the case of i = 1, since there is no corresponding search history information, the index group is not specified, so the data is empty.

Returning to the description of the flowchart, in S706, the determination unit 205 determines whether the number of divisions specified for the season i is greater than the threshold value Nd. The threshold value Nd is a threshold value set in advance to provide a population sufficient for subsequent search processing. If the number of identified segments is greater than the threshold value Nd, the process proceeds to S505 (Yes in S706). On the other hand, if the number of identified sections is equal to or less than the threshold value Nd, the process proceeds to S707 (No in S706).

In S707, the determination unit 205 searches season i for a segment having an index in the range of W (j) starting from the segment with the same index x as the designated target segment in the partial time series data of season i. Specify as a range. Here, W (j) is a function for calculating an increase / decrease range of the index, and is represented by the following Expression 3, for example.
W (j) = [− j, j] Equation 3

However, at the stage where S707 is processed, j = n is given to the variable j in S703. Therefore, the search range of season i is a section having an index of the range of W (n) values starting from the index x, and is added to the table 806. For example, in the table 806, since the number of sections corresponding to i = 1 is 0, since x = 5 and n = 3, the ranges of 2, 3, 4, 5, 6, 7, and 8 are added as search ranges. The

The above is the processing for determining the search range based on the relative position of the target section in the present embodiment. In the present embodiment, the amount of calculation can be reduced by using the past search history in the process of determining the search range. However, the narrowing down method is not limited to this. Moreover, Formula 2 and Formula 3 are examples, and do not limit the present invention.

<Example of search processing>
FIG. 9 is a flowchart illustrating an example of search processing executed by the server apparatus 10 in S504. In S901, the search unit 207 reads a data group included in the section specified as the target section from the time series data held. That is, the data group of the section of index x in the partial time series data of season s is extracted. In S902, the search unit 207 assigns values to the variables i and k so that the processes in S902 to S904 are executed for all the search range categories determined by the process in S503. A value that is an identifier of a season held in the table 806 is given to i, and an index identifier value is given to k.

In S903, the search unit 207 reads data belonging to the category of index k in the partial time series data of season i from the time series data. In S904, the search unit 207 calculates the similarity between the data group read in S901 and the data group read in S903. The similarity between data uses a general method for measuring the similarity of time series data. Specifically, each extracted data is arranged in order of time, and a distance representing the similarity between the data is calculated. For example, Euclidean distance, Dynamic time warping or Earth Move's distance, edit distance, Hamming distance, Jaccard coefficient distance, Manhattan distance, Mahalanobis distance, etc. are used. The data group may be divided according to the type of data, and the similarity between each data may be calculated. When the data is empty, a predetermined sufficiently large value may be output as a distance representing the similarity. The search unit 207 generates similarity information shown in FIG. 8C based on the resulting similarity and stores it in the RAM 102. FIG. 8C shows an example of generated similarity information. In the table 811, a distance 814 representing the calculated similarity is associated with each category represented by the season 812 and the index 813.

In S905, the search unit 207 specifies a category similar to the target category for each season based on the similarity information generated in S704. For example, a segment with the shortest distance 814 representing the similarity is selected for each season. Note that a distance threshold for selection may be defined in advance, and a plurality of sections below the threshold may be selected, or may not be selected if they are above the threshold. The search unit 207 causes the RAM 102 to store information indicating the classification specified for each season. FIG. 8D shows an example of information indicating the classification specified for each season. As a search result for the target category of the conditions shown in FIG. 6 (season s = 6, index x = 5, number of segments representing the influence range n = 3), one segment is identified from

seasons

5, 4 and 2, respectively. It shows that. This information is output in the subsequent S505.

According to the first embodiment described above, a data group similar to the data group obtained in a predetermined unit time is efficiently identified from the accumulated time series data and presented to the user. can do. In particular, it is possible to efficiently specify a past similar data group from time-series data repeated approximately at a predetermined cycle, such as various data collected regarding cultivation of agricultural products.

In the present embodiment, an example in which a single category is designated as the target category to be used as a query for search processing has been shown. However, a plurality of categories may be designated, or non-contiguous categories may be designated. Good. In that case, a plurality of sections output as search results can be displayed in a list, or a time range including a plurality of sections can be displayed as a similar range. In the present embodiment, a system including a plurality of devices has been described. However, any device may include other devices. For example, in the information processing apparatus described as the server apparatus 10, the present embodiment can be realized even with a configuration in which time-series data is input and search results are displayed.

<Second Embodiment>
In the first embodiment, the client device 20 receives and displays the search result performed in the server device 10 every time the client device 20 instructs execution of the search process. The client device 20 can be a portable device such as a tablet PC. If the communication environment is in place, the user is not limited in the location where the client device 20 is operated, Data search can be performed. However, there are cases where the communication environment of the client device 20 is not sufficiently maintained in the entire field, such as when the field where the crop is grown is vast. Therefore, in the second embodiment, the client device 20 further includes a configuration that caches the search results received from the server device 10, and thereafter browses the minimum search results even in an offline environment. Here is an example where The user, for example, operates the client device 20 in a communicable state before instructing to update the cache before heading to the field. If necessary in the field, a data group acquired in the past can be browsed.

<System configuration>
The system configuration according to the second embodiment and the hardware configuration of each device are the same as those of the first embodiment described with reference to FIG. FIG. 11 is a diagram illustrating an example of a software configuration of the system according to the second embodiment. The functional units common to those in FIG. 2 are denoted by the same reference numerals, and detailed description thereof is omitted. The client device 20 according to the second embodiment includes a creation unit 1101. The creation unit 1101 is realized by the CPU 111 developing a program stored in the ROM 113 in the RAM 112 and executing processing to be described later. The creation unit 1101 creates cache data that stores search results acquired via the communication unit 209 and holds the cache data in the holding unit 1102. The holding unit 1102 is a functional unit of the HDD 117 or the RAM 112. Further, in the second embodiment, the accepting unit 210 accepts an input of an operation instructing to update and display a cache in the client device 20 in addition to an operation of designating a target category to be a search query.

<Process flow>
In the present embodiment, the flow of processing executed by each element of the information processing system will be described with reference to the sequence diagram of FIG. However, the steps common to those in FIG. 5 are denoted by the same reference numerals, and detailed description thereof is omitted. As in the first embodiment, the following is an example of processing time series data obtained by crop cultivation that repeats the one-year period from April to March of the following year as a first unit time. This embodiment will be described as an example.

In S1201, the accepting unit 210 of the client device 20 accepts an input for instructing cache update. In this embodiment, the user gives an instruction to update the cache by operating the user interface screen displayed on the display device 115.

FIG. 13A is an example of a user interface screen in the present embodiment. In FIG. 13, the reference date can be set by pressing a reference date setting button 1301. Here, the reference date corresponds to a category transmitted to the server device 10 as a search target category. In this embodiment, the date of the day is input as an initial value. This is because the data that the user who subsequently takes out the client device 20 and wants to perform farming work wants to confirm immediately even in a place where the communication environment is not prepared is likely to be time-series data similar to that day. However, the initial value may be the date of the previous day, for example. In addition, when there is little data obtained on the current day and it is difficult to calculate the similarity in the subsequent search process, it is possible to add modifications such as using the previous day as a reference.

Further, a range selection slider 1302 is displayed on the screen of FIG. 13A. Similar to the range adjustment slider 604 in FIG. 6, the range selection slider 1302 can set the range of categories to be considered in the search process in addition to the category corresponding to the reference date. For example, in FIG. 13A, the range for the previous three days is designated with 2017/4/6 as a reference. In response to pressing of the update button 1303 by the user, an instruction can be given to update the cache data stored in the client device 20 under the set conditions.

Returning to the description of the sequence in FIG. 12, in step S1202, the communication unit 209 transmits information on the designated reference date to the communication unit 206 of the server device 10 as information on the target category. For example, in the case of FIG. 13A, the information that the season s = 6, the index x = 5, and the number of divisions n = 3 indicating the influence range is sent to the server device 10.

In the second embodiment, when the search result transmitted in S507 is received, in S1202, the creation unit 1101 acquires the difference between the received cache data and the cache data already held in the holding unit 1102. For example, the table 1401 in FIG. 14A is an example of the latest search result received from the server device 10, and the table 1402 in FIG. 14B is an example of cache data held in the holding unit 1102. The creation unit 1101 acquires a category that exists only in the table 1401 that is the latest search result as a category to be added to the cache. Then, a section that exists only in the table 1402 that is old cache data is acquired as a section to be deleted from the cache. A table 1403 in FIG. 14C is an example of difference data acquired by the creation unit 1101.

According to the second embodiment described above, only a data group similar to the data group acquired on the reference date is cached in the client device 20 from the enormous time series data. As a result, the user can browse the minimum necessary data even when working in an offline environment. FIG. 13B is an example of a user interface screen displayed on the client device 20 when the user inputs an instruction to call and display a cache. While displaying in the same format as the search result display screen shown in FIG. 10, the icon 1304 explicitly indicates that the cache data is being displayed. Users can browse past data groups with the same operational feeling as searching in the online environment, and at the same time easily identify search results in the online environment and display of cached data in the offline environment. Can do.

<Other embodiments>
The present invention supplies a program that realizes one or more functions of the above-described embodiments to a system or apparatus via a network or a storage medium, and one or more processors in a computer of the system or apparatus read and execute the program This process can be realized. It can also be realized by a circuit (for example, ASIC) that realizes one or more functions.

The present invention is not limited to the above embodiment, and various changes and modifications can be made without departing from the spirit and scope of the present invention. Therefore, in order to make the scope of the present invention public, the following claims are attached.

This application claims priority on the basis of Japanese Patent Application No. 2018-114902 filed on June 15, 2018, the entire contents of which are incorporated herein by reference.

Claims

Acquisition means for acquiring time-series data in which data acquired every first unit time is continuous;
A management means for dividing the time series data into partial time series data for each second unit time longer than the first unit time and holding the divided time series data in a holding unit;
A specifying means for specifying, as a search target, any of the data acquired for each first unit time among the time-series data;
(A) a temporal relative position from a reference position in the partial time-series data for each second unit time of the data acquired for each first unit time specified as the search target; (B) A temporal relative position from the reference position in the partial time-series data for each second unit time of each of the data acquired for each first unit time included in the time-series data. And a determination means for determining a search range for searching for data similar to the search target among the time series data,
Search means for searching for data similar to the determined data from data included in the determined search range;
An information processing apparatus comprising:
The determining means includes, among the partial time series data for each second unit time, out of the partial time series data for each second unit time of the data determined by the determining means as a target. The information processing apparatus according to claim 1, wherein data existing at a relative position close to a temporal relative position is determined as the search range.
The management means includes information indicating partial time-series data for each second unit time belonging to each of the data acquired in the first unit time included in the time-series data, and the second The information processing apparatus according to claim 1 or 2, wherein information indicating a relative position in partial time-series data for each unit time is associated with and held in the holding unit.
The time series data is obtained by collecting a plurality of types of data in time series,
The search means obtains data similar to the specified data based on the result of calculating the similarity between the specified data and each of the data included in the search range for each type of the data. The information processing apparatus according to claim 1, wherein the information processing apparatus is specified.
5. The information processing according to claim 1, wherein the second unit time is a length of time corresponding to one cycle in which the content of the data is approximately repeated. apparatus.
The data acquired in the first unit time includes data obtained by collecting information representing at least one of the temperature, weather, humidity, growth state, image, and work record acquired regarding the cultivation of the crop. The information processing apparatus according to claim 1, wherein:
The information processing apparatus according to claim 6, wherein the information processing apparatus is a management apparatus that manages data relating to cultivation of agricultural crops.
A step of acquiring time-series data in which data acquired every first unit time is continuous by an acquisition unit;
Dividing the time series data by the management means into partial time series data for each second unit time longer than the first unit time and holding the divided time series data in a holding unit;
A step of specifying, as a search target, any one of the data acquired for each first unit time among the time series data by an specifying unit;
(A) the time acquired from the reference position in the partial time-series data for each second unit time of the data acquired for each first unit time specified as the search target by the determining means; A relative position, and (b) time from the reference position in the partial time-series data for each second unit time of the data acquired for each first unit time included in the time-series data. Determining a search range for searching for data similar to the search target in the time series data based on a relative position;
Searching for data similar to the determined data from the data included in the determined search range by a search means;
An information processing apparatus control method comprising:
Computer
Acquisition means for acquiring time-series data in which data acquired every first unit time is continuous;
A management means for dividing the time series data into partial time series data for each second unit time longer than the first unit time and holding the divided time series data in a holding unit;
A specifying means for specifying, as a search target, any of the data acquired for each first unit time among the time-series data;
(A) a temporal relative position from a reference position in the partial time-series data for each second unit time of the data acquired for each first unit time specified as the search target; (B) A temporal relative position from the reference position in the partial time-series data for each second unit time of each of the data acquired for each first unit time included in the time-series data. And a determination means for determining a search range to be searched out of the time series data based on
Search means for searching for data similar to the search target from data included in the determined search range,
A program for causing an information processing apparatus to function.
A computer-readable storage medium storing the program according to claim 9.
An information processing system that has a server device and a client device and retrieves data similar to data designated as a target from time-series data in which data acquired in a first unit time is continuous,
The client device is
Receiving means for receiving an input for designating any of the data acquired in the first unit time of the time series data;
The server device
Acquisition means for acquiring time-series data in which data acquired every first unit time is continuous;
A management unit for dividing the time series data into partial time series data for each second unit time longer than the first unit time and holding the divided time series data in a holding unit;
Among the time-series data, a specifying unit that specifies, as a search target, any of data specified by an input received by the receiving unit;
(A) a temporal relative position from a reference position in the partial time-series data for each second unit time of the data acquired for each first unit time specified as the search target; (B) A temporal relative position from a reference position in the partial time-series data for each second unit time of each of the data acquired for each first unit time included in the time-series data. A determination means for determining a search range to be searched out of the time-series data based on
Search means for searching for data similar to the search target from data included in the determined search range,
Furthermore, the client device
Display control means for displaying information indicating data output as a search result by the search means on a display screen;
An information processing system comprising:
The display control means further displays, on the display screen, a user interface for designating any one of the data acquired in the first unit time among the time series data.
12. The information processing system according to claim 11, wherein a date corresponding to the data designated as the target and a date corresponding to one or more data are further displayed on the user interface.
The date and time information corresponding to the specified data is arranged side by side on the display screen displayed on the client device, with information corresponding to data output as a search result by the search means. Item 13. The information processing system according to Item 11 or 12.
The client device further includes holding means for holding cache data of information indicating data output as a search result by the search means,
The said display control means displays the information which shows the data output as a search result by the said search means on the said display screen based on the cache data hold | maintained at the said holding means. Information processing system.
The information according to any one of claims 11 to 14, wherein the second unit time is a length of time corresponding to one cycle in which the content of the data is approximately repeated. Processing system.
The data acquired in the first unit time includes data obtained by collecting information representing at least one of the temperature, weather, humidity, growth state, image, and work record acquired regarding the cultivation of the crop. The information processing system according to claim 11, wherein:
The information processing system according to any one of claims 11 to 16, wherein the information processing system is a system that manages data relating to cultivation of agricultural crops.
A program that causes a computer to function as the client device according to claim 11.
A computer-readable storage medium storing the program according to claim 18.
A control method for an information processing system that includes a server device and a client device, and retrieves data similar to data designated as a target from time-series data obtained by connecting data acquired in a first unit time. ,
In the client device,
A step of receiving an input designating any one of the data acquired in the first unit time among the time series data by an accepting unit;
In the server device,
A step of acquiring time-series data in which data acquired every first unit time is continuous by an acquisition unit;
Dividing the time series data by the management means into partial time series data for each second unit time longer than the first unit time and holding the divided time series data in a holding unit;
A step of specifying, as a search target, any one of the data specified by the input among the time series data by an specifying unit;
(A) the time acquired from the reference position in the partial time-series data for each second unit time of the data acquired for each first unit time specified as the search target by the determining means; Relative position and (b) time from the reference position in the partial time-series data for each second unit time of the data acquired for each first unit time included in the time-series data Determining a search range to be searched out of the time series data based on the relative position;
A step of searching for data similar to the search target from data included in the determined search range by a search means,
Furthermore, in the client device,
A step of displaying information indicating data output as a search result in the server device on a display screen by a display control unit;
A method for controlling an information processing system comprising: