JP2010009112A - Apparatus and method for generating time series data - Google Patents

Apparatus and method for generating time series data Download PDF

Info

Publication number
JP2010009112A
JP2010009112A JP2008164540A JP2008164540A JP2010009112A JP 2010009112 A JP2010009112 A JP 2010009112A JP 2008164540 A JP2008164540 A JP 2008164540A JP 2008164540 A JP2008164540 A JP 2008164540A JP 2010009112 A JP2010009112 A JP 2010009112A
Authority
JP
Japan
Prior art keywords
time series
state
partial
event
generation unit
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
JP2008164540A
Other languages
Japanese (ja)
Inventor
Shigeaki Sakurai
茂明 櫻井
Original Assignee
Toshiba Corp
株式会社東芝
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Toshiba Corp, 株式会社東芝 filed Critical Toshiba Corp
Priority to JP2008164540A priority Critical patent/JP2010009112A/en
Publication of JP2010009112A publication Critical patent/JP2010009112A/en
Application status is Pending legal-status Critical

Links

Images

Abstract

<P>PROBLEM TO BE SOLVED: To provide a data generation apparatus for generating data suitable for analysis of factors of time variation. <P>SOLUTION: The data generation apparatus includes: an input unit 101 for inputting input data having identification information, an event, and an occurrence date of the event in association with each other; an event time series generation unit 102 for generating an event time series in which events are arranged in order of occurrence date; a state time series generation unit 103 for generating a state time series representing transition of state of an object following occurrence of events included in the event time series; a partial time series generation unit 104 for extracting a time series including a plurality of states matching a state transition pattern from the state time series and generating a partial state time series having the extracted time series in association with identification information corresponding to the state time series; and an attribute value time series generation unit 106 for acquiring attribute information corresponding to the identification information from an attribute value storage unit 125 and generating an attribute time series having the acquired attribute information in association with each of states included in the partial state time series. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an apparatus and a method for generating time-series data used for analysis from data representing a time change of an object detected from an RFID (Radio Frequency Identification) tag or the like attached to the object of data analysis.

  There are various techniques for generating time-series data that changes with time for data analysis and the like. Patent Document 1 proposes a technique for generating time-series data obtained by integrating a plurality of time-series data by interpolating missing data by combining a plurality of time-series data. In Patent Document 2, Web pages collected by a search key are clustered based on information on the URL, and series data is generated by dividing pages included in a specific cluster based on time information. Techniques to do this have been proposed.

  On the other hand, a technique for analyzing temporal changes such as movement of an object by detecting an RFID tag attached to the object to be analyzed is also known. With a conventional analysis technique using data collected from an RFID tag, the relevance of a plurality of RFID tags measured simultaneously can be analyzed. In addition, it is possible to analyze the transition of the position of the same RFID tag with time and the result.

JP 2003-44518 A JP 2006-331089 A

  However, the conventional method has a problem that a plurality of operations (factors) cannot be analyzed from the time series of one tag. For example, when a plurality of users move products with the same tag, it has been impossible to distinguish and analyze that they have been moved separately by a plurality of users.

  The present invention has been made in view of the above, and provides an apparatus and a method capable of generating data suitable for analyzing a factor of time change from time-series data representing time change of an object. With the goal.

  In order to solve the above-described problems and achieve the object, the present invention relates to identification information for identifying an object and attribute storage that stores attribute information representing the attribute of the object in association with each other, and the identification For each of the identification information included in the input data, an input unit that inputs a plurality of input data that associates information, an event that has occurred with respect to the object of the identification information, and an occurrence date and time of the event A time series in which the events are arranged in order of occurrence date and time, an event time series generation unit that generates an event time series in which the identification information is associated, and the occurrence of the event included in the event time series A state time series generation unit that generates a state time series that represents the state transition of the target object and that associates the identification information corresponding to the event time series, and a plurality of the states are determined in advance. A partial state in which a time series including a plurality of the states that match a state transition pattern arranged in a predetermined order is extracted from the state time series, and the identification information corresponding to the state time series is associated with the extracted time series A partial time series generating unit for generating a time series; and for each partial state time series, acquiring the attribute information corresponding to the identification information associated with the partial state time series from the attribute storage unit, An attribute time series generation unit that generates an attribute time series in which the acquired attribute information is associated with each state included in the state time series.

  The present invention is also a method capable of executing the above apparatus.

  According to the present invention, there is an effect that it is possible to generate data suitable for analyzing the cause of the time change from the time series data representing the time change of the object.

  Exemplary embodiments of an apparatus and a method according to the present invention will be described below in detail with reference to the accompanying drawings.

(First embodiment)
For example, when a product to which an RFID tag is attached is used as an object for data analysis, data representing the progress of product movement can be collected by using a plurality of sensors capable of detecting the passage of the RFID tag. . However, as described above, with such data alone, for example, it is impossible to distinguish and analyze the movements of products caused by a plurality of users.

  In addition, data collected from individual RFID tags may not always be measured at the same time even if the transition is caused by the same factor. It was difficult to analyze the chronological relationship of.

  Therefore, the data generation device according to the first embodiment generates a state time series representing the transition of the state of the product from the data representing the movement progress of the product, and the partial state obtained by dividing the state time series according to a predetermined rule Generate a time series. This makes it possible to generate data suitable for analyzing the cause of time change. In addition, the data generation device according to the first embodiment combines a plurality of partial state time series related to a plurality of RFID tags, which are considered to have changed due to the same factor. Thereby, the relationship between a plurality of factors corresponding to a specific product can be analyzed.

  In the following, a data generation apparatus that generates time-series data for analysis based on input data representing the movement history of products collected at a clothing store that handles products with RFID tags will be described as an example. The applicable data collection environment is not limited to this, and may be a retail store such as a supermarket. More generally, this implementation is performed on input data collected from an environment where there is a movable object (object) on which the RFID tag can be installed and the specific RFID reader can record the date and time when the RFID tag passes. The method of the form can be applied.

  Here, an example of an environment in which the above input data is collected will be described with reference to FIG. FIG. 1 is a diagram illustrating an example of a data collection environment. As shown in the figure, in the first embodiment, a data collection environment including a warehouse 10 for storing products and a store 20 for displaying and selling products will be described as an example.

  In the store 20, shelves 1 to 4 for displaying products, a fitting room for customers to try on products, and a cash register for purchasing products are installed. Further, the passage of the RFID tag can be detected between the warehouse 10 and the store 20, between the shelf 1 to the shelf 4 and the store 20, between the fitting room and the store 20, and between the cash register and the store 20. Gates 31 to 37 are installed.

  It is assumed that each product is previously provided with an RFID tag storing a tag ID that can identify the product individually. By using this environment, each time a product passes through the gate, the date and time of passage, the gate passed, and the tag ID of the RFID tag can be collected. Collected data (hereinafter referred to as tag ID data) is input as input data to the data generation apparatus of the first embodiment.

  Hereinafter, the gate 31 may be referred to as an inter-warehouse gate, the gates 32 to 35 may be referred to as a shelf 1 gate to a shelf 4 gate, the gate 36 may be referred to as a fitting room gate, and the gate 37 may be referred to as a register gate.

  FIG. 2 is a block diagram illustrating a configuration of the data generation device 100 according to the first embodiment. As shown in FIG. 2, the data generation device 100 includes a tag ID data storage unit 121, a state rule storage unit 122, a division rule storage unit 123, a combination rule storage unit 124, an attribute value storage unit 125, and an input Unit 101, event time series generation unit 102, state time series generation unit 103, partial time series generation unit 104, combined time series generation unit 105, attribute value time series generation unit 106, output unit 107, It has.

  The tag ID data storage unit 121 stores tag ID data collected in the data collection environment as shown in FIG. FIG. 3 is a diagram illustrating an example of the data structure of tag ID data stored in the tag ID data storage unit 121. As illustrated in FIG. 3, the tag ID data storage unit 121 stores tag ID data in which an event has occurred, the event that has occurred, and the tag ID.

  In the present embodiment, since the detection of the RFID tag corresponds to an event, the date and time when the RFID tag is detected at each gate is set. In addition, an event (event information) indicating that one of the gates has been passed is set in the event. In addition, you may comprise so that the name of the gate which passed may be set instead of an event. As the tag ID, the tag ID read from the RFID tag at the corresponding gate is set.

  Returning to FIG. 2, the state rule storage unit 122 stores a state rule that defines a rule for state transition of an object caused by an event that has occurred. FIG. 4 is a diagram illustrating an example of the data structure of the state rule stored in the state rule storage unit 122. As illustrated in FIG. 4, the state rule storage unit 122 stores a state rule that associates a state before transition, an event, and a state after transition by the event (state after transition). The state represents whether the item exists in the warehouse or the store. For example, the first state rule in FIG. 3 represents that the state where the commodity exists in the warehouse transitions from the state where the commodity passes through the inter-warehouse gate to the state where the commodity exists in the store. In this embodiment, the initial state is “warehouse”.

  Returning to FIG. 2, the division rule storage unit 123 stores a division rule for dividing the state time series generated by the state time series generation unit 103 (described later). FIG. 5 is a diagram illustrating an example of the data structure of the division rule stored in the division rule storage unit 123. As illustrated in FIG. 5, the division rule storage unit 123 stores a state transition pattern in which a plurality of states are arranged in a predetermined order as a division rule. For example, the first state transition pattern in FIG. 4 represents a state transition pattern in which the vehicle moves from the warehouse to the store and then moves from the store to any shelf (shelf x (x = 1 to 4)). The partial time series generation unit 104 described later generates a partial state time series including a plurality of states that match the state transition pattern. The division rule is not limited to this.

  Returning to FIG. 2, the combination rule storage unit 124 stores a combination rule that defines a condition regarding a state for combining a plurality of partial state time series. FIG. 6 is a diagram illustrating an example of the data structure of the combination rules stored in the combination rule storage unit 124. As illustrated in FIG. 6, the combination rule storage unit 124 stores a combination rule that associates a tag ID, a position, a combined state, and a time condition.

  Tag ID represents the conditions regarding tag ID corresponding to the some partial state time series to combine. In the figure, an example is shown in which “different” is set, which indicates that the tag IDs corresponding to the partial state time series to be combined are different.

  The position represents whether a plurality of partial state time series are combined in a state existing in a state included in the partial state time series. In the figure, there is an example in which “Final” indicating that the state is combined at the end of the partial state time series and “Intermediate” indicating that the state is combined at a state other than the beginning / end of the partial state time series are set. It is shown.

  The time condition represents a condition related to the time of the combined state. As the time condition, for example, a condition related to the occurrence date and time of the event that causes the transition to the combined state is set. In the first combination rule in the figure, an example of a time condition indicating that the difference in the occurrence date and time of the event that caused the transition to the combined state is within one minute is shown. Also, in the second combination rule in the figure, the time from the occurrence date and time of the event that caused the transition to the combined state to the occurrence date and time of the event that caused the transition to the next state is duplicated. An example of a time condition to represent is shown.

  Returning to FIG. 2, the attribute value storage unit 125 stores attribute information (attribute value) representing the attribute of the product to which the tag ID is assigned. FIG. 7 is a diagram illustrating an example of a data structure of data stored in the attribute value storage unit 125. As illustrated in FIG. 7, the attribute value storage unit 125 stores data in which the tag ID is associated with the product type to which the tag ID is assigned and the color of the product as an attribute value.

  The tag ID is identification information for identifying individual products. That is, when there are a plurality of products having the same product type and color, a different tag ID is assigned to each product. Therefore, as in the first data in the figure, a plurality of tag IDs may be associated with the same attribute value.

  The tag ID data storage unit 121, the state rule storage unit 122, the division rule storage unit 123, the combination rule storage unit 124, and the attribute value storage unit 125 are an HDD (Hard Disk Drive), an optical disk, a memory card, a RAM (Random It can be constituted by any storage medium generally used such as (Access Memory).

  Returning to FIG. 2, the input unit 101 inputs tag ID data from the tag ID data storage unit 121 as input data. The input source is not limited to the tag ID data storage unit 121, but may be configured to input from another storage medium. Further, for example, it may be configured to input from another device via a network.

  The event time series generation unit 102 divides the tag ID data for each tag ID, and generates an event time series that is time series data arranged in the order of event occurrence.

  The state time series generation unit 103 applies a state rule to the event time series, and generates a state time series representing the state transition of the product with the tag ID.

  The partial time series generation unit 104 divides the state time series into a plurality of partial state time series with reference to the division rule. Specifically, the partial time series generation unit 104 extracts a plurality of states that match a state transition pattern stored in the division rule storage unit 123 from a plurality of states included in the state time series, and extracts the plurality of extracted states. A partial state time series including the states of is generated.

  The combined time series generation unit 105 generates a combined time series obtained by combining the partial state time series with reference to the combination rule. For example, the combined time series generation unit 105 refers to the combination rule as shown in FIG. 6, the corresponding tag IDs are different from each other (different types), and the last states of the partial state time series are both “purchase”, A combined time series is generated by combining a plurality of partial state time series whose corresponding event occurrence dates and times are within one minute.

  The attribute value time series generation unit 106 sends the attribute value of the tag ID corresponding to each state included in the time series from the attribute value storage unit 125 for each of the combined time series generated by the combined time series generation unit 105. Acquire and generate an attribute value time series that is a time series in which the acquired attribute values are associated instead of the tag ID.

  The output unit 107 outputs the generated attribute value time series. For example, the output unit 107 outputs the attribute value time series to a storage unit (not shown) prepared for outputting the attribute value time series. The output method is not limited to this, and for example, the attribute value time series may be output to another device connected via a network.

  Next, data generation processing by the data generation apparatus 100 according to the first embodiment configured as described above will be described with reference to FIGS. 8 and 9. 8 and 9 are flowcharts showing the overall flow of data generation processing in the first embodiment.

  First, the input unit 101 acquires unprocessed tag ID data from the tag ID data storage unit 121 (step S801). Next, the event time series generation unit 102 determines whether or not the tag ID included in the acquired tag ID data has already been acquired (step S802).

  For example, it is assumed that “04/01 15:00, shelf 1, r1” which is the fourth tag ID data in FIG. 2 is acquired. In this case, the first tag ID data “04/01 15:00, between warehouse stores, r1” has already been taken out. Therefore, the event time series generation unit 102 determines that the tag ID = r1 has already been acquired.

  If the tag ID has not already been acquired (step S802: NO), the event time series generation unit 102 initializes the newly acquired event time series for the tag ID (step S803). For example, when “04/01 15:01, between warehouse stores, r2” which is the second tag ID data in FIG. 2 is acquired, the event time series generation unit 102 sets the event time series corresponding to the tag ID = r2. Is newly generated, and the tag ID data is registered as the first data of the generated event time series.

  If the tag ID has already been acquired (step S802: YES), the event time series generation unit 102 extends the event time series by adding tag ID data to the already generated event time series ( Step S804). For example, when the fourth tag ID data “04/01 15:00, shelf 1, r1” in FIG. 2 is acquired, the first tag ID data “04/01 15:00, warehouse” is acquired. The event time series is extended by adding the tag ID data to the event time series having “in-store, r1” as the top data.

  Next, the event time series generation unit 102 determines whether or not unprocessed tag ID data exists in the tag ID data storage unit 121 (step S805). If unprocessed tag ID data exists (step S805: YES), the unprocessed tag ID data is acquired and the process is repeated (step S801).

  When there is no unprocessed tag ID data (step S805: NO), the state time series generation unit 103 generates a state time series from the generated event time series (steps S806 to S811). ).

  For example, after acquiring the last tag ID data “04/04 19:00, warehouse store interior, r3” in FIG. 2 and extending the event time series, there is no unprocessed tag ID data. Therefore, the state time series generation process is executed.

  FIG. 10 is a diagram showing an example of an event time series generated by the event time series generation unit 102 from the tag ID data as shown in FIG. As illustrated in FIG. 10, the event time series generation unit 102 generates an event time series in which an event occurrence date / time, an event, and a tag ID are arranged in order of occurrence date / time. In the figure, for convenience of explanation, event time series for each tag ID in units of three columns are shown in a horizontal direction.

  In the state time series generation process, first, the state time series generation unit 103 acquires an event time series of an unprocessed tag ID from the generated event time series (step S806). Next, the state time series generation unit 103 initializes a state time series corresponding to the acquired event time series (step S807). For example, when the initial state of the state time series is “warehouse”, the state time series generation unit 103 sets “warehouse” as the initial state and sets a tag ID corresponding to the event time series. The state time series is initialized by

  Next, the state time series generation unit 103 extracts an event from the acquired event time series (step S808). Next, the state time series generation unit 103 refers to the state rules stored in the state rule storage unit 122 and the current state that is the state included in the acquired event time series, and determines the state after the transition. The state time series is set (step S809).

  For example, it is assumed that a state rule as illustrated in FIG. 4 is stored in the state rule storage unit 122. Further, it is assumed that “warehouse” is set in the initial state in all event time series. Then, it is assumed that an event is extracted from the data positioned first in the event time series corresponding to the tag ID = r1 in FIG. Since this data matches the state rule positioned first in FIG. 4, the state time series generation unit 103 determines the state after the transition as “inside the store”.

  Similarly, it is assumed that an event is extracted from data positioned second in the event time series corresponding to the tag ID = r1 in FIG. This data matches the third state rule in FIG. 4 because the current state is “in-store”. Therefore, the state time series generation unit 103 determines the state after the transition as “shelf 1”.

  Next, the state time series generation unit 103 determines whether there is an unprocessed event in the event time series (step S810). If it exists (step S810: YES), an unprocessed event is extracted and the process is repeated (step S808).

  For example, when the event time series corresponding to the tag ID = r1 in FIG. 10 is processed, immediately after the data “04/03 15:00, cash register, r1” located at the end of the event time series is extracted, It is determined that there is no unprocessed event in the event time series.

  When there is no unprocessed event (step S810: NO), the state time series generation unit 103 further determines whether an event time series with an unprocessed tag ID exists (step S811). If there is an event time series of an unprocessed tag ID (step S811: YES), an event time series of an unprocessed tag ID is acquired and the process is repeated (step S806).

  When there is no event time series of unprocessed tag IDs, that is, when a state time series is generated for event time series of all tag IDs (step S811: NO), the partial time series generation unit 104 A partial time series generation process (steps S812 to S817) for generating a partial state time series from the generated state time series is executed.

  For example, if the event time series corresponding to the tag ID = r3 in FIG. 10 is processed, the event time series of the unprocessed tag ID does not exist, and the partial time series generation process is executed.

  FIG. 11 is a diagram illustrating an example of a state time series generated by the state time series generation unit 103 from the event time series as illustrated in FIG. As shown in FIG. 11, the state time series generation unit 103 generates a state time series in which the occurrence date / time of the event that caused the state transition, the state, and the tag ID are arranged in the order of occurrence date / time.

  In the partial time series generation process, first, the partial time series generation unit 104 acquires a state time series of an unprocessed tag ID from the generated state time series (step S812). Next, the partial time series generation unit 104 extracts one state in order from the last state of the state time series (step S813). The reason why the states are sequentially extracted from the last state in the state time series is that the matching state transition pattern can be narrowed down earlier. The partial time series generation unit 104 may be configured to extract from the first state of the state time series.

  Next, the partial time series generation unit 104 refers to the division rule stored in the division rule storage unit 123, and selects a state transition pattern that matches the extracted state (step S814). Further, the partial time series generation unit 104 extracts, as a partial state time series, a partial state time series from the extracted state to the first state of the state transition pattern in accordance with the selected state transition pattern (step S815). .

  For example, assume that the division rule of FIG. 5 is stored in the division rule storage unit 123. Further, it is assumed that “purchase”, which is the state of the last data in the state time series of the tag ID = r1 in FIG. 11, is extracted as the state to be processed. In this case, the partial time series generation unit 104 compares the last state of each state transition pattern of the division rule with the extracted state, and extracts the third state transition pattern and the eighth state transition pattern. Furthermore, since the state before the extracted state “purchase” is “shelf 2 → inside store → fitting room → inside store”, the partial time series generation unit 104 compares these states with the state transition pattern, It is determined that the third state transition pattern matches the extracted state. Therefore, the partial time series generation unit 104 extracts “shelf 2 → inside store → fitting room → inside store → purchase” as the partial state time series. In addition, “shelf 2” is set as the next state. As described above, when there is a matching state transition pattern, the partial time series generation unit 104 repeats the process by setting the first state of the state transition pattern as the state to be processed next.

  Similarly, it is assumed that “purchase” that is the state of the last data in the state time series of the tag ID = r2 in FIG. 11 is extracted as the state to be processed. In this case, the partial time series generation unit 104 compares the last state of each state transition pattern of the division rule with the extracted state, and extracts the third state transition pattern and the eighth state transition pattern. Furthermore, since the state before the extracted state “purchase” is “shelf 2 → in the store”, the partial time series generation unit 104 compares the state with the state transition pattern to thereby obtain the eighth state transition. It is determined that the pattern matches the extracted state. Therefore, the partial time series generation unit 104 extracts “shelf 2 → inside store → purchase” as the partial state time series. In addition, “shelf 2” is set as the next state.

  After extracting the partial state time series, the partial time series generation unit 104 determines whether there is an unprocessed state in the state time series (step S816). For example, when the state time series corresponding to the tag ID = r1 in FIG. 11 is a processing target, immediately after the first data “04/01 15:00, in-store, r1” of this state time series is processed, It is determined that there is no unprocessed state in the state time series.

  If there is an unprocessed state in the state time series (step S816: YES), the next state is acquired and the process is repeated (step S813).

  If there is no unprocessed state (step S816: NO), the partial time series generation unit 104 further determines whether there is a state time series of an unprocessed tag ID (step S817). If there is an unprocessed tag ID state time series (step S817: YES), an unprocessed tag ID state time series is acquired and the process is repeated (step S812).

  When there is no unprocessed tag ID state time series (step S817: NO), the combined time series generation unit 105 executes combined time series generation processing (steps S818 to S826) for combining partial state time series. To do.

  For example, in the case of the state time series of FIG. 11, the state time series of the unprocessed tag ID does not exist after the state time series corresponding to the tag ID = r3 is extracted. Executed.

  FIG. 12 is a diagram illustrating an example of a partial state time series generated by the partial time series generation unit 104 from the state time series as illustrated in FIG. As illustrated in FIG. 12, the partial time series generation unit 104 generates a partial state time series in which tag IDs are associated.

  In the combined time series generation process, first, the combined time series generation unit 105 acquires an unprocessed partial state time series from the generated partial state time series (step S818). Further, the combined time series generation unit 105 acquires an unprocessed partial time series from the partial state time series different from the partial state time series acquired in step S818 (step S819).

  Next, the combined time series generation unit 105 refers to the combination rule stored in the combination rule storage unit 124 to determine whether or not the two extracted partial state time series can be combined (step S820). ).

  For example, it is assumed that a combination rule as illustrated in FIG. 6 is stored in the combination rule storage unit 124. Further, it is assumed that the first partial state time series and the second partial state time series in FIG. 12 are extracted. At this time, the same tag ID is given to these two partial state time series. On the other hand, since all of the combining rules in FIG. 6 are based on the condition that the tag IDs are different, the combined time series generation unit 105 determines that the partial state time series cannot be combined.

  As another example, assume that the first partial state time series and the sixth partial state time series of FIG. 12 are extracted. At this time, the two partial state time series have different tag IDs, but the final state does not match either “purchase” or “fitting room” described in the combination rule. For this reason, the combined time series generation unit 105 determines that the partial state time series cannot be combined.

  In contrast, it is assumed that the fifth partial state time series and the tenth partial state time series of FIG. 12 are extracted. At this time, the two partial state time series have different corresponding tag IDs, and the final state is “purchased”. Also, the difference between the date and time of occurrence of this final state is 1 minute (15: 00-15: 01). Therefore, since these two partial state time series match the condition of the first combination rule in FIG. 6, the combined time series generation unit 105 determines that these partial state time series can be combined.

  Further, as another example, the ninth partial state time series of FIG. 12 and the partial state time series corresponding to the tag ID = r4 not shown in FIG. 12 are “(04/02 09:00, shelf 2 ) → (04/02 11:10, in store) → (04/02 11:16, fitting room) → (04/02 11:20, in store) → (04/02 11:25, shelf 2) ”(below , Partial state time series d1) is extracted. At this time, the two partial state time series have different tag IDs, and there is a “fitting room” state in between. The time from this state “fitting room” to the next state “inside the store” is “04/02 11:05 to 04/02 11:20” in the ninth partial state time series. In the time series d1, it is “04/02 11:15 to 04/02 11:20”. That is, there is an overlapping part in the date and time range of both. Therefore, since these two partial state time series match the condition of the second combination rule in FIG. 6, the combined time series generation unit 105 determines that these partial state time series can be combined.

  When it is determined that the two extracted partial state time series can be combined (step S820: YES), the combined time series generation unit 105 extracts the partial state time series for which the combination rule is established in step S818 and extracts the partial state time series in step S818. Is temporarily stored as a partial state time series to be combined with the partial state time series extracted in step S821.

  For example, in the case of the fifth partial state time series and the tenth partial state time series in FIG. 12, the tenth partial state time series is stored as a partial state time series corresponding to the fifth partial state time series. The In the case of the ninth partial state time series and the partial state time series d1 in FIG. 12, the partial state time series d1 is stored as a partial state time series corresponding to the ninth partial state time series.

  If it is determined that the two extracted partial state time series cannot be combined (step S820: NO), the combined time series generation unit 105 does not establish a combination rule for the partial state time series extracted in step S818. The partial state time series is temporarily stored (step S822).

  For example, if the first partial state time series of FIG. 12 is extracted in step S818, the second partial state time series is a part where the combination rule is not satisfied with respect to the first partial state time series. Stored as state time series. In addition, when the first partial state time series and the sixth partial state time series in FIG. 12 are extracted, the sixth partial state time series is combined with the first partial state time series. Is stored as a partial state time series in which is not established.

  Next, the combined time series generation unit 105 determines whether or not there is an unprocessed partial state time series among the partial state time series different from the partial state time series acquired in step S818 (step S823). If there is an unprocessed partial state time series (step S823: YES), the next unprocessed partial state time series is acquired and the process is repeated (step S819).

  If there is no unprocessed partial state time series (step S823: NO), the combined time series generation unit 105 stores that the combined rule is established for the partial state time series extracted in step S818. The partial state time series that are present are combined (step S824). At this time, when there is no partial state time series to be combined, the combined time series generation unit 105 outputs a state combined time series composed of only the partial state time series extracted in step S818.

  For example, it is assumed that the partial state time series of FIG. 12 is given as all of the partial state time series to be processed, and the first partial state time series is to be combined. In this case, for the first partial state time series, since there is no other partial state time series that matches the combination rule in FIG. 6, the combined time series generation unit 105 performs the first partial state time series alone. Construct a state combination time series.

  Specifically, the combined time series generation unit 105 deletes information related to the date and time from the partial state time series, and generates a state combined time series as shown in FIG. 13 in which a tag ID is assigned to each state. To do. FIG. 13 shows an example of a state combination time series generated by the combined time series generation unit 105 in this way.

  On the other hand, it is assumed that the partial state time series of FIG. 12 is given as all of the partial state time series to be processed, and the fifth partial state time series is to be combined. In this case, as described above, the tenth partial state time series in FIG. 12 is to be combined. The combined time series generation unit 105 first combines the states of the final positions determined to be combined. Next, the combined time series generation unit 105 compares the states before the final position of each partial state time series. In this example, since the previous state is “in-store”, the combined time series generation unit 105 further combines this state. Subsequently, the combined time series generation unit 105 further compares the state before “in store”. In this case, since “matching room” and “shelf 2” do not match, the combined time series generation unit 105 compares the date and time associated with each state in order to determine the order in which the unconnected states are arranged. In this example, the date “04/03 14:35” associated with “Fitting Room” occurs after the date “04/02 11:25” associated with “Shelf 2”. The state of “shelf 2” is inserted before the state of “fitting room”.

  Regarding the tenth partial state time series, the state of “shelf 2” is the first state. Therefore, the combined time series generation unit 105 finally deletes the date and time information corresponding to each state, and generates a state combined time series as shown in FIG. 13 with the corresponding tag ID assigned to the state. .

  As another example of the state combination time series, a case where the ninth partial state time series and the partial state time series d1 in FIG. 12 are combined will be described. In the two partial state time series, the second coupling rule of FIG. 6 is established in the “fitting room” located in the middle. For this reason, the combined time series generation unit 105 first combines the state “fitting room”. Next, the combined time series generation unit 105 combines the previous states with this state as a reference. In this example, since the previous state matches with “inside the store”, the combined time series generation unit 105 combines the state “inside the store”. Furthermore, since the previous state “shelf 2” also coincides, the combined time series generation unit 105 also combines this state. Similarly, the states after the reference state “fitting room” are sequentially combined. For example, since the state after the state “fitting room” matches in order “in the store” and “shelf 2”, the combined time series generation unit 105 combines these states. As a result, the combined time series generation unit 105 finally selects “(r2: shelf 2, r4: shelf 2) → (r2: in store, r4: in store) → (r2: fitting room, r4: fitting room) → (r2: in store, r4: in store) → (r2: shelf 2, r4: shelf 2) "is generated.

  After generating the state combination time series in step S824, the combination time series generation unit 105 sets the partial state time series stored as the partial state time series that cannot be combined as a set of partial state time series to be processed next. (Step S825).

  For example, a combination rule as in FIG. 6 and a partial state time series as in FIG. 12 are given, and the first partial state time series in FIG. 12 is given as the partial state time series extracted in step S818. Suppose that In this case, in the first partial state time series, there is no other partial state time series that matches the combination rule of FIG. 6, so all the partial state time series except for the first partial state time series are partial It is set as a set of state time series.

  Next, the combined time series generation unit 105 determines whether there is a set of partial state time series to be processed (step S826). If there is a set of partial state time series to be processed (step S826: YES), a new partial state time series is acquired from this partial state time series set and the process is repeated (step S818).

  When there is no set of partial state time series to be processed (step S826: NO), the attribute value time series generation unit 106 generates attribute value time series from the generated state combination time series. Processing (step S827 to step S831) is executed.

  For example, after the last partial state time series in FIG. 12 is processed, the attribute value time series generation process is executed because there is no partial state time series to be processed.

  FIG. 13 shows an example of a state combination time series generated by the combined time series generation unit 105 from the partial state time series as shown in FIG.

  In the attribute value time series generation process, first, the attribute value time series generation unit 106 acquires an unprocessed state combination time series from the generated state combination time series (step S827). Next, the attribute value time series generation unit 106 acquires an unprocessed tag ID from the acquired state combination time series (step S828).

  Next, the attribute value time series generation unit 106 acquires an attribute value corresponding to the acquired tag ID from the attribute value storage unit 125, and converts the tag ID extracted from the state combination time series into the acquired attribute value. The attribute value time series is generated (step S829).

  For example, assume that attribute values as shown in FIG. 7 are stored in the attribute value storage unit 125. Further, it is assumed that tag ID = r1 is extracted from the first state combination time series of FIG. At this time, when the tag ID is converted by an attribute value obtained by concatenating all the attribute values of the tag ID, the attribute value time series generation unit 106 converts the tag ID = r1 as “skirt 1 / black”. Similarly, when the tag ID = r2 is extracted from the sixth state combination time series of FIG. 13, the attribute value time series generation unit 106 converts the tag ID = r2 to “jacket 1 / black”.

  Next, the attribute value time series generation unit 106 determines whether or not there is an unprocessed tag ID (step S830). For example, assume that the first state combination time series in FIG. 13 is extracted. After the tag ID = r1 is acquired and processed from this state combination time series, there is no other tag ID, so the attribute value time series generation unit 106 determines that there is no unprocessed tag ID. .

  If there is an unprocessed tag ID (step S830: YES), the attribute value time series generation unit 106 acquires the next unprocessed tag ID and repeats the process (step S828). If there is no unprocessed tag ID (step S830: NO), the attribute value time series generation unit 106 further determines whether or not an unprocessed state combination time series exists (step S831).

  If there is an unprocessed state combination time series (step S831: YES), the attribute value time series generation unit 106 acquires the next unprocessed state combination time series and repeats the process (step S827). When there is no unprocessed state combination time series (step S831: NO), the output unit 107 outputs the generated attribute value time series to the storage unit or the like (step S832), and the data generation process ends.

  FIG. 14 is a diagram illustrating an example of the attribute value time series generated by the attribute value time series generation unit 106 from the state combination time series as shown in FIG.

  As described above, the data generation apparatus according to the first embodiment can generate attribute value time series that can be used for analysis of time series patterns from the collected tag ID data. In the data generation device according to the first embodiment, a plurality of partial state time series related to a plurality of RFID tags can be combined by using a combination rule. Thereby, the relationship between a plurality of factors corresponding to a specific product can be analyzed.

(Second Embodiment)
For the attribute value time series generated by the data generation apparatus of the first embodiment, it is possible to apply a time series data analysis method for finding a characteristic time series pattern from time series data. It becomes. Therefore, in the second embodiment, an example of a data generation device having a function of analyzing attribute value time series will be described.

  FIG. 15 is a block diagram illustrating a configuration of a data generation device 1500 according to the second embodiment. As illustrated in FIG. 15, the data generation device 1500 has a configuration in which a time series pattern extraction unit 1508 and a time series pattern storage unit 1526 are added to the data generation device 100 according to the first embodiment. Other configurations and functions are the same as those in FIG. 2, which is a block diagram showing the configuration of the data generation device 100 according to the first embodiment.

  The time series pattern extraction unit 1508 regards the attribute value time series as one time series data by regarding the data consisting of the attribute value and state set as one item, and from this time series data, the characteristic item set A time series pattern that is a series is extracted.

  For example, the time series pattern extraction unit 1508 can regard the first attribute value time series in FIG. 14 as time series data by regarding “skirt 1 / black: warehouse” as one item.

The time series pattern extraction unit 1508 is, for example, “Shigeaki Sakurai, Yoichi Kitahara, Ryohei Orihara: Proposal of a new index for discovering characteristic time series patterns, Database Society of Japan Letters, 5, 1, 8, 153-157 (2006). ”(Hereinafter referred to as a reference), all the values of the series interest level, which is the evaluation value of the time series pattern defined by the following formula (1), are not less than the specified threshold value. Extract time-series patterns.

However, time series pattern s, s p partial time series pattern of the time series pattern s a function that calculates the number of time-series data including a time series pattern f (s), the number of time-series data N, alpha (≧ 0) is a series interest degree parameter.

  Note that the time series pattern extraction unit 1508 may be configured to analyze the attribute value time series by a method other than the method of the reference document.

  The time series pattern storage unit 1526 stores a time series pattern extracted by the time series pattern extraction unit 1508 and having a series interest degree value equal to or greater than a threshold value. The time-series pattern storage unit 1526 can be configured by any commonly used storage medium such as an HDD, an optical disk, a memory card, and a RAM.

  As described above, in the data generation apparatus according to the second embodiment, the time-series pattern extraction unit 1508 and the time-series pattern storage unit 1526 are provided to analyze the time-series relationship between a plurality of items. It becomes possible.

  The present invention is not limited to the above embodiment. For example, the attribute value time series generation unit 106 converts tag IDs by concatenating all attribute values stored in the attribute value storage unit 125, but concatenates only part of the attribute values. The tag ID may be converted.

  In the above description, the attribute value stored in the attribute value storage unit 125 is limited to a discrete value, but a numerical value may be used as the attribute value.

  Further, as a combination rule stored in the combination rule storage unit 124, for example, a combination rule using information other than the date and time is used, such as a rule for combining tag IDs described in the same receipt at the cash register. You may comprise. In this case, for example, data indicating that the tag ID is described in the same receipt is separately input and is configured to be referred to when the combination rule is applied.

  Further, the time series pattern extraction unit 1508 is configured to extract only the attribute value time series including only a part of the attribute values and then use the extracted attribute value time series as an input of the method of the reference document. May be.

  Next, the hardware configuration of the data generation apparatus according to the first or second embodiment will be described with reference to FIG. FIG. 16 is an explanatory diagram illustrating a hardware configuration of the data generation device according to the first or second embodiment.

  The data generation device according to the first or second embodiment communicates with a control device such as a CPU (Central Processing Unit) 51 and a storage device such as a ROM (Read Only Memory) 52 and a RAM 53 by connecting to a network. The communication I / F 54, an external storage device such as an HDD (Hard Disk Drive) and a CD (Compact Disc) drive device, a display device such as a display device, and an input device such as a keyboard and a mouse. A bus 61 is provided and has a hardware configuration using a normal computer.

  A data generation program executed by the data generation apparatus according to the first or second embodiment is a file in an installable format or an executable format, and is a CD-ROM (Compact Disk Read Only Memory), a flexible disk (FD). ), A CD-R (Compact Disk Recordable), a DVD (Digital Versatile Disk), and the like.

  Further, the data generation program executed by the data generation apparatus according to the first or second embodiment is stored on a computer connected to a network such as the Internet and is provided by being downloaded via the network. It may be configured. The data generation program executed by the data generation apparatus according to the first or second embodiment may be provided or distributed via a network such as the Internet.

  The data generation program according to the first or second embodiment may be provided by being incorporated in advance in a ROM or the like.

  The data generation program executed by the data generation apparatus according to the first or second embodiment includes the above-described units (input unit, event time series generation unit, state time series generation unit, partial time series generation unit, combined time) A module generation unit including a series generation unit, an attribute value time series generation unit, and an output unit). As actual hardware, the CPU 51 (processor) reads out and executes a data generation program from the storage medium. Are loaded on the main storage device, and the above-described units are generated on the main storage device.

  As described above, the apparatus and the method according to the present invention are suitable for an apparatus and a method for generating data used for analysis by inputting time-series data representing the movement progress of a product or the like to which an RFID tag is attached.

It is a figure which shows an example of a data collection environment. It is a block diagram which shows the structure of the data generation apparatus concerning 1st Embodiment. It is a figure which shows an example of the data structure of the tag ID data memorize | stored in the tag ID data storage part. It is a figure which shows an example of the data structure of the state rule memorize | stored in the state rule memory | storage part. It is a figure which shows an example of the data structure of the division rule memorize | stored in the division rule memory | storage part. It is a figure which shows an example of the data structure of the combination rule memorize | stored in the combination rule memory | storage part. It is a figure which shows an example of the data structure of the data memorize | stored in the attribute value memory | storage part. It is a flowchart which shows the whole flow of the data generation process in 1st Embodiment. It is a flowchart which shows the whole flow of the data generation process in 1st Embodiment. It is a figure which shows an example of the produced | generated event time series. It is a figure which shows an example of the produced | generated state time series. It is a figure which shows an example of the produced | generated partial state time series. It is a figure which shows an example of the produced | generated state connection time series. It is a figure which shows an example of the produced | generated attribute value time series. It is a block diagram which shows the structure of the data generation apparatus concerning 2nd Embodiment. It is explanatory drawing which shows the hardware constitutions of the data generation apparatus concerning 1st or 2nd embodiment.

Explanation of symbols

10 warehouse 20 store 31-37 gate 51 CPU
52 ROM
53 RAM
54 Communication I / F
61 Bus 100 Data Generation Device 101 Input Unit 102 Event Time Series Generation Unit 103 State Time Series Generation Unit 104 Partial Time Series Generation Unit 105 Combined Time Series Generation Unit 106 Attribute Value Time Series Generation Unit 107 Output Unit 121 Tag ID Data Storage Unit 122 State rule storage unit 123 Division rule storage unit 124 Join rule storage unit 125 Attribute value storage unit 1500 Data generation device 1508 Time series pattern extraction unit 1526 Time series pattern storage unit

Claims (10)

  1. An attribute storage unit that stores identification information for identifying the object and attribute information that represents the attribute of the object in association with each other;
    An input unit that inputs a plurality of input data in which the identification information, an event that has occurred with respect to the object of the identification information, and an occurrence date and time of the event are associated;
    For each of the identification information included in the input data, an event time series generation unit that generates an event time series in which the events are arranged in order of occurrence date and time, the event time series corresponding to the identification information,
    A time series that represents a transition of the state of the object associated with the occurrence of the event included in the event time series, and that generates a state time series that associates the identification information corresponding to the event time series A sequence generation unit;
    A time series including a plurality of states that match a state transition pattern in which a plurality of the states are arranged in a predetermined order is extracted from the state time series, and the extracted time series corresponds to the state time series. A partial time series generation unit for generating a partial state time series in which information is associated;
    For each partial state time series, the attribute information corresponding to the identification information associated with the partial state time series is acquired from the attribute storage unit, and for each of the states included in the partial state time series An attribute time series generation unit that generates an attribute time series in which the acquired attribute information is associated;
    A data generation device comprising:
  2. The partial time series generation unit generates a time series including the plurality of states that match the state transition pattern that defines the state transition pattern of the object according to a predetermined factor from the state time series. Extracting and generating the partial state time series in which the identification information corresponding to the state time series is associated with the extracted time series;
    The data generation device according to claim 1.
  3. The partial time series generation unit extracts a time series including a plurality of the states that match the state transition pattern from the state time series, and corresponds to the state time series for each state included in the extracted time series. Generating the partial state time series corresponding to the identification information;
    A combined time series generation unit for generating a state combined time series combining a plurality of the partial state time series satisfying a predetermined condition related to the plurality of states included in each of the plurality of partial state time series;
    The attribute time series generation unit acquires the attribute information corresponding to the identification information associated with the state from the attribute storage unit for each state included in the state combination time series, and for the state Generating the attribute time series associating the attribute information acquired by
    The data generation device according to claim 1.
  4. The combined time series generation unit includes a plurality of conditions that the states included in the partial state time series match and the occurrence date and time of the event that has been changed to the matching state satisfies a predetermined time condition regarding time. Generating the state combined time series by combining the partial state time series;
    The data generation device according to claim 3.
  5. The combined time series generation unit includes a plurality of the differences in the occurrence dates and times of the events that have been matched in a state included in the partial state time series and have transitioned to a matching state. Generating the state combined time series by combining partial state time series;
    The data generation device according to claim 4.
  6. The combined time series generation unit matches the states included in the partial state time series, and from the occurrence date and time of the event that has been changed to the matching state, the transition to the next state of the matching state Generating the state combination time series obtained by combining a plurality of the partial state time series that overlap the time until the occurrence date and time of the event;
    The data generation device according to claim 4.
  7. The combined time series generation unit matches the states included in the partial state time series, and matches the positions in the partial state time series of the matching states, and makes a transition to the matching state. Generating the state combination time series obtained by combining a plurality of partial state time series in which the occurrence date and time of the event satisfies the time condition;
    The data generation device according to claim 4.
  8. A time-series pattern extracting unit that extracts a time-series pattern representing the characteristics of the attribute time-series from the attribute time-series;
    The data generation device according to claim 1.
  9. The attribute storage unit stores the identification information, which is a tag ID for identifying an RFID (Radio Frequency Identification) tag attached to each of the objects, and the attribute information in association with each other,
    The input unit is detected by the tag ID and the event indicating that the passage of the RFID tag of the tag ID is detected by any one of a plurality of detection devices capable of detecting the passage of the RFID tag. A plurality of the input data in association with the occurrence date and time representing the date and time,
    The event time series generation unit is a time series in which the events are arranged in the order of occurrence date and time for each tag ID included in the input data, and generates the event time series in which the tag IDs are associated with each other.
    The state time series generation unit is a time series representing a state transition of the object accompanying the occurrence of the event included in the event time series, and associates the tag ID corresponding to the event time series Generating the state time series;
    The first partial time series generation unit extracts a time series including a plurality of the states that match the state transition pattern from the state time series, and associates the tag ID corresponding to the state time series with the extracted time series Generating the partial state time series,
    The attribute time series generation unit acquires, for each partial state time series, the attribute information corresponding to the tag ID associated with the partial state time series from the attribute storage unit, and generates the partial state time series. Generating the attribute time series in which the acquired attribute information is associated with each of the states included;
    The data generation device according to claim 1.
  10. An input step for inputting a plurality of pieces of input data in which identification information for identifying an object, an event that has occurred with respect to the object of the identification information, and an occurrence date and time of the event are associated;
    An event time series generating unit is a time series in which the events are arranged in the order of occurrence date and time for each of the identification information included in the input data, and an event time series in which the identification information is associated is generated. Generation step;
    The state time series generation unit is a time series representing a state transition of the object associated with the occurrence of the event included in the event time series, and is a state in which the identification information corresponding to the event time series is associated A state time series generation step for generating a time series;
    The partial time series generation unit extracts a time series including a plurality of states that match a state transition pattern in which a plurality of the states are arranged in a predetermined order from the state time series, and the states are extracted into the time series. A partial time series generation step for generating a partial state time series in which the identification information corresponding to the time series is associated;
    The attribute time series generation unit is associated with the partial state time series from an attribute storage unit that stores the identification information and attribute information representing the attribute of the object in association with each other for each partial state time series. An attribute time series generation step of acquiring the attribute information corresponding to identification information and generating an attribute time series in which the acquired attribute information is associated with each of the states included in the partial state time series;
    A data generation method characterized by comprising:
JP2008164540A 2008-06-24 2008-06-24 Apparatus and method for generating time series data Pending JP2010009112A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2008164540A JP2010009112A (en) 2008-06-24 2008-06-24 Apparatus and method for generating time series data

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP2008164540A JP2010009112A (en) 2008-06-24 2008-06-24 Apparatus and method for generating time series data

Publications (1)

Publication Number Publication Date
JP2010009112A true JP2010009112A (en) 2010-01-14

Family

ID=41589566

Family Applications (1)

Application Number Title Priority Date Filing Date
JP2008164540A Pending JP2010009112A (en) 2008-06-24 2008-06-24 Apparatus and method for generating time series data

Country Status (1)

Country Link
JP (1) JP2010009112A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9299033B2 (en) 2012-06-01 2016-03-29 Fujitsu Limited Method and apparatus for detecting abnormal transition pattern

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2001134554A (en) * 1999-11-04 2001-05-18 Nippon Telegr & Teleph Corp <Ntt> Method and system for providing multimedia information and medium where program thereof is recorded
JP2006318191A (en) * 2005-05-12 2006-11-24 Hitachi Ltd Commodity information provision system
WO2008001550A1 (en) * 2006-06-27 2008-01-03 Murata Kikai Kabushiki Kaisha Audio guidance apparatus, audio guidance method and audio guidance program

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2001134554A (en) * 1999-11-04 2001-05-18 Nippon Telegr & Teleph Corp <Ntt> Method and system for providing multimedia information and medium where program thereof is recorded
JP2006318191A (en) * 2005-05-12 2006-11-24 Hitachi Ltd Commodity information provision system
WO2008001550A1 (en) * 2006-06-27 2008-01-03 Murata Kikai Kabushiki Kaisha Audio guidance apparatus, audio guidance method and audio guidance program

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9299033B2 (en) 2012-06-01 2016-03-29 Fujitsu Limited Method and apparatus for detecting abnormal transition pattern

Similar Documents

Publication Publication Date Title
Seuring et al. From a literature review to a conceptual framework for sustainable supply chain management
Zhou RFID and item-level information visibility
Fulcher et al. Highly comparative time-series analysis: the empirical structure of time series and their methods
RU2323475C2 (en) Method (variants) and device (variants) for automated detection of intentional or incidental disruptions of technological procedure by operator
Duttagupta et al. Anatomy of banking crises in developing and emerging market countries
Bohn Nielsen Cointegration analysis in the presence of outliers
US9418352B2 (en) Image-augmented inventory management and wayfinding
JP4621716B2 (en) Human behavior analysis apparatus, method and program
US20120075450A1 (en) Activity determination as function of transaction log
Swait et al. Context dependence and aggregation in disaggregate choice analysis
JP4368767B2 (en) Abnormal operation detection device and abnormal operation detection method
JPWO2015033577A1 (en) Customer behavior analysis system, customer behavior analysis method, customer behavior analysis program, and shelf system
US7778863B2 (en) System and method for customer behavior movement frequency prediction in a store
Liu et al. Fashion landmark detection in the wild
WO2013090603A1 (en) Systems and methods for managing product location information
JP2009163331A (en) Merchandise sales data processor and computer program
Zunino et al. Commodity predictability analysis with a permutation information theory approach
Lenartowicz et al. Practices and performance of small retail stores in developing economies
JP2010002997A (en) Personal behavior analysis apparatus and personal behavior analysis program
Tuma et al. A survey of the challenges and pifalls of cluster analysis application in market segmentation
JP5687806B1 (en) Color estimation apparatus, color estimation method, and color estimation program
JP2002189597A (en) Customer data analysis method
JP2005309951A (en) Sales promotion support system
JP4753616B2 (en) Product information provision system
JP5553866B2 (en) Product recognition device and recognition dictionary addition program

Legal Events

Date Code Title Description
A621 Written request for application examination

Free format text: JAPANESE INTERMEDIATE CODE: A621

Effective date: 20110614

A977 Report on retrieval

Free format text: JAPANESE INTERMEDIATE CODE: A971007

Effective date: 20121105

A131 Notification of reasons for refusal

Free format text: JAPANESE INTERMEDIATE CODE: A131

Effective date: 20121113

A02 Decision of refusal

Free format text: JAPANESE INTERMEDIATE CODE: A02

Effective date: 20130611