Information Processing Method, Program, Storage Medium, and Information Processing Apparatus

Abstract

To properly analyze an action history of a person.SOLUTION: An information processing method using an information processing apparatus includes: an extraction step (S12) which extracts time-series action histories from start to end of an action, for each identifier identifying an individual, from data including actions of persons; a mapping step (S13) which maps the action histories; a calculation step (S14) which calculates one or both of closeness between the temporally first and last actions in the action histories and the number of stays in the action histories; and a classification step (S15) which classifies the action histories into action transition patterns, on the basis of one or both of the closeness and the number of stays.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present invention relates to an information processing method, a program, Memory medium and an information processing apparatus.

Background Art

[0002] Conventionally, an action history has been extracted from data including human actions and the action history has been analyzed. The analysis results of the action history are used for various purposes such as marketing, recommendation of products or services, or traffic planning.

[0003] As an example of the above-described method for analyzing an action history, for example, Patent Document 1 discloses a method for analyzing the behavior tendency of a user based on the usage history of a non-contact transportation system IC card or a portable terminal having an equivalent function. Specifically, based on the usage history, the usage frequency of stations, the staying time at stations, and the usage tendency as to whether the stores used in the station building or around the station are entertainment stores or convenience stores are calculated, and the users of the stations are characterized and classified by the index values. Then, based on the classification results of the users, information provision such as recommendation targeting the purchasing behavior in the station building or around the station is performed.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] Here, as a result of intensive studies by the present inventors, it has been found that for the analysis of behavior history, classifying a series of behavior histories into predetermined behavior transition patterns is useful for achieving the various purposes described above. However, conventional behavior analysis methods such as the method disclosed in Conventional Document 1 do not consider such classification and there is room for improvement.

[0006] The present invention has been made in view of such points and aims to appropriately analyze a person's behavior history.

Means for Solving the Problems

[0007] To achieve the above object, the present invention Information processing method is an information processing apparatus is an extraction step of extracting, for each identifier for identifying an individual, a temporally continuous behavior history from the start to the end of the behavior from data including a person's behavior; mapping the behavior history vectorized and mapped into a vector space a mapping step; using the mapped behavior history a calculation step of calculating one or both of the proximity between the first behavior and the last behavior in terms of time in the behavior history and the number of stays in the behavior history; and a classification step of classifying the behavior history into a behavior transition pattern based on one or both of the proximity and the number of stays. to execute It is characterized by this. Further, the information processing method of the present invention includes: an extraction step in which an information processing apparatus extracts a temporally continuous behavior history from the start to the end of the behavior for each identifier that identifies an individual from data including human behavior; a mapping step in which the behavior history is mapped into a geographical space; a calculation step in which, using the mapped behavior history, the proximity between the first behavior and the last behavior in the behavior history and / or the number of stays in the behavior history is calculated; and a classification step in which the behavior history is classified into behavior transition patterns based on the proximity and / or the number of stays.

[0008] In the information processing method, in the calculation step, both the proximity and the number of stays may be calculated, and in the classification step, the behavior history may be classified into a behavior transition pattern according to the shape based on the proximity and the number of stays. Note that the shape is formed by mapping the behavior history on a geographical space or a vector space and connecting the mapped behavior histories with lines.

[0010] In the information processing method, in the calculating step, the physical distance between the positions where the first action and the last action are performed may be calculated as the proximity. Alternatively, in the calculating step, the similarity between the first action and the last action may be calculated as the proximity.

[0011] In the information processing method, the calculating step includes a step of calculating the proximity between two actions that are temporally continuous and one or both of the times between the two actions, a step of determining whether the period between the two actions is a stay or a movement based on the proximity between the two actions and one or both of the times between the two actions, and a step of calculating the number of stays. Alternatively, in the calculating step, the number of actions in the action history may be calculated as the number of stays.

[0012] In the information processing method, the data may include identification information of the person who performed the action.

[0013] In the information processing method, the data may include data acquired from a transportation IC card or a mobile terminal.

[0014] According to another aspect of the present invention, there is provided a program for causing an information processing apparatus to execute the information processing method. for A program is provided.

[0015] According to still another aspect of the present invention, there is provided a storage medium storing the program. a computer-readable A storage medium is provided.

[0016] According to still another aspect of the present invention, there is provided an information processing apparatus including: a storage unit that stores a program for controlling the information processing apparatus, and an arithmetic unit that controls the information processing apparatus according to the program an extraction unit that extracts, for each identifier for identifying an individual, an action history that is temporally continuous from the start to the end of the action from data including the actions of a person; and a mapping unit that maps the action history. vectorized and mapped into a vector space a mapping unit that maps the action history. using the mapped behavior historyA calculation unit that calculates one or both of the proximity between the first action and the last action in terms of time and the number of stays in the action history, and a classification unit that classifies the action history into action transition patterns based on one or both of the proximity and the number of stays. It is characterized by having these components. Further, the present invention is an information processing apparatus including: a storage unit that stores a program for controlling the information processing apparatus; an arithmetic unit that controls the information processing apparatus according to the program; an extraction unit that extracts a temporally continuous behavior history from the start to the end of the behavior for each identifier that identifies an individual from data including human behavior; a mapping unit that maps the behavior history into a geographical space; a calculation unit that calculates, using the mapped behavior history, the proximity between the first behavior and the last behavior in the behavior history and / or the number of stays in the behavior history; and a classification unit that classifies the behavior history into behavior transition patterns based on the proximity and / or the number of stays.

[0017] In the information processing apparatus, the calculation unit may calculate both the proximity and the number of stays, and the classification unit may classify the action history into action transition patterns according to the shape based on the proximity and the number of stays. Note that the shape is formed by mapping the action history onto a geographical space or a vector space and connecting the mapped action histories with lines.

[0019] In the information processing apparatus, the calculation unit may calculate the physical distance between the positions where the first action and the last action were performed as the proximity. Alternatively, the calculation unit may calculate the similarity between the first action and the last action as the proximity.

[0020] In the information processing apparatus, the calculation unit may calculate the proximity between two temporally consecutive actions and one or both of the times between the two actions, and based on one or both of the proximity between the two actions and the times between the two actions, determine whether the interval between the two actions is a stay or a movement, and calculate the number of stays. Alternatively, the calculation unit may calculate the number of actions in the action history as the number of stays.

[0021] In the information processing apparatus, the data may include identification information of the person who performed the action.

[0022] In the information processing apparatus, the data may include data acquired from a transportation IC card or a mobile terminal.

Advantages of the Invention

[0023] According to the present invention, a person's behavioral history is classified into behavioral transition patterns, so that the behavioral history can be appropriately analyzed. As a result, the analysis results can be used for a variety of purposes, such as marketing and recommendation of products and services, event planning, and transportation planning. [Brief explanation of the drawings]

[0024]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Figure 17

Figure 18

Figure 19

Figure 20

Figure 21

Figure 22

Figure 23

Figure 24

Figure 25

Figure 26

Figure 27

Figure 28

Figure 29

[0025] Hereinafter, embodiments of the present invention will be described with reference to the drawings. In this specification and the drawings, components having substantially the same functional configuration are denoted by the same reference numerals, and redundant description is omitted.

[0026] <Configuration of Information Processing System 1> FIG. 1 is an explanatory diagram showing an outline of the configuration of the information processing system 1 according to the present embodiment.

[0027] The information processing system 1 includes a database device 10 and an information processing device 20. The database device 10 and the information processing device 20 are configured to be connectable via a network 30. The network 30 is not particularly limited as long as it can perform communication. For example, the Internet, a wired LAN, a wireless LAN, or the like is used. In the present embodiment, only one database device 10 is shown in the drawing for simplicity of explanation, but the information processing system 1 includes a plurality of database devices 10.

[0028] (Configuration of Database Device 10) The database device 10 includes an arithmetic unit 11, a storage unit 12, and a communication unit 13.

[0029] The arithmetic unit 11 is a central processing unit such as a circuit (hardware) or a CPU. The arithmetic unit 11 controls the database device 10 according to a program (software) stored in the storage unit 12.

[0030] The storage unit 12 stores a database of data to be output to the information processing device 20. This data includes data including human behavior (hereinafter sometimes referred to as "behavior history data"). The behavior history data includes, in addition to human behavior, identification information of the person.

[0031] A person's identification information is an identifier that can distinguish one individual from another and includes, for example, a user ID. In addition to the user ID, the identifier may be a combination of other factors such as date and time or location. For example, A on January 1 and A on January 2 may be treated as different individuals and assigned different identifiers. Also, a person's identification information may include attribute information (such as age, gender, etc.) associated with the individual.

[0032] Action history data, in other words, is data (transaction data) that includes actions taken by an individual with intention and is associated with that individual. Also, action history data includes actions represented as points on the time axis, and a series of these temporally continuous actions constitutes the action history. Note that the type of action history data is not particularly limited and includes various types of data.

[0033] The action history data includes, for example, information obtained from a transportation IC card or a portable terminal having an equivalent function (hereinafter collectively referred to as "transportation IC card"). For example, when using a transportation IC card at a ticket gate of a railway station or a transportation facility such as a bus, the usage history is transmitted to the database device 10 and stored in the storage unit 12. Also, when using a transportation IC card at a card reader terminal, a POS register terminal, a vending machine, etc. installed in a station building or a store around the station other than a transportation facility, the usage history is similarly transmitted to the database device 10 and stored in the storage unit 12. When the usage history is stored in this way, the user ID for identifying the transportation IC card and the location information of the place where the transportation IC card was used are also stored in association with the usage history.

[0034] In addition, the action history data includes, for example, the operation history of a website. The operation history of a website includes, for example, the purchase history of goods and the usage history of services on an EC (electronic commerce) website, the browsing history of the website, etc. When these operation histories of the website are stored in the storage unit 12, the user ID on the website is also stored in association with the operation history.

[0035] Furthermore, the action history data may include, for example, purchase data (POS data) in a retail store or the like, usage data of credit cards and electronic money, transaction data in a financial institution (such as deposits, withdrawals, settlements, etc.), dining data in a restaurant, accommodation data in an accommodation facility, playback / play data of music / video services and games, point usage data on a point site, and the like.

[0036] In addition, the storage unit 12 stores various programs for controlling the database device 10.

[0037] The communication unit 13 is a communication interface that mediates communication with the network 30 and performs data communication with the information processing device 20.

[0038] (Configuration of the information processing device 20) The information processing device 20 includes an arithmetic unit 21, a storage unit 22, an extraction unit 23, a mapping unit 24, a calculation unit 25, a classification unit 26, a visualization unit 27, and a communication unit 28.

[0039] The arithmetic unit 21 is a central processing unit such as a circuit (hardware) or a CPU. The arithmetic unit 21 controls the information processing device 20 according to a program (software) stored in the storage unit 22. [[ID=!20]]

[0040] The storage unit 22 stores various data processed by the information processing device 20. For example, the storage unit 22 stores input data input from the database device 10 to the information processing device 20, extraction result data in the extraction unit 23, mapping result data in the mapping unit 24, calculation result data in the calculation unit 25, classification result data in the classification unit 26, and data visualized in the visualization unit 27. In addition, the storage unit 22 stores various programs for controlling the information processing device 20.

[0041] It should be noted that there is an error in the original text where "!20" is used. It should likely be "20" in the "ID" tag. This has been left as is in the translation to maintain consistency with the original.The extraction unit 23 extracts the behavior history from the behavior history data input from the database device 10 to the information processing device 20. As described above, the behavior history is linked to an individual, and the extraction unit 23 extracts the behavior history for each identifier that identifies the individual. Further, the behavior history to be extracted is a temporally continuous behavior history from the start to the end of the behavior. The time from the start to the end of the behavior can be arbitrarily set. For example, the time may be one day, or for example, when a person's behavior is entering and leaving a station, the time from when the station opens to when it closes may be used. Also, for example, when a person's behavior is recorded, the time may be the time from the start record to the end record. Also, for example, when a person's behavior is a purchase history of goods or a usage history of services on an EC site, the time may be the time from logging in to the EC site to logging out. Also, the time may be the time from when a person wakes up to when they go to sleep, or the time from when a person leaves home to when they return home. Note that the raw data input from the database device 10 may contain unnecessary information, and the extraction unit 23 also performs cleansing processing such as removing such unnecessary information.

[0042] The mapping unit 24 maps the behavior history extracted by the extraction unit 23. Specifically, the mapping unit 24 vectorizes the behavior history and maps it to a vector space. In vectorizing the behavior history, among the plurality of parameters included in each behavior, the parameters necessary for the processing are selected to vectorize each behavior. The method of parameter selection is arbitrary. For example, the parameters may be selected by an operator, or the parameters may be selected by machine learning. The vector space may be one that represents the physical real space as a vector, or one that represents a virtual space that does not physically exist as a vector. For example, when the behavior history to be processed is a behavior in a transportation facility or a physical store, etc., the vector space is one that represents the real space as a vector. On the other hand, for example, when the behavior history to be processed is a behavior on a website, etc., the vector space is one that represents the virtual space as a vector.

[0043] The parameter to be vectorized for each action may be a single parameter, or may be a parameter obtained by reducing the dimension of multiple parameters. When reducing the dimension of multiple parameters, the multiple parameters are input into a matrix, and principal components are extracted by, for example, performing principal component analysis to generate a parameter with reduced dimension.

[0044] Furthermore, if the behavior history data can be linked to geographic information, the mapping unit 24 may map the behavior history in geographic space.

[0045] The calculation unit 25 calculates one or both of the temporal proximity between the first and last behaviors in the behavior history (hereinafter, sometimes referred to as "start-end behavior proximity") and the number of stays in the behavior history.

[0046] When calculating the proximity between two actions, for example, if the action history to be processed is an action on a public transport system or in a brick-and-mortar store and can be linked to geographic information, the calculation unit 25 calculates the physical distance between the locations where the two actions were performed as the proximity. The physical distance may be, for example, a geographical straight-line distance or a distance along a geographical route.

[0047] Furthermore, when calculating the closeness between two actions, if the action history to be processed is an action on a website or the like and the action history can be vectorized, the calculation unit 25 calculates the similarity between the two actions as the closeness. The similarity may be any value, and examples thereof include cosine similarity, Euclidean distance, and Manhattan distance.

[0048] When calculating the number of stays in the behavior history, the calculation unit 25 performs, for example, the following steps (1) to (3). (1) Calculate the proximity of two consecutive actions in time and / or the time between the two actions. Hereinafter, two consecutive actions in time will be referred to simply as "two actions," and the proximity of the two actions will be referred to as "consecutive action proximity." (2) Based on one or both of the continuity of actions proximity calculated in (1) above and the time between two actions (hereinafter sometimes referred to as "continuous action time"), it is determined whether the period between two actions is a stay or a movement. (3) Calculate the number of stays determined in (2) above as the number of stays.

[0049] In (1) above, the method for calculating the continuity of actions proximity is the same as the method for calculating the start and end action proximities. That is, the calculation unit 25 may calculate the physical distance between the positions where two actions are performed as the continuity of actions proximity, or alternatively, may calculate the similarity between two actions as the continuity of actions proximity.

[0050] In (2) above, when determining whether the period between two actions is a stay or a movement, the period between the two actions includes the state of a person between two temporally continuous actions. A stay indicates a state where a person remains between two actions in vector space. On the other hand, it indicates a state where a person is moving between two actions in vector space. In the first embodiment, (2) above is performed by the calculation unit 25, but it may also be performed by the classification unit 26.

[0051] When calculating the continuity of actions proximity in (1) above, in (2), the calculation unit 25 compares the calculated continuity of actions proximity with a preset threshold value to determine whether the period between two actions is a stay or a movement. For example, when the calculated continuity of actions proximity is less than the threshold value, the period between two actions is determined to be a stay, and when the calculated continuity of actions proximity is equal to or greater than the threshold value, the period between two actions is determined to be a movement.

[0052] When calculating the continuous action time in (1) above, in (2), the calculation unit 25 compares the calculated time with a preset threshold value to determine whether the period between two actions is a stay or a movement. For example, when the calculated continuous action time is less than the threshold value, the period between two actions is determined to be a stay, and when the calculated continuous action time is equal to or greater than the threshold value, the period between two actions is determined to be a movement.

[0053] When calculating the continuous action proximity and continuous action time in (1) above, in (2) above, the calculation unit 25 first derives a time derivative of the continuous action proximity from the calculated continuous action proximity and continuous action time. If the continuous action proximity is a distance, the time derivative of the continuous action proximity is a velocity, and if the continuous action proximity is an angle, the time derivative of the continuous action proximity is an angular velocity. The calculation unit 25 then compares the derived time derivative of the continuous action proximity with a preset threshold to determine whether the two actions are a stay or a move. For example, if the derived time derivative of the continuous action proximity is less than the threshold, the two actions are determined to be a stay, and if the derived time derivative of the continuous action proximity is equal to or greater than the threshold, the two actions are determined to be a move.

[0054] Furthermore, when calculating the number of stays in the behavior history, the calculation unit 25 uses, for example, the number of actions in the behavior history as the number of stays. That is, if a person's action directly indicates a stay, such as when purchasing a product, the number of actions of this person becomes the number of stays.

[0055] As described above, the number of stays calculated by the calculation unit 25 may be the number of stays between two temporally consecutive actions, the number of actions of a person, etc. In other words, the number of stays is the number of stays of a person in the action history.

[0056] The classification unit 26 classifies the behavior history into behavior transition patterns based on one or both of the start-end behavior proximity and the number of stays calculated by the calculation unit 25. A behavior transition pattern is a classification of a series of behavioral histories from when a person starts a behavior to when the person finishes the behavior into a predetermined pattern.

[0057] When the calculation unit 25 calculates both the start / end behavior proximity and the number of stays, the classification unit 26 classifies the behavior history into behavior transition patterns by shape based on the start / end behavior proximity and the number of stays. Specifically, the classification unit 26 regards the locations where the behavior started, ended, and stayed as "points" and derives the shape by connecting the points with "lines." In other words, the shape is formed by mapping the behavior history onto a geographic space or vector space and connecting the mapped behavior histories with lines.

[0058] The visualization unit 27 visualizes the action transition pattern classified by the classification unit 26.

[0059] The communication unit 28 mediates communication with the network 30 and performs data communication with the database device 10.

[0060] In the information processing apparatus 20 of the present embodiment, the above program is stored in the storage unit 22. However, for example, it may be stored in a computer-readable storage medium such as a computer-readable hard disk drive (HDD), solid state drive (SSD), flexible disk (FD), compact disk (CD), magneto-optical disk (MO), or various memories. Further, the above program can be stored in the above storage medium or the like by being downloaded via a communication line network such as the Internet.

[0061] <Information Processing Method> Next, an information processing method performed using the information processing system 1 configured as described above will be described. Hereinafter, the information processing method of the present invention will be described using an embodiment including four specific cases.

[0062] (First Embodiment) An information processing method according to the first embodiment will be described. FIG. 2 is a flowchart showing the main steps of the information processing method according to the first embodiment. In the first embodiment, a case where the action history data to be processed is data including the usage history when a transportation IC card is used at a ticket gate of a station will be described.

[0063] [S11: Storage Step] First, when a transportation IC card is used at a ticket gate of a station, the usage history is transmitted to the database device 10 and stored in the storage unit 12. The action history data stored in the storage unit 12 is input to the information processing apparatus 20 via the network 30 and stored in the storage unit 22.

[0064] Figure 3 is a table showing an example of action history data. The action history data includes, for example, user ID, date and time, station, and entry / exit, and these are arranged in chronological order. Also, the storage unit 22 stores station master data shown in Figure 4. The station master data includes geographical information of the station, that is, latitude and longitude.

[0065] [S12: Extraction step] Next, in the extraction unit 23, the action history is extracted from the action history data stored in the storage unit 12 in step S11. In step S12, as shown in Figure 5, the action history is extracted for each identifier. The identifier in the first embodiment is a combination of user ID and date. For example, although user ID 0004 has performed actions on April 2 and April 3, 2020, the identifiers are respectively "0004_20200402" and "0004_20200403", and different identifiers are assigned. Also, in step S12, the action history to be extracted is an action history that is temporally continuous from the start to the end of the action. For example, for the identifier "0001_20200401", 10:54 is the start of the action and 14:33 is the end of the action.

[0066] [S13: Mapping step] Next, in the mapping unit 24, the action history extracted in step S12 is mapped to the geographical space. In step S13, as shown in Figure 6, for the stations in the action history of Figure 5, the geographical information (latitude and longitude) of the stations in the station master data of Figure 4 is associated. In such a case, since geographical information is associated with the action history, the action history is mapped to the geographical space.

[0067] Note that the order of step S12 and step S13 may be reversed. That is, geographical information may be associated with the action history, and after mapping the action history to the geographical space, an identifier may be assigned.

[0068] [S14: Calculation step] Next, in the calculation unit 25, using the action history mapped in step S13, the proximity between the first action and the last action in the action history is calculated. In step S14, first, as shown in FIG. 7, for each identifier, a vector having two consecutive actions as elements is generated. Hereinafter, in two consecutive actions, the previous action is referred to as the "previous action", and the subsequent action is referred to as the "subsequent action". The elements of the previous action and the subsequent action include time, station, entry / exit, and geographical information of the station. Next, as the proximity between two actions, i.e., the consecutive action proximity, for each identifier, using the geographical information (latitude and longitude) of the station of the previous action and the geographical information of the station of the subsequent action, the inter-station distance between these previous and subsequent actions is calculated. In the first embodiment, the straight-line distance between the stations of the previous and subsequent actions was calculated, but the distance along the geographical route may be calculated. Next, for each identifier, the sum of the above distances is calculated. Then, as shown in FIG. 8, for each identifier, as the proximity between the first action and the last action, i.e., the start-end action proximity, the distance between the station of the first action (the first station) and the station of the last action (the last station) is calculated.

[0069] [S15: Classification Step] Next, in the classification unit 26, based on the start-end action proximity calculated in step S14, the action history is classified into action transition patterns. In step S15, as shown in FIG. 9, based on the start-end action proximity, it is determined whether the first station and the last station are the same.

[0070] If the first station and the last station are different stations, the action transition pattern of the identifier is determined to be a "one-way pattern". As shown in FIG. 10(a), the one-way pattern is a pattern that moves from the first station to the last station and does not return to the first station.

[0071] On the other hand, if the first station and the last station are the same station, the action transition pattern of the identifier is determined to be a "round-trip pattern" or a "circular tour pattern". As shown in FIG. 10(b), the round-trip pattern is a pattern that moves from the first station (departure place) to the last station (destination) and then returns to the first station. As shown in FIG. 10(c), the circular tour pattern is a pattern that moves from the first station to the last station via a plurality of stations (destinations), and more specifically, enters and exits at a plurality of stations.

[0072] According to the first embodiment described above, since the behavioral history is classified into behavioral transition patterns in step S15, it is possible to appropriately understand the actual usage status of each station and the flow of people. For example, in passenger travel, by distinguishing between use of transportation that returns to the departure point and use of transportation for the purpose of travel (not returning to the departure point), it is possible to assume the purpose of passenger travel and tally the number of passengers. Furthermore, since people's behavioral history can be appropriately analyzed in this way, the analysis results can be used for a variety of purposes, such as marketing and recommendations of products and services, event planning, and transportation planning.

[0073] Here, OD (Origin: departure point, Destination: destination) data has conventionally been used to analyze the flow of people. However, analysis using data from leaving a station to entering the station in addition to data from entering the station to exiting the station (the OD data) as in the first embodiment has not conventionally been performed. According to the first embodiment, by classifying the behavioral history into behavioral transition patterns in this way, it becomes possible to analyze people's behavioral history more appropriately than before.

[0074] Furthermore, as described in Patent Document 1, for example, it has been customary to analyze usage trends within and around stations based on the usage history of transportation IC cards. In contrast, in the first embodiment, data from the time a person leaves a station until the time they enter is also used to analyze behavioral transition patterns, making it possible to understand stays and movements in a wider area than the area around the station, including the area between stations. From this perspective, the first embodiment makes it possible to analyze people's behavioral history more appropriately than ever before.

[0075] Also, in the first embodiment, the action history is classified into action transition patterns for each identifier. By aggregating and analyzing the action history in units of identifiers in this way, it is possible to suppress excessive compression of the large amount of information recorded on the transportation IC card. That is, it becomes easier to aggregate the action history, and it is possible to suppress the loss of detailed information. Also, it is possible to suppress excessive expansion of a large amount of information. That is, it is possible to grasp detailed information, and it is possible to suppress the inability to quantitatively aggregate it. In other words, there is significance in classifying the action history into action transition patterns in the first embodiment in that it is possible to balance the compression and expansion of information by aggregating and analyzing the action history in units of identifiers.

[0076] Also, the identifier in the first embodiment is a combination of the user ID and the date, and the action history is classified into action transition patterns every day. By separating time in meaningful sections in this way, it is possible to appropriately analyze a person's action history. Note that in the first embodiment, the action history is classified every day, but the time unit is not limited to this. For example, an identifier may be assigned in units of one trip and the action history may be classified into action transition patterns.

[0077] Furthermore, in the first embodiment, since the action history data including actions taken by an individual with will is analyzed, there is significance in classifying the action history into action transition patterns as a person's action. For example, even if the action transition pattern is classified by analyzing data automatically acquired like the log data acquired using GPS, it is not possible to appropriately analyze a person's action.

[0078] (Second Embodiment) An information processing method according to the second embodiment will be described. FIG. 11 is a flowchart showing the main steps of the information processing method according to the second embodiment. In the second embodiment, as in the first embodiment, a case will be described in which the action history data to be processed is data including the usage history when a transportation IC card is used at a ticket gate of a station. Also, in the second embodiment, the same action history data as in the first embodiment is used.

[0079] [S21: Memory process, S22: Extraction process, S23: Mapping process] The processes S21 to S23 in the second embodiment are the same as the processes S11 to S13 in the first embodiment respectively. That is, first, in process S21, the action history data shown in FIG. 3 and the station master data shown in FIG. 4 are stored in the storage unit 22 of the information processing apparatus 20. Next, in process S22, as shown in FIG. 5, the action history is extracted for each identifier from the action history data. Next, in process S23, as shown in FIG. 6, the geographical information of the stations in the station master data of FIG. 4 is associated with the stations in the action history of FIG. 5, and the action history is mapped to the geographical space.

[0080] [S24: Calculation process] Next, in the calculation unit 25, the number of stays in the action history is calculated using the action history mapped in process S23. In process S24, the number of stays is calculated by performing the following three processes: process S24-1 to process S24-3.

[0081] In process S24-1, first, similar to process S24 in the first embodiment, as shown in FIG. 12, a vector having two consecutive actions as elements is generated for each identifier. Next, as the proximity of consecutive actions, which is the proximity of two actions, for each identifier, the distance between the stations of the previous and subsequent actions is calculated using the geographical information of the stations of the previous and subsequent actions. The calculation of this proximity of consecutive actions is the same as in the first embodiment. In the second embodiment, further, using the time of the previous action and the time of the subsequent action, the time between these two actions is calculated as the time of consecutive actions between the previous and subsequent actions.

[0082] Next, in step S24-2, for each identifier, based on the proximity of consecutive actions and the consecutive action time, it is determined whether the period between two actions is a stay or a movement. Here, the threshold for the proximity of consecutive actions (distance between stations) for determining stay or movement is set to, for example, 3000 m, and the threshold for the consecutive action time is set to 30 minutes. If the proximity of consecutive actions is less than 3000 m and the consecutive action time is less than 30 minutes, it is determined as "stay", and if the proximity of consecutive actions is 3000 m or more or the consecutive action time is 30 minutes or more, it is determined as "movement". In step S24-1, the distance between stations may be divided by the time to derive the speed, and the derived speed may be compared with a predetermined threshold to determine whether the period between two actions is a stay or a movement.

[0083] Next, in step S24-3, for each identifier, the number of stays is counted and calculated as the stay count.

[0084] In step S24 of the second embodiment, both the proximity of consecutive actions and the consecutive action time are used to determine whether the period between two actions is a stay or a movement, but either one may be used. For example, when using the proximity of consecutive actions, the threshold for the proximity of consecutive actions is set to 3000 m, and if the proximity of consecutive actions is less than 3000 m, it is determined as "stay", and if it is 3000 m or more, it is determined as "movement". Also, for example, when using the consecutive action time, if the consecutive action time is less than 30 minutes, it is determined as "stay", and if it is 30 minutes or more, it is determined as "movement".

[0085] Also, the determination of stay and movement in step S24 of the second embodiment was performed by the calculation unit 25, but it may be performed by the classification unit 26.

[0086] [S25: Classification step] Next, in the classification unit 26, based on the stay count calculated in step S24, for each identifier, the action history is classified into action transition patterns. In step S25, as shown in FIG. 13, if the stay count is 0 times, it is classified into the action transition pattern of "no stay". If the stay count is 1 time, it is classified into the action transition pattern of "staying at one place". If the stay count is 2 times or more, it is classified into the action transition pattern of "staying at multiple places".

[0087] [S26: Visualization step] Next, the behavioral transition patterns classified in step S25 are visualized in the visualization unit 27. In step S26, the number of passengers for each length of stay in the table shown in Fig. 13 is tallied by identifier. Then, the tallied result of the number of passengers for each length of stay is visualized in a graph as shown in Fig. 14.

[0088] The second embodiment described above can also achieve the same effects as the first embodiment described above. That is, since the behavioral history is classified into behavioral transition patterns in step S25, it is possible to appropriately grasp the actual usage of each station and the flow of people. Specifically, by tallying the number of destinations where passengers stay, it is possible to grasp the purpose of passenger travel and how they stop at multiple stations. Furthermore, since the behavioral transition patterns are visualized in step S26, it becomes even easier to grasp the actual usage of each station and the flow of people.

[0089] (Third embodiment) An information processing method according to the third embodiment will be described. Fig. 15 is a flow diagram showing the main steps of the information processing method according to the third embodiment. In the third embodiment, as in the first and second embodiments, a case will be described in which the behavior history data to be processed is data including a usage history when a transportation IC card is used at a ticket gate in a station. Furthermore, in the third embodiment, the same behavior history data as in the first and second embodiments is used.

[0090] [S31: storage step, S32: extraction step, S33: mapping step] Steps S31 to S33 in the third embodiment are the same as steps S11 to S13 in the first embodiment. That is, first, in step S31, the behavior history data shown in FIG. 3 and the station master data shown in FIG. 4 are stored in the storage unit 22 of the information processing device 20. Next, in step S32, behavior history is extracted for each identifier from the behavior history data as shown in FIG. 5. Next, in step S33, as shown in FIG. 6, the stations in the behavior history in FIG. 5 are linked with the geographic information of the stations in the station master data in FIG. 4, and the behavior history is mapped to geographic space.

[0091] [S34: Derivation process] Next, the calculation unit 25 derives the stay station with the longest elapsed time for each identifier using the behavior history mapped in step S33. In step S34, first, as shown in FIG. 16, the relationship between two consecutive behaviors is determined as a stay or a move for each identifier. This determination of stay or move is the same as in the second embodiment shown in FIG. 12, and the relationship between the two behaviors is determined as a stay or a move based on the proximity of the consecutive behaviors and the consecutive behavior time. Next, when the relationship between the two behaviors is a stay, it is determined whether the consecutive behavior time (elapsed time) is the longest. Then, the stay with the longest elapsed time is extracted, and the corresponding stay station is derived. In the example of FIG. 16, the station for which the stay with the longest elapsed time is "true" is the stay station.

[0092] Although the calculation unit 25 derives the station where the maximum elapsed time has passed in step S34 of the third embodiment, the classification unit 26 may also derive it.

[0093] [S35: Visualization process] Next, the visualization unit 27 visualizes the number of major passengers staying at each station based on the station with the longest elapsed time calculated in step S34. In step S35, the number of passengers staying at each station is calculated for each identifier, assuming that the major stations during the day are the destinations. The results of the calculation of the number of passengers staying at each station are visualized in a graph, as shown in FIG. 17.

[0094] According to the third embodiment described above, in step S34, the main destinations of passengers' travels can be identified by deriving the station where the longest elapsed time has passed. As a result, the actual usage status of each station and the flow of people can be appropriately identified.

[0095] (Fourth embodiment) An information processing method according to the fourth embodiment will be described. Fig. 18 is a flow diagram showing main steps of the information processing method according to the fourth embodiment. In the fourth embodiment, a case will be described in which the behavior history data to be processed is data including a product purchase history on an EC site. Note that the same applies even if the behavior history data is a product purchase history at a shopping center rather than a product purchase history on an EC site.

[0096] [S41: Memory process] First, when a product is purchased on an EC site, the purchase history is transmitted to the database device 10 and stored in the memory unit 12. The behavior history data stored in the memory unit 12 is input to the information processing device 20 via the network 30 and stored in the memory unit 22.

[0097] FIG. 19 is a table showing an example of behavior history data. The behavior history data includes, for example, a user ID, purchase date and time, a purchase store, and a purchased item, which are arranged in chronological order. The storage unit 22 also stores store master data shown in FIG. 20 and item master data shown in FIG. 21. Store topics in the store master data are created by extracting topics from the behavior history data using a topic model. Similarly, item topics in the item master data are created by extracting topics from the behavior history data using a topic model.

[0098] [S42: Extraction process] Next, the extraction unit 23 extracts behavioral history from the behavioral history data stored in the memory unit 12 in step S41. In step S42, first, behavioral history is extracted for each user ID as shown in FIG. 22, and then behavioral history is extracted for each identifier. The identifier in the fourth embodiment is a combination of the user ID and the date. For example, user ID 0001 performed purchasing behavior on December 1st and December 23rd, 2020, but is assigned different identifiers, "0001" and "0002," respectively. Furthermore, in step S42, the behavioral history extracted is a temporally continuous behavioral history from the start to the end of the behavior.

[0099] [S43: Mapping process] Next, the behavior history extracted in step S42 is vectorized and mapped to a vector space in the mapping unit 24. In step S43, as shown in Fig. 23, the behavior history data is linked to the store and item master data to generate a vector whose elements are the store and item where the purchase was made.

[0100] [S44: Calculation process] Next, the calculation unit 25 calculates the start-end action closeness, which is the temporal closeness between the first action and the last action in the action history, and the number of stays in the action history, using the action history vectorized in step S43. In step S44, the start-end action closeness is calculated in step S44-1, and the number of stays is calculated in step S44-2.

[0101] In step S44-1, first, the first and last actions in a series of action histories are extracted for each identifier, as shown in Fig. 24. Next, as shown in Fig. 25, for each identifier, the cosine similarity between the vectors of the previous and next actions is calculated as the start-end action closeness, which is the closeness between the first and last actions, using a vector consisting of the store topic and item topic of the previous action and a vector consisting of the store topic and item topic of the next action. Note that, although cosine similarity is calculated in the fourth embodiment, Euclidean distance or Manhattan distance may also be calculated.

[0102] In step S44-2, the number of stays is calculated as the number of actions (number of vectors) in the action history for each identifier as shown in Fig. 26. Note that either step S44-1 or S44-2 may be performed first.

[0103] [S45:Classification process] Next, in the classification unit 26, based on the start / end behavior proximity and the number of stays calculated in step S44, the behavior history is classified into behavior transition patterns according to the shape. In step S45, according to the rules shown in FIG. 27, the behavior history is classified into five shapes. That is, the number of vectors, which is the number of stays, is regarded as points, and the points are connected by lines to derive the shape. Therefore, when the number of vectors is 1, the shape is "point", when the number of vectors is 2, the shape is "line", and when the number of vectors is 3 or more, the shape is "polygon". At this time, the threshold of the cosine similarity for determining the size of the shape is set to 0.707, for example. Then, according to this rule, as shown in FIG. 28, for each identifier, the behavior history is classified into behavior transition patterns according to the shape.

[0104] [S46: Visualization step] Next, in the visualization unit 27, the behavior transition pattern classified in step S45 is visualized. In step S27, as shown in FIG. 29, the shape representing the behavior transition pattern is visualized.

[0105] Also in the above-described fourth embodiment, the same effects as those of the first and second embodiments described above can be obtained. That is, since the behavior history is classified into behavior transition patterns in step S45, the actual usage situation of each station and the flow of people can be appropriately grasped, and as a result, the analysis results can be used for various purposes such as marketing and recommendation of products and services.

[0106] In the fourth embodiment, the behavior history is classified into behavior transition patterns according to the shape. This shape mainly has two elements. The first element is the start / end behavior proximity, for example, whether one end of a polygon is closed or not in the case of a polygon. This makes it possible to grasp whether it is a one-way pattern or a round-trip / touring pattern. The second element is the number of stays, for example, the points in a polygon. This makes it possible to grasp whether it is a round-trip pattern or a touring pattern.

[0107] In addition, by classifying the action history according to the shape in this way, it is possible to grasp the correlation between the variation and the number of actions. For example, when the purpose is to increase the variation of actions, products or services may be recommended so that the number of points of the polygon increases. Further, if the relationship between the points and the shape in the polygon is understood, it can be utilized for the recommendation of products or services and the design of the layout of stores. For example, when the compatibility between Product A and Product B is good, Product B can be recommended to the purchasers of Product A. Alternatively, when the compatibility between Store C and Store D is good, it is also possible to arrange Store C and Store D in proximity to each other.

[0108] As described above, the preferred embodiments of the present invention have been explained with reference to the accompanying drawings, but the present invention is not limited to such examples. It is obvious that those skilled in the art can conceive of various modification examples or correction examples within the scope of the idea described in the claims, and it is naturally understood that those also belong to the technical scope of the present invention.

Industrial Applicability

[0109] The present invention is useful when analyzing the action history of a person.

Explanation of Signs

[0110] 1 Information processing system 10 Database device 11 Arithmetic unit 12 Storage unit 13 Communication unit 20 Information processing device 21 Arithmetic unit 22 Storage unit 23 Extraction unit 24 Mapping unit 25 Calculation unit 26 Classification unit 27 Visualization unit 28 Communication unit 30 Network

Claims

1. An information processing apparatus, from data including human behavior, for each identifier that identifies an individual, an extraction step of extracting a temporally continuous behavior history from the start to the end of the behavior; a mapping step of vectorizing the behavior history and mapping it to a vector space; a calculation step of calculating, using the mapped behavior history, one or both of the proximity between the earliest and the latest behaviors in the behavior history and the number of stays in the behavior history; a classification step of classifying the behavior history into behavior transition patterns based on one or both of the proximity and the number of stays, characterized in that it executes an information processing method.

2. An information processing apparatus, from data including human behavior, for each identifier that identifies an individual, an extraction step of extracting a temporally continuous behavior history from the start to the end of the behavior; a mapping step of mapping the behavior history to a geospatial space; a calculation step of calculating, using the mapped behavior history, one or both of the proximity between the earliest and the latest behaviors in the behavior history and the number of stays in the behavior history; a classification step of classifying the behavior history into behavior transition patterns based on one or both of the proximity and the number of stays, characterized in that it executes an information processing method.

3. In the calculation step, both the proximity and the number of stays are calculated, In the classification step, based on the proximity and the number of stays, the behavior history is classified into behavior transition patterns according to the shape, the information processing method according to claim 1 or 2.

4. In the calculation step, the physical distance between the positions where the earliest and the latest behaviors are performed is calculated as the proximity, the information processing method according to any one of claims 1 to 3.

5. In the calculation step, the similarity between the earliest and the latest behaviors is calculated as the proximity, the information processing method according to any one of claims 1 to 3.

6. The calculation step is a step of calculating the proximity between two temporally continuous behaviors and one or both of the times between the two behaviors; a step of determining whether the two behaviors are a stay or a movement based on one or both of the proximity between the two behaviors and the time between the two behaviors; a step of calculating the number of stays, the information processing method according to any one of claims 1 to 5.

7. The information processing method according to any one of claims 1 to 5, characterized in that, in the calculation step, the number of the actions in the action history is calculated as the number of stays.

8. The information processing method according to any one of claims 1 to 7, characterized in that the data includes identification information of the person who performed the action.

9. The information processing method according to any one of claims 1 to 8, characterized in that the data includes data acquired from a transportation IC card or a mobile terminal.

10. A program for causing an information processing apparatus to execute the information processing method according to any one of claims 1 to 9.

11. A computer-readable storage medium storing the program according to claim 10.

12. An information processing apparatus, a storage unit that stores a program for controlling the information processing apparatus; an arithmetic unit that controls the information processing apparatus according to the program; an extraction unit that extracts a temporally continuous action history from the start to the end of the action for each identifier that identifies an individual from data including human actions; a mapping unit that vectorizes the action history and maps it to a vector space; a calculation unit that calculates one or both of the proximity between the first action and the last action in the action history and the number of stays in the action history using the mapped action history; An information processing apparatus, comprising: a classification unit that classifies the action history into action transition patterns based on one or both of the proximity and the number of stays.

13. An information processing apparatus, a storage unit that stores a program for controlling the information processing apparatus; an arithmetic unit that controls the information processing apparatus according to the program; an extraction unit that extracts a temporally continuous action history from the start to the end of the action for each identifier that identifies an individual from data including human actions; a mapping unit that maps the action history to a geographical space; a calculation unit that calculates one or both of the proximity between the first action and the last action in the action history and the number of stays in the action history using the mapped action history; An information processing apparatus, comprising: a classification unit that classifies the action history into action transition patterns based on one or both of the proximity and the number of stays.

14. The calculation unit calculates both the proximity and the number of stays. The information processing apparatus according to claim 12 or 13, wherein the classification unit classifies the behavior history into behavior transition patterns by shape based on the proximity and the number of stays.

15. The information processing apparatus according to any one of claims 12 to 14, wherein the calculation unit calculates the physical distance between the positions where the first behavior and the last behavior are performed as the proximity.

16. The information processing apparatus according to any one of claims 12 to 14, wherein the calculation unit calculates the similarity between the first behavior and the last behavior as the proximity.

17. The calculation unit calculates the proximity between two behaviors that are temporally continuous and one or both of the times between the two behaviors, determines whether the period between the two behaviors is a stay or a movement based on the proximity between the two behaviors and one or both of the times between the two behaviors, and calculates the number of stays, the information processing apparatus according to any one of claims 12 to 16.

18. The information processing apparatus according to any one of claims 12 to 16, wherein the calculation unit calculates the number of the behaviors in the behavior history as the number of stays.

19. The information processing apparatus according to any one of claims 12 to 18, wherein the data includes identification information of the person who performed the behavior.

20. The information processing apparatus according to any one of claims 12 to 19, wherein the data includes data acquired from a transportation IC card or a mobile terminal.

Images