JP6408248B2 - Data analysis method and data analysis system using data visualization - Google Patents

Description

  The present invention relates to a data analysis method and a data analysis system, and more particularly, to a data analysis method and a data analysis system for performing data analysis using data visualization.

  Attempts have been made to visualize and utilize the data. For example, Patent Document 1 discloses that a population distribution map is created based on user position information received from a GPS-compatible mobile phone and used in the event of a disaster or the like.

JP 2003-044969 A

  By visualizing simple numerical data on a map or the like, it becomes easier to understand what the data means. However, as the amount of data to be processed increases, for example when handling big data, complicated processing is required for data processing and visualization, so a computer with high processing power and a long time are required. Cost. Therefore, it took a lot of resources and time to capture and analyze the meaning of the data.

  The present invention has been made in view of the above points, and an object of the present invention is to provide a data analysis method and a data analysis system that enable efficient data analysis using data visualization.

The data analysis method of the present invention is a data analysis method using data visualization using a data analysis system including a calculation unit, a storage, a memory, a display, and a user interface, and the calculation unit stores the data in the storage. An extraction step for extracting a part of the data set, and a first condition including at least one of a method and a parameter used for the method, which is set through the user interface, by the calculation unit. a step of storing in the arithmetic unit, a portion of the extracted the data set was analyzed with the first condition, a first visualization step Ru display the analysis result on the display, the arithmetic unit , on the basis of the input via the user interface, the first condition and different in A step of the cell stored in a memory is changed to a second condition that said operation portion is a portion of the data set was analyzed in the second condition, the analysis results Ru is again displayed on the display and 2 visualize process, the calculation unit obtains the whole of the data set from the storage, and analyzed in the second condition stored in the memory, third visual which Ru display the analysis result on the display A rise step, and the data of the data set includes position information of a plurality of subjects, and any one of the first visualize step, the second visualize step, or the third visualize step, The calculation unit has a plurality of movement directions and movement amounts corresponding to the movement direction and the movement amount from the first position at each time T1 to the second position at another time T2. An anonymous vector calculation step for calculating an anonymous vector at a predetermined point on a plane on which a visualized result is displayed based on a body movement vector, and the calculation unit outputs to the display based on the anonymous vector. characterized by chromatic and display step of performing display, the.

The data analysis system of the present invention includes a calculation unit, a storage, a memory, a display, and a user interface, and the calculation unit extracts a part of a data set stored in the storage and is set via the user interface. and, a first condition comprising at least one of the parameters used to the technique and the technique to be stored in the memory, a portion of the extracted the data set was analyzed with the first condition, the analysis results display is displayed, based on an input through the user interface, said to be stored in the memory is changed to a different second condition from the first condition, a portion of the data set, the second were analyzed under the conditions, to display the analysis result again on the display, the whole of the data set from the storage Acquired, the memory and analyzed by the stored second condition, results of the analysis a data analyze system using data visualization to Ru is displayed on the display, data of the data set of the plurality of entities A plurality of subject movement vectors, including position information, wherein the calculation unit corresponds to the direction and amount of movement of the plurality of subjects from the first position at the time T1 to the second position at the other time T2. based on, in a given point of the plane to display the result of the visualizing, calculates anonymous vector, characterized Rukoto to display to the display on the basis of the anonymous vector.

  With such a configuration, it is possible to visualize a part of the data set, analyze and verify the result, and change the visualization condition to visualize again when a characteristic event cannot be found. . If a characteristic event can be found from the result, the entire data set can be visualized under the same conditions as a part of the data set. Since a part of the data set can be visualized in a shorter time than when the entire data set is processed, it is possible to determine a visualization condition for finding a characteristic event from the data set in a short period of time. Furthermore, if the data set is analyzed / verified and analyzed thereafter, trial and error for finding social issues from the big data can be performed in a short period of time. That is, it becomes easy to use data socially.

  According to the present invention, it is possible to provide a data analysis method and a data analysis system that can efficiently perform data analysis using data visualization.

It is explanatory drawing which shows the whole image of the data analysis which concerns on this Embodiment. It is explanatory drawing for demonstrating the data analysis which concerns on this Embodiment. It is a block diagram which shows the hardware constitutions of the data analysis system which concerns on this Embodiment. It is a block diagram which shows the function of the data analysis system which concerns on this Embodiment. It is a flowchart figure which shows the data analysis method which concerns on this Embodiment. It is explanatory drawing for demonstrating the gate analysis in the data analysis system which concerns on this Embodiment. It is explanatory drawing for demonstrating the heat map analysis in the data analysis system which concerns on this Embodiment. It is explanatory drawing for demonstrating the heat map analysis in the data analysis system which concerns on this Embodiment. It is a flowchart figure which shows the anonymous flow visualization method in the data analysis system which concerns on this Embodiment. It is explanatory drawing for demonstrating the anonymous flow visualization method in the data analysis system which concerns on this Embodiment. It is explanatory drawing for demonstrating the calculation method of the personal movement vector field in the anonymous flow visualization method in the data analysis system which concerns on this Embodiment.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

(Data analysis using data visualization)
By visualizing the big data, an expert (for example, an expert in various fields such as city planning, administration, security, and marketing) can see the whole big data.

  However, it is extremely difficult to read and understand a huge amount of big data and grasp the remarkable events and changes. Experts are required to find and solve social issues by analyzing big data.

  FIG. 1 is an explanatory diagram showing an overall image of data analysis according to the present embodiment. As shown in FIG. 1, the big data A is visualized, a rough tendency is grasped from the result (S1), and abnormality or change is detected (S2). In other words, in order to understand which points in the big data A should be analyzed, it is possible to roughly grasp the “parts that are likely to happen” from the visualization results, To find characteristic events. In the visualization, it is also effective to combine the big data A with other data B for visualization.

  Next, if a characteristic event is found by visualization, it is analyzed and verified (S3). At this time, the big data A is viewed in time series, another data C is added, or the extraction condition is changed to verify characteristic events. By proceeding with such a deeper analysis, the characteristic events and their causes are identified, and key factors that lead to the solution of the problem are extracted. Then, paying attention to the key factor, information necessary for an expert (hereinafter referred to as an analysis result) is generated (S4). Using the analysis result, the expert can identify a social problem, propose a specific solution for solving it, and aim for social utilization (S5).

  In the data analysis using the data visualization as described above, it is effective to find a condition for performing the visualization effective for the analysis while repeating the visualization and the analysis many times.

  Data analysis using such data visualization will be described with a specific example.

(Tokyo Marathon)
For example, there is location information of a mobile phone as big data. The case where social utilization is aimed at from the flow of people on the day of the Tokyo Marathon using this will be explained. First, the location information of mobile phones in the Tokyo 23 wards on the day of the Tokyo Marathon is extracted from the location information of mobile phones. Next, on the map of Tokyo 23 wards, the position of the mobile phone, that is, the position of the person holding the mobile phone is displayed as a light spot in time series, and the flow of the person is visualized. At this stage, even if the expert analyzes and verifies the result of the visualization, it is only possible to see a huge flow of light spots, and it is difficult to find some characteristic movement.

  Therefore, if only the flow of people moving at a specific speed (for example, 6 to 20 km / h) is extracted and visualized again, the flow of people moving at that speed can be grasped. Furthermore, by extracting the people who were at the starting point at the starting time, it becomes possible to visualize the flow of runners who participated in the Tokyo Marathon.

  In addition to such changes in extraction conditions, for example, a marathon course is displayed on the map, SNS post data including the keyword “marathon” is superimposed, other data is added, It is also effective to perform visualization.

  In this way, visualization and analysis can be repeated while changing the conditions for data visualization, and a visualization result can be obtained that allows an expert to find characteristic human movements on the day of the Tokyo Marathon. As a result of the analysis, the final visualization result is used as it is, or a graph or the like that is processed again from the big data is generated. Can be used to make decisions.

(Transportation status)
Next, the case where the flow of people using transportation around a certain station is used socially will be described. FIG. 2 is an explanatory diagram for explaining data analysis according to the present embodiment. FIG. 2A shows the result of visualizing the flow of people around a certain station. A plurality of white circles 21 in FIG. 2A indicate the positions of people at a certain point around the ground station 22. Such a person's position can be acquired based on position information acquired from a mobile phone equipped with a GPS function, for example. By moving the white circle 21 indicating the position of the person as time passes, it is possible to express the position of the person by animation. In FIG. 2A, since the positions of all persons are displayed, it is difficult to find some value even if the white circle 21 is moved, and the time required for processing is enormous.

  FIG. 2B shows the result of visualizing the data after performing the gate analysis. Details of the gate analysis will be described later. In the gate analysis, gates are installed at the ground station 22, the underground station 24, the bus stop 25, and the taxi stand 26, which are lift points of other transportation facilities. Only a position of a person who has passed at least two of these gates within a predetermined time is displayed as a black circle 23. A person who does not pass through any gate and a person who passes through any one gate are not displayed. Furthermore, the movements of the train 27, bus 28 and taxi 29 are additionally displayed. For example, the movement of the train 27, the bus 28, and the taxi 29 can be obtained from a company that operates each of them.

  By performing gate analysis in this way, adding lift points 24 to 26 of other transportation facilities, adding movements of train 27, bus 28 and taxi 29, it is possible to find characteristic human movements. It becomes possible. From this visualization result, for example, it is possible to achieve social use such as determining the installation location of the vending machine, or adding or removing the bus stop 25 or the taxi stand 26.

(Data analysis system)
Hereinafter, the data analysis system according to the present embodiment will be described with reference to the drawings.

(Hardware configuration)
FIG. 3 is a block diagram showing a hardware configuration of the data analysis system according to the present embodiment. As shown in FIG. 3, the data analysis system 30 according to the present embodiment includes a calculation unit 31, a power supply unit 32, a memory 33, a storage 34, a user interface 35, and an external interface 36. Details of each part will be easily understood by those skilled in the art from the names.

(function)
FIG. 4 is a block diagram showing functions of the data analysis system according to the present embodiment. As shown in FIG. 4, the data set 41 stored in the storage 34 is input to the calculation unit 31. The calculation unit 31 executes a computer program to extract a part of the data set 41, a visualization part 43 that visualizes a part of the data set 41 or the whole data set 41, a user A condition setting unit 44 for setting the visualization condition in the visualization unit 43 via the interface 35, and further, the data set 41 is analyzed based on the visualization result, and the analysis result necessary for social use is obtained. And an analyzing unit 46 to be generated. Further, the visualization condition 45 set by the condition setting unit 44 can be stored in the memory 33. Further, the processing result in the visualizer 43 can be displayed on the display 47, for example, via the external interface 36.

  Although the case where processing is performed by one device has been described here, it goes without saying that a plurality of devices may process the data set 41 in parallel on the cloud.

(Overview)
Hereinafter, an outline of the data analysis method according to the present embodiment will be described. FIG. 5 is a flowchart showing a data analysis method according to the present embodiment. As shown in FIG. 5, first, the extraction unit 42 extracts a part from the entire data set 41, for example, at random (S11). This extraction is performed to reduce the processing target and the processing load in order to shorten the time required for visualization, which will be described later. Next, the condition setting unit 44 sets the visualization condition 45 (S12). The set visualization condition 45 is stored in the memory 33.

  Next, the visualizer 43 executes the visualize process using the visualize condition 45 set in S2 (S13). The processing result is displayed on the display 47 by the visualizer 43, for example.

  Based on the result of the visualization, the above analysis / verification is performed (S14). In the analysis / verification, a general tendency of a part of the data set 41 is grasped from the visualization result, and it is verified whether abnormality or change can be detected, that is, whether a characteristic event can be found. As a result, if it is determined that a characteristic event cannot be found or is not clear, the condition setting unit 44 accepts a change of the visualization condition by the operator performed via the user interface 35, and the visualization condition 45 is changed (S15). The changed visualization condition 45 is stored in the memory 33.

  Next, the visualization unit 43 visualizes a part of the data set 41 again using the changed visualization condition 45 (S13).

  On the other hand, if a characteristic event is found in S14, the entire data set 41 is acquired (S16). The visualizer 43 visualizes the entire data set 41 using the final visualize condition 45 stored in the memory 33 (S17). The processing result is displayed on the display 47 by the visualizer 43, for example.

  Next, the result of visualizing the entire data set 41 is analyzed and verified (S18). In the process so far, visualization, analysis and verification are repeatedly performed on a part of the data set 41, the general tendency of the data set 41 is grasped, and a characteristic phenomenon can be found. It is easy to further analyze and verify from the visualization result of the entire data set 41.

  After analyzing and verifying the entire data set 41 in S18, an analysis process is executed on the entire data set 41, and an analysis result necessary for social use is generated (S19).

  As described above, according to the data analyzing method according to the present embodiment, a part of the data set 41 is visualized (S13), and the result is analyzed and verified (S14). If it is not found, the visualization condition is changed (S15) and visualized again (S13). If a characteristic event is found from the result (S14), the entire data set 41 is visualized under the same conditions as those for visualizing a part of the data set 41 (S16, S17). Since it is possible to perform visualization processing in a shorter time by processing a part of the data set 41 than processing the entire data set 41, a visualization condition 45 for finding a characteristic event from the data set 41 is determined in a short period of time. can do. Thereafter, analysis / verification (S18) of the data set 41 is performed, and an analysis result can be generated by an analysis process (S19).

(Details)
Hereinafter, details of the data analysis system according to the present embodiment will be described.

(data set)
The data set 41 is so-called big data. Big data usually refers to a collection of data that exceeds the capabilities of commonly used software tools to collect, select, manage, and process the data in an acceptable amount of time. It is not limited.

  In the present embodiment, the data set 41 includes at least position information and time information. Here, the “position” is a position on the earth, for example, and includes latitude and longitude. The “position” includes a position in the human body. For example, the position of red blood cells or viruses in the whole body blood flow, or the measurement position of brain wave data in the brain.

Although the specific example of the data set 41 is illustrated below, it is not limited to these.
<Environment>
Weather: For example, rainfall CO ₂ concentration

<Economic Activity>
Purchasing information:
By collecting purchase data from the POS terminal, it is possible to grasp which store the individual has moved from which store to which store. In this case, the position can be grasped from the store address. Similarly, credit card usage history can also be used. In addition, cloud-type household account book services are provided that capture receipts, record purchase information, and store them in the cloud. In this service, it is possible to identify a store from purchase information and grasp a user's action history from the store address.

Vending machine: Installation location of vending machines and the quantity of products sold at that point.
<Health>
Location of infection:
<Energy>
Solar cell installation location:

<Society>
Network social media: Twitter (registered trademark) Tweet sending location and Foursquare (registered trademark) check-in location Search query: Location of location included in search query Blog article: Location included in blog article and location of photography Transportation: Car movement record, airplane flight record

<Human movement>
For example, a person's movement can be grasped from the following position information.
Location information of mobile communication terminals such as mobile phones and wearable terminals (use of GPS and base station data)
Position of connected Wi-Fi hotspot Position of geofence captured using communication such as Wi-Fi or ultrasonic wave Position that passed automatic ticket gate

(Data set extraction)
To extract a part of the data set 41 is to reduce the processing target and shorten the processing time. For the extraction, for example, a method of randomly extracting a part of data in which position information and time information are set, a method of extracting data of a specific area or time zone, and a difference from reference data is larger than a predetermined value A method for extracting data is used, but the method is not limited to this.

(Visualize)
Data visualization means visualizing data. By visualizing the data, it is possible for an expert to grasp the general tendency of the data and find a characteristic phenomenon.

(Visualization conditions)
The visualization conditions include, for example, “methods” for analysis and expression and “parameters” necessary for each method.

  Visualization can be broadly divided into analyzing data and expressing analysis results.

Although the following can be illustrated as an analysis method, it is not specifically limited.
Analysis method using a gate: The analysis target is selected based on whether the analysis target passes through a gate (gate) arbitrarily set on the map of the target area. For example, an analysis target that has passed through the gate is a display target.
Analysis method using mesh: A map of the target area is divided into a plurality of areas, and the number of analysis objects existing in one area is grasped.

Although the following can be illustrated as an expression method, it is not specifically limited.
Map: The movement of the light spot and the locus are displayed on the map corresponding to the movement of the person. Alternatively, the change in the number of people is displayed by the density of light spots, the color shading, or the brightness. It is also possible to express the fluctuation of the number of people by the height of the point like an elevation line.
Graph: Bar graph, line graph, pie chart (also called π chart)

  The parameters vary depending on the analysis and expression method. Although the following can be illustrated, it is not specifically limited.

<Parameters of analysis method>
Location or size of gate in analysis method using gate Size of divided area (heat map) in analysis method using heat map <Parameters of expression method>
Light spot: Lightness / darkness / lightness of color (for example, light spot is displayed in dark color if the moving speed is fast), shape, size Color of divided area, lightness / darkness / lightness in analysis method using heat map ( (For example, when there are many people, the mesh is displayed in bright red with a high brightness)

(Drawing visualization results)
The visualization result is displayed on the display 47. The method of drawing data on the display 47 is the same as that of a general application program.

(Analysis and verification of visualization results)
In the analysis / verification of the visualization result, for example, it is determined whether or not a characteristic event can be found in the processing result displayed on the display 47. For example, in the case of the Tokyo Marathon, the characteristic event is that at the nearest station near the goal on the day of the Tokyo Marathon, the flow of people to the station increases from the time the runner's goal starts. Event. The determination that a characteristic event can be found may be made by an expert or an analyst, or may be made by the visualizer 43 using a threshold value or the like.

(Change visualization conditions)
The change of the visualization condition 45 includes at least one of a change of the analysis method and a parameter used for the analysis method. Visualize again using the changed visualization condition and evaluate the result.

  The visualization condition 45 is changed by, for example, selecting at least one of a plurality of prepared methods or inputting a parameter value change to the condition setting unit 44.

(All data set processing)
If a characteristic event is found from the visualization result, the entire data set 41 is obtained, and the same method and the same parameters as part of the data set 41 are used to visualize the entire data set 41. To do.

(Analysis method)
Next, the analysis method will be described in detail. In the following description, a case where data indicating movement of a large number of people is analyzed as an analysis target will be described as an example. As the movement data of this person, for example, the movement history of a mobile phone terminal with GPS owned by each person can be collected via a mobile phone company.

(Gate)
FIG. 6 is an explanatory diagram for explaining an example of an analysis method using a gate in the data analysis system according to the present embodiment. As shown in FIG. 6, two gates G1 and G2 are installed on the map. A plurality of white circle marks (61), black circle marks (62), and white triangle marks (63) in FIG. 6 indicate the positions of persons, and arrows indicate the movements. When a person passes through both the two gates G1 and G2 within a predetermined time, it is indicated by a white circle (61). Thus, only those who have passed both gates G1 and G2 are to be displayed, and the black circle (62) and the white triangle (63) pass through either one of the gates G1 or G2. However, although the case where it does not pass through the other gate G2 or G1 is shown, such a person's movement is not displayed.

  The passage of the gates G1 and G2 is, for example, whether or not a person has passed between two positions (latitude and longitude) assumed at a predetermined distance at a predetermined location on the map. It can be determined from the movement data of the person. Also, for example, if this person has grasped from the usage history of the FeliCa (registered trademark) card that he has passed an automatic ticket gate installed at a certain station, and if there is a usage history, he has passed the gate installed at that station It can also be judged.

  It is possible to express that only movement data indicating a person who has passed through both the gates G1 and G2 is displayed on the map.

(Heat map)
7 and 8 are explanatory diagrams for explaining heat map analysis in the data analysis system according to the present embodiment. As shown in FIG. 7, a plurality of grids 71 are drawn on the map at regular intervals in the east-west (left and right in the figure) and north-south (up and down in the figure), and the map is divided into a plurality of regions 72.

  In FIG. 7, a person's movement is indicated by an arrow, and a white circle 73 indicates the current position. When the person enters the area A through the point X, the weight of the area A is incremented by “1”. When the person passes the point Y and leaves the area A, the weight of the area A is decremented by “1”. In this way, the weighting of the region 72 varies with the movement of the person.

  FIG. 8 illustrates a case where the area A is displayed with a color corresponding to the weight. As shown in FIG. 8, when there is a person, the number of people existing in the area A and the number of persons existing in the other area 72 are made different from those of the other areas 72 by changing the color shade and color of the area A. Compared with the number of people, it becomes possible to grasp visually. A region B in FIG. 8 is given a different color from the region A when the weighting is larger than that of the region A. Thus, the difference in weight can be expressed by the difference in color. Furthermore, since the change in weighting over time appears as a change in the color assigned to each region 72, the movement of the person can be grasped by the change in color.

(Anonymous flow visualization method)
When a person's movement is expressed by moving a point on the map, the point represents an individual's action history. At this time, personal information is included even if the individual cannot be immediately identified. Therefore, it is desirable to visualize so as not to include personal information when using the visualization result.

  Hereinafter, in order to perform visualization without including personal information, a visualization method (anonymous flow visualization method) of movement of a person who has discarded personal information will be described. In the anonymous flow visualize method, a plurality of subject movement vectors are obtained corresponding to the direction of movement and the amount of movement from the first position at the time T1 to the second position at the other time T2, respectively. An anonymous vector is calculated from the subject movement vector at a predetermined point on the plane where the result of visualization is generated and displayed. Then, display is performed based on the anonymous vector. Thereby, a person's movement can be displayed, without displaying the subject movement vector itself corresponding to a person's movement.

  FIG. 9 is a flowchart showing an anonymous flow visualization method in the data analysis system according to the present embodiment. FIG. 10 is an explanatory diagram for explaining an anonymous flow visualization method in the data analysis system according to the present embodiment.

  First, the subject movement vector indicating the movement of the individual who is the subject of movement is calculated from the data of the movement of the person included in the data set 41 (S21). As shown in FIG. 10A, on the map, for at least two “individuals”, the moving directions and movements of the persons A and B moving from positions A and B at time T1 to positions A ′ and B ′ at time T2 Assume that there is a subject movement vector (A → A ′, B → B ′) indicating the distance. As an example, in FIG. 10A, the subject movement vector is represented by a vector in which an arrow indicating the movement direction is attached to the A-A ′ and B-B ′ line segments.

  These subject movement vectors (A → A ′, B → B ′) include personal information indicating where they were at time T1. Therefore, the following abstraction is performed. First, as shown in FIG. 10B, a plurality of points (indicated by small white circles 91 in FIG. 10B) arranged at predetermined intervals on the map are provided (S22).

  Next, as shown in FIG. 10C, at each point 91, when there is a subject movement vector (A → A ′, B → B ′) whose distance d from the point 91 is equal to or less than a predetermined value R, Corresponding to such a subject movement vector, a point vector 94 having the same direction as the subject movement vector and having a magnitude corresponding to the distance d is calculated (S23). The distance d between the point 91 and the subject movement vector corresponds to the length of the perpendicular drawn from the point 91 to the subject movement vector. The size of the point vector 94 is, for example, maximum when the distance d is zero, and decreases as the distance d increases, and is zero when the distance d is equal to or greater than R. That is, among these points 91, the subject movement vector for a point P (indicated by a small black circle 93 in FIG. 10C) centered on the point O on the subject movement vector and within the range of the circle 92 having the radius R. Point vectors corresponding to the distance from the line AA ′ and the line BB ′ to the point P93 representing (A → A ′, B → B ′) are calculated (S23). In FIG. 10D, the point vector at the point P93 is indicated by an arrow 94. The presence of two arrows 94 at one point P93 indicates that both of the two individual movement vectors (A → A ′, B → B ′) are affected.

  FIG. 11 is an explanatory diagram for explaining a calculation method of the subject movement vector field in the anonymous flow visualization method in the data analysis system according to the present embodiment. As shown in FIG. 11, when the point vector (p) corresponding to the subject movement vector (a, that is, A → A ′) at the point P93 is calculated, the following equation (1) can be used. In Expression (1), d indicates the distance of a perpendicular line drawn from the point P93 to the A-A ′ line segment.

  Next, at point P93, an anonymous vector is calculated based on the point vector calculated based on each of the subject movement vectors (A → A ′, B → B ′) (S24). For example, for the point 91, when there is no subject movement vector whose distance from the point 91 is within R, the point vector and the anonymous vector are both zero in magnitude, and the point vector and the anonymous vector are not exist. Further, regarding the point P93, when there is one subject movement vector whose distance from the point 91 is within R, there is one point vector, and the point vector and the anonymous vector may coincide. When there are a plurality of subject movement vectors whose distance from the point 91 is within R with respect to the point P93, there are a plurality of point vectors, and the anonymous vector may be the sum of a plurality of point vectors. In FIG. 10E, the anonymous vector at the point P93 is indicated by an arrow 95. The anonymous vector 95 is an abstraction of the subject movement vector (A → A ′, B → B ′). That is, the anonymous vector field formed by the anonymous vector 95 does not include the personal information that the subject movement vector has. In other words, personal information is discarded in the anonymous vector field.

  Finally, a light spot is placed on a map where there is an anonymous vector calculated based on all subject movement vectors, and the placed light spot is moved according to the anonymous vector (S25). For example, the light spot is moved in the direction of the anonymous vector by a distance corresponding to the magnitude of the anonymous vector. That is, the light spot moves at a speed corresponding to the anonymous vector. Over the necessary period, S21 to S25 are repeated to obtain the movement of the light spot as described above, and for example, by drawing on the map, the personal information is “discarded” from the movement of a plurality of individuals and abstracted. "Human movement" can be visualized by "Light spot movement".

  Here, the display corresponding to the anonymous vector is represented by the movement of a plurality of scattered light spots, but this is only an example and is not limited. For example, an arrow indicating the direction of the anonymous vector may be displayed with a thickness corresponding to the magnitude of the anonymous vector, or may be expressed with a three-dimensional height.

(Analyze)
The analysis of the data set 41 refers to generating an analysis result as information necessary for social utilization from the data set 41 based on the result of analyzing and verifying the visualization result.

  The analysis result may be the entire visualization result of the data set 41 described above. The analysis result may be a graph obtained by processing the entire data set 41 (an abstract expression obtained by visualizing information using a two-dimensional geometric model or a diagram).

  The analysis method can be a conventionally known method and is not particularly limited.

(Social use of analysis results)
The analysis results generated by the above-described data analysis system can be used socially as follows.

  For example, from the viewpoint of a business plan or business plan, it is possible to determine an increase or decrease in resources to be input from a person's movement as an analysis result.

  Also, from a marketing perspective, it is possible to diagnose product entry and exit / exit from the purchaser's travel route as an analysis result, and to determine the location of an efficient vending machine. .

  In terms of city planning, it is possible to infer roads, railroads, and houses that should be developed in the future from the movement of people as an analysis result.

  Moreover, from the viewpoint of disaster countermeasures, the movement of evacuees at the time of the occurrence of a large earthquake as an analysis result can be used for planning evacuation routes and disaster prevention facilities.

  In addition, the security system can be examined from the movement of the audience as an analysis result at events such as city marathons and fireworks displays.

  As described above, according to the data analyzing system 30 according to the present embodiment, when a part of the data set 41 is visualized and the result is analyzed and verified, a characteristic event cannot be found. Change the visualization condition 45 and visualize again. If a characteristic event is found from a part of the data set 41, visualize the part of the data set 41 with the same visualization condition 45 as the part of the data set 41 is visualized. Thus, it is possible to determine the visualization condition 45 in which a characteristic event can be found from the data set 41 in a short period of time. As a result, the analysis result necessary for social utilization can be obtained from the data set 41 in a short time.

  In addition, this invention is not limited to the said embodiment, It can change and implement variously. In the above-described embodiment, the configuration and the like illustrated in the accompanying drawings are not limited to this, and can be appropriately changed within a range in which the effect of the present invention is exhibited.

  The present invention can be suitably used for socially utilizing big data such as movement of people, for example.

30 Data Analyzing System 41 Data Set 42 Extraction Unit 43 Visualization Unit 44 Condition Setting Unit 45 Visualization Condition 46 Analyzing Unit

Claims (7)

A data analysis method using data visualization using a data analysis system including a calculation unit, storage, memory, display, and user interface ,
An extraction step in which the arithmetic unit extracts a part of the data set stored in the storage ;
Storing the first condition including at least one of a technique and a parameter used for the technique, which is set through the user interface, in the memory, and
The arithmetic unit, a portion of the extracted the data set was analyzed with the first condition, a first visualization step Ru display the analysis result on the display,
The arithmetic unit, based on an input through the user interface, to change to a second condition different from the first condition, and to store in the memory,
The arithmetic unit, and said part of the data set, the analyzed in a second condition, analysis second visualization step Ru is displayed again on the display the results,
The arithmetic unit acquires the entirety of the data set from the storage, and analyzed in the second condition stored in the memory, and a third visualization step Ru display the analysis result on the display,
Equipped with,
The data of the data set includes location information of a plurality of subjects;
Any of the first visualize process, the second visualize process, or the third visualize process,
Based on a plurality of subject movement vectors corresponding to the direction and amount of movement of the plurality of subjects from the first position at time T1 to the second position at another time T2, An anonymous vector calculation step for calculating an anonymous vector at a predetermined point on the plane displaying the rise result;
A display step in which the arithmetic unit performs display on the display based on the anonymous vector;
Data Analyze wherein to have a.   The data analysis method according to claim 1, wherein the data of the data set includes position information indicating a position.   3. The data analyzing method according to claim 2, wherein data is displayed on a map in any one of the first visualizing step, the second visualizing step, and the third visualizing step. In the anonymous vector calculation step,
The computing unit has the same direction as the subject movement vector at the predetermined point, corresponding to the subject movement vector whose distance d from the predetermined point is equal to or less than a predetermined value. Calculating a point vector having a magnitude according to
The arithmetic unit calculating the anonymous vector based on the point vector;
The data analysis method according to claim 2 or 3, wherein: Said display based on anonymous vector is a light spot in the direction of the anonymous vector, data analyze process according to claim 1, wherein the moving distance corresponding to the magnitude of the anonymous vector. Computation unit, storage, memory, display and user interface,
The computing unit is
Extract a portion of the data set stored in the storage ,
Storing in the memory a first condition including at least one of a method and a parameter used for the method set through the user interface;
Some of the extracted the data set was analyzed with the first condition, to display the analysis result on the display,
Based on the input through the user interface, the second condition different from the first condition is changed and stored in the memory,
A part of the data set is analyzed under the second condition, and the analysis result is displayed again on the display ;
The Get the entirety of the data set from the storage, the memory and analyzed by the stored second condition, the analysis result A data analyze system using data visualization to Ru is displayed on the display,
The data of the data set includes location information of a plurality of subjects;
Based on a plurality of subject movement vectors corresponding to the direction and amount of movement of the plurality of subjects from the first position at time T1 to the second position at another time T2, Calculate an anonymous vector at a predetermined point on the plane displaying the rise result,
A data analysis system for displaying on the display based on the anonymous vector . The computing unit has the same direction as the subject movement vector at the predetermined point, corresponding to the subject movement vector whose distance d from the predetermined point is equal to or less than a predetermined value. Calculate a point vector with a corresponding size,
The arithmetic unit calculates the anonymous vector based on the point vector.
The data analysis system according to claim 6.

Images