NETWORK ANALYSIS METHOD, NETWORK ANALYSIS SYSTEM AND SERVER
The network analysis method addresses the challenge of understanding time-series changes in group networks by collecting activity data, extracting cliques, and calculating a group consciousness index to visualize activity and diversity, facilitating easy comprehension of group dynamics.
Patent Information
- Application Number
- JP2022009334
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2022-01-25
- Publication Date
- 2024-12-09
- Estimated Expiration
- 2042-01-25
AI Technical Summary
Existing technologies struggle to easily understand time-series changes in the network structure between individuals within a group, particularly in expressing the degree of activity and diversity of group communication.
A network analysis method using a computer system that collects time-series activity data from individuals, extracts cliques, and calculates a group consciousness index to visualize changes in network structure and activity level.
Enables the visualization of time-series changes in group network structure by quantifying activity and diversity, allowing easy grasping of group consciousness.
Smart Images

Figure 0007595885000011 
Figure 0007595885000012 
Figure 0007595885000013
Abstract
Description
[Technical field]
[0001] The present invention relates to a technique for analyzing a population network. [Background technology]
[0002] With the advancement of sensing devices and communication technology, techniques for analyzing group behavior have been attracting attention in the fields of sports, business, and education. There are many known cases where the results of group analysis have been used to activate groups.
[0003] In the analysis of groups, groups that communicate are broadly defined by considering not only conscious communication between individuals, such as face-to-face contact and conversation, but also unconscious communication, such as the synchronization of body movements between individuals. It is also known that the situation of communication within a group can be expressed using a network diagram.
[0004] In the field of sports, a technology for analyzing the relationships between individuals during a game is known as a technology for analyzing networks within a group (Patent Document 1). In the field of business, a technology for extracting communities within an organization is known as Patent Document 2.
[0005] Patent Document 1 discloses a technique for extracting individuals that act in synchronization with a group and visualizing the group network. Patent Document 2 discloses a technique for extracting communities within an organization and measuring and visualizing the quality of the extracted communities. [Prior art documents] [Patent documents]
[0006] [Patent Document 1] JP 2018-81406 A [Patent Document 2] JP 2012-27589 A Summary of the Invention [Problem to be solved by the invention]
[0007] The time series changes in a group network are very complex, and it is difficult to easily grasp the structural changes in the network between individuals. In other words, the above conventional technology can calculate the relationship and activity between individuals at a certain time, and can grasp the number of relationships between individuals. However, the above conventional technology has a problem in that it cannot easily express the changes in the activity and diversity of communication in a group.
[0008] The present invention has been made in consideration of the above problems, and aims to visualize the activity level and changes in network structure of a group that is engaged in purposeful activities. [Means for solving the problem]
[0009] The present invention is a network analysis method in which a computer including a processor and a memory analyzes a network between individuals in a population, the method including: a sensing data collection step in which the computer collects time-series information on the activities of individuals constituting the population and stores the information in sensor data; a network analysis step in which the computer extracts activated individuals from the sensor data as cliques and generates a network composed of the cliques in time series; and a group consciousness analysis step in which the computer calculates the occurrence probability of the cliques and the information entropy of the cliques to calculate an index indicating the activity level of the population and the diversity of the population as a group consciousness index. Effect of the Invention
[0010] Therefore, by quantifying indicators showing the activity level and diversity of a group, the present invention makes it possible to visualize changes over time in the network structure of a group that is engaged in purposeful activities, thereby making it possible to easily grasp the consciousness of the group.
[0011] The details of at least one implementation of the subject matter disclosed herein are set forth in the accompanying drawings and the description below. Other features, aspects, and advantages of the disclosed subject matter will become apparent from the following disclosure, drawings, and claims. [Brief description of the drawings]
[0012] [Figure 1] FIG. 1 illustrates a first embodiment of the present invention, and is a diagram illustrating an example of a network analysis system applied to the field of sports. [Diagram 2] 1 is a flowchart illustrating an outline of a process performed by a server according to the first embodiment of the present invention. [Diagram 3] FIG. 2 illustrates an example of sensor data according to the first embodiment of the present invention. [Figure 4] FIG. 2 illustrates an example of feature amount information according to the first embodiment of the present invention. [Diagram 5] 5 is a flowchart illustrating an example of processing performed by the network analysis unit according to the first embodiment of the present invention. [Figure 6] FIG. 2 illustrates an example of synchronous network time series data according to the first embodiment of the present invention. [Figure 7] FIG. 2 illustrates a first embodiment of the present invention and is a diagram illustrating an example of a relationship between sensor data and synchronous network time-series data. [Figure 8] FIG. 1 illustrates a first embodiment of the present invention and is a diagram illustrating an example of extracting a clique from synchronous network time series data. [Figure 9] FIG. 11 illustrates the first embodiment of the present invention, and is a diagram illustrating an example of occurrence probability data according to the type of clique. [Figure 10] 10 is a flowchart illustrating an example of processing performed by the collective consciousness analysis unit according to the first embodiment of the present invention. [Figure 11] FIG. 1 shows Example 1 of the present invention and is a plot diagram showing areas of collective consciousness. [Figure 12] FIG. 1 shows Example 1 of the present invention and is a plot diagram showing trends in group consciousness. [Figure 13]FIG. 1 shows Example 1 of the present invention and is a plot diagram of three cliques obtained by analyzing a population of athletic fields. [Figure 14] FIG. 11 illustrates a second embodiment of the present invention, and illustrates an example of a network analysis system applied to the field of education. [Figure 15] FIG. 11 illustrates a second embodiment of the present invention and is a diagram illustrating an example of the relationship between the meeting matrix at time T1 and communication network time-series data. [Figure 16] FIG. 11 illustrates a second embodiment of the present invention, and is a diagram illustrating an example of the relationship between the meeting matrix at time T2 and communication network time-series data. [Figure 17] FIG. 13 shows Example 2 of the present invention and is a plot diagram of one clique analyzed in a population in the education field. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0013] Hereinafter, an embodiment of the present invention will be described with reference to the accompanying drawings. EXAMPLES
[0014] 1 is a diagram showing an example in which a network analysis system is applied to a group sport. The network analysis system of the first embodiment is a computer system that analyzes relationships based on the movements of individuals (players) 60 constituting a group, extracts a network (connections between individuals), generates time-series data, and visualizes changes in the activity level and diversity (structure) of the group.
[0015] In the following, a computer system for visualizing human networks will be described, but the present invention may be applied to any system that analyzes and visualizes networks between moving individuals, such as animals.
[0016] The network visualization system includes a sensing device 1, a network 4, and a server 5.
[0017] An example of a group that is an analysis target of the network analysis system in the first embodiment is a group that plays soccer. The type of exercise may be any sport in which two or more individuals participate, and the network analysis system in the first embodiment may be applied to group sports such as rugby, basketball, volleyball, goalball, baseball, tennis, handball, lacrosse, track and field, and speed skating.
[0018] Furthermore, the network analysis system of Example 1 is not limited to networks between individuals, and may be used to display networks between human bodies and objects, or between objects, by measuring the movements of objects such as tools and implements such as balls, bats, and rackets using the sensing device 1.
[0019] In the first embodiment, one or more sensing devices 1 are attached to the body of an individual 60 performing exercise. The individual 60 in the first embodiment shows an example in which a wristwatch-type wearable device equipped with a three-axis acceleration sensor is directly attached to the human body as the sensing device 1. The individual 60 performs exercise on a field, but the place where the individual performs exercise is not limited to a field and may be any place.
[0020] In addition, by specifying the location where the exercise will be performed, even if the individual wearing the sensing device 1 is in multiple locations, it is possible to display a network of only the individuals in the specified location. In this embodiment, the individual 60 is an individual displayed in the network and is also an individual measured by the sensing device 1.
[0021] The sensing device 1 in this embodiment is equipped with at least one or more sensors among various sensors that measure the activity of an individual. The various sensors include, but are not limited to, an acceleration sensor, a gyro sensor, a pedometer, a heart rate monitor, a position sensor, a shock sensor, a magnetometer, a temperature sensor, a humidity sensor, a wind sensor, a sound sensor, a pressure sensor, a positioning sensor, an infrared sensor, and a position information sensor. The results measured by these various sensors may be transmitted to the server 5 as sensing data.
[0022] The individual 60 may also wear the sensing device 1 so as to be physically connected to a body part of the individual 60, such as the head, neck, shoulders, back, arms, wrists, hands, fingers, waist, hips, legs, ankles, feet, heels, and toes. If there is one or more pieces of clothing, footwear, or athletic protective gear between the sensing device 1 and the body of the individual 60, the individual 60 may wear the sensing device 1 by various removable or non-removable connecting means such as straps, adhesives, pockets, and clips, in a state where the sensing device 1 is integrated with the clothing, footwear, or athletic protective gear used in the athletic activity. The sensing device 1 may also be attached to or embedded in an object such as an implement or tool, such as a ball, bat, or racket.
[0023] In addition, the sensing device 1 is not limited to a wearable sensor, but may be any device that can measure the movement of an individual or object, such as an instrument that measures the movement of an individual or object through image analysis, a laser radar that can detect the position of an object by reflecting laser light, a force plate that can measure pressure and impact, or other installed sensors such as an acoustic sensor, a magnetic sensor, an RGB camera, a depth sensor, or a multi-array microphone.
[0024] The sensing data relating to the movement measured by the sensing device 1 is transmitted to the server 5 via the network 4, where analysis processing and the like are performed. The analysis results are displayed on the display of the input / output device 56 or the like.
[0025] The server 5 of this embodiment analyzes the sensing data received from the sensing device 1 to generate synchronized network time series data (described later) that indicates the cooperation or relationships between the individuals 60, and calculates the time-series changes in the activity and diversity (structure) of the group as a group consciousness index from the synchronized network time series data. The server 5 visualizes the changes in the activity and structure of the group over time by outputting the group consciousness index as a plot diagram on a display or the like.
[0026] The server 5 is a computer having a CPU 51, a memory 54, an auxiliary storage device (storage device) 55, a communication unit 52, and an input / output device 56. In the following description, a network without a symbol is used to indicate cooperation between individuals, and is distinguished from the network 4 that connects computers.
[0027] The server 5 realizes the functions of the analysis unit 10 and the information input unit 53 by executing, with the CPU 51, a program loaded from the auxiliary storage device 55 to the memory 54. The memory 54 of the server 5 can be configured with a ROM, which is a non-volatile storage element, or a RAM, which is a volatile storage element.
[0028] The ROM stores programs (e.g., BIOS), etc. The RAM is a high-speed, volatile storage element such as a DRAM (Dynamic Random Access Memory), and temporarily stores the programs stored in the auxiliary storage device 55 and data used when the programs are executed.
[0029] The database 20 is stored in the auxiliary storage device 55. The auxiliary storage device 55 is, for example, a large-capacity non-volatile storage device such as a magnetic storage device (HDD) or a flash memory (SSD). The auxiliary storage device 55 also stores the programs executed by the CPU 51 and data used by the programs. That is, the programs are read from the auxiliary storage device 55, loaded into the memory 54, and executed by the CPU 51.
[0030] The server 5 has an information input unit 53 that stores the sensing data received from the sensing device 1 in the sensor data 22 of the database 20, and collects the sensing data transmitted at a predetermined cycle.
[0031] The analysis unit 10 is a functional unit that calculates acceleration sensor and position information of each individual 60 in a time series based on the sensor data 22, calculates a network between individuals based on the movement and trajectory of the individual detected from the acceleration sensor information and position information, and calculates a collective consciousness index W from the network. When cooperation or relationships between individuals in a competition or the like are expressed as edges, a node containing one or more edges is considered to be a clique. A node in which one node is activated can also be treated as a clique. A clique can be treated as a complete partially connected network.
[0032] The analysis unit 10 mainly has functional units of a feature amount calculation unit 11, a network analysis unit 12, a group consciousness analysis unit 13, and a display unit 14. Note that the position of an individual may be measured using position information detected by the GNSS 35 or the like in addition to the acceleration of the sensing device 1 attached to the individual.
[0033] The display unit 14 outputs the data recorded in the database 20 to a computer or device connected via the network 4, or to an input / output device 56, thereby providing the data to a user (a user of the analysis system) who evaluates the group. The input / output device 56 is composed of a keyboard, a mouse, a display, etc.
[0034] The CPU 51 operates as a functional unit that provides a predetermined function by processing according to the program of each functional unit. For example, the CPU 51 functions as the analysis unit 10 by processing according to an analysis program. The same applies to other programs. Furthermore, the CPU 51 also operates as a functional unit that provides each function of a plurality of processes executed by each program. A computer and a computer system are devices and systems that include these functional units.
[0035] The communication unit 52 is a network interface that controls communication with other devices in accordance with a predetermined protocol, and is capable of communicating with other devices via the network 4 and transmitting and receiving data.
[0036] The information input unit 53 functions as a sensing data collection unit that receives sensing data transmitted from the sensing device 1 and stores it in the sensor data 22 of the database 20. The information input unit 53 can receive input from the input / output device 56 and reflect it in the database 20.
[0037] The analysis unit 10 and the information input unit 53 may be implemented by a program, or may be implemented by a physical integrated circuit. In particular, the analysis unit 10 may be implemented by multiple programs or multiple integrated circuits for executing the functional units included in the analysis unit 10. Furthermore, the feature calculation unit 11, the network analysis unit 12, the collective consciousness analysis unit 13, and the display unit 14 may each be implemented by multiple programs or multiple integrated circuits for each process that they execute.
[0038] The programs executed by the CPU 51 are provided to the server 5 via a removable medium (such as a CD-ROM or a flash memory) or a network, and are stored in the auxiliary storage device 55, which is a non-transitory storage medium. For this reason, the server 5 may have an interface for reading data from the removable medium.
[0039] The server 5 is a computer system configured on one physical computer, or on multiple logical or physical computers, and the above-mentioned programs may be executed in different threads on the same computer, or may be executed on virtual computers constructed on multiple physical computers.
[0040] The database 20 stores sensor data 22, feature amount information 21, synchronous network time series data 24, occurrence probability data 25, entropy data 30, and a collective consciousness graph 40.
[0041] The sensing data measured by the sensing device 1 of each individual 60 is collected by the information input unit 53 and stored as sensor data 22. The sensor data 22 in this embodiment indicates the results of movement measurements for each individual 60.
[0042] The feature amount information 21 indicates a feature amount related to the movement of each individual 60, which is calculated by executing the feature amount calculation unit 11 on the sensor data 22. Details will be described later.
[0043] The synchronization network time series data 24 indicates information about the network between the individuals 60 calculated by the network analysis unit 12. Details will be described later.
[0044] The occurrence probability data 25 is data obtained by calculating the probability of occurrence of a clique constituting a network in a time series. The entropy data 30 is time-series data obtained by calculating the information entropy of a group using the occurrence probability data 25, and will be described in detail later.
[0045] The group consciousness graph 40 is composed of a plot diagram in which the activity level and structure of the group calculated based on the information entropy are plotted in time series. The group consciousness graph 40 can store plot diagrams for each number of cliques. The display unit 14 of the server 5 displays the plot diagrams read from the group consciousness graph 40 on the display of the input / output device 56, etc., thereby functioning as a visualization unit that visualizes the consciousness of the group, consisting of the activity level of the group and changes in the network structure, in time series.
[0046] The sensing device 1 is a device that measures the movement of a target individual for analyzing a network. In this embodiment, a wearable device worn by the target individual is taken as an example. Note that the sensing device 1 may measure any content as sensing data as long as the content changes in value as a result of the user's movement.
[0047] The sensing device 1 includes an acceleration sensor 31, a Global Navigation Satellite System (GNSS) 35, a memory 32, a microcomputer (MCU in the figure) 33, and a communication unit 34.
[0048] The acceleration sensor 31 measures the acceleration of the user, for example, about 20 to 1000 times per second. The GNSS 35 can be configured as a position information system such as a Global Positioning System (GPS) or a Quasi-Zenith Satellite System (QZSS), and measures the position information of the individual. In this embodiment, the position information from the GNSS 35 is used as the position information of the individual 60 at the start of the competition, and the sensing data of the acceleration sensor 31 is used to measure the movement of the individual 60 during the competition.
[0049] The microcomputer 33 records the measurement results obtained by the acceleration sensor 31 and the GNSS 35 as sensing data in the memory 32. Here, the microcomputer 33 records an identifier (user ID) unique to each user and the measurement results as the sensing data.
[0050] The microcomputer 33 transmits the sensing data and the position information recorded in the memory 32 to the server 5 via the communication unit 34. The communication unit 34 transmits the sensing data to the server 5, wirelessly or via wired communication, at a timing when communication is possible or at a predetermined timing set by a user or an administrator.
[0051] <Processing Overview> FIG. 2 is a flowchart showing an example of processing performed by the analysis unit 10 of the server 5.
[0052] First, the feature amount calculation unit 11 acquires sensing data to be processed from the sensor data 22, and calculates feature amount information 21 related to the movement of each individual 60 from the target sensor data 22 (S101).
[0053] The feature amount information 21 indicates a feature amount whose value changes according to the strength of the movement (or the magnitude of the energy of the movement) of the individual 60, such as the amount of body movement, and further indicates the change in the feature amount in a time series.
[0054] In this embodiment, the physical activity intensity is calculated as the characteristic amount, but any characteristic amount related to exercise such as amount of exercise, amount of body movement, moving speed, activity frequency, pitch, number of steps, etc. Furthermore, characteristic amounts such as calorie consumption, heart rate, pulse rate, amount of sweat, etc. that change with exercise may be calculated as the characteristic amount.
[0055] Next, the network analysis unit 12 uses the calculated feature amount information 21 to perform network analysis between individuals 60 to generate synchronization network time series data 24 (S102). Then, the group consciousness analysis unit 13 executes the processes of steps S21 to S23 to extract cliques from the network between individuals 60 generated, calculates the information entropy for each type of clique as described below, calculates the group consciousness index W (Index of group consciousness), and generates a plot diagram (S103).
[0056] The collective consciousness analysis unit 13 first extracts cliques, which are activated nodes, from the synchronization network time series data 24 generated in step S102, calculates the occurrence probability for each type of clique, and calculates the sum of the occurrence probabilities k i is generated as occurrence probability data 25 (S21). As will be described later, the total occurrence probability k i is an index showing the activity of the population. Note that the subscript i of the sum of the occurrence probabilities k indicates the number of cliques.
[0057] Next, the group consciousness analysis unit 13 calculates the information entropy Hi at each time for each type of clique from the occurrence probability data 25. The group consciousness analysis unit 13 calculates the ratio of the information entropy Hi at each time to the information entropy Hi (max) of the network in a random state calculated for each type of clique, as an index showing the diversity (composition) of the group (S22).
[0058] The collective consciousness analysis unit 13 calculates the sum of occurrence probabilities k iThe set of ratios of the information entropy Hi calculated in step S22 is taken as the collective consciousness index W, and plot diagrams 400 shown in FIGS. 11 and 13 are generated for each number of cliques and stored in the collective consciousness graph 40 (S23).
[0059] The display unit 14 outputs the plot diagram stored in the collective consciousness graph 40 to the display of the input / output device 56, and provides the user of the network analysis system with a time series of changes in the activity level and network structure of groups that are acting with a purpose.
[0060] By executing the process shown in FIG. 2, the analysis unit 10 uses the sensor data 22 to analyze the network between the individuals 60 to be analyzed, generates the synchronous network time series data 24, and extracts cliques.
[0061] The group consciousness analysis unit 13 calculates the total occurrence probability k for each type of clique. i The ratio (or difference) of the information entropy Hi at each time to the information entropy Hi(max) of the network state in a random state is calculated as a group consciousness index W, and the group consciousness index W is plotted in space to generate a plot diagram. The plot diagram makes it possible to visualize the activity level of a group that is acting with a purpose and changes in the network structure over time.
[0062] <Network analysis details> 3 to 6 show data and processing contents used in the processing performed by the feature amount calculation unit 11 and the network analysis unit 12. Fig. 3 is a diagram showing sensor data 22 in the first embodiment.
[0063] The sensor data 22 includes a user information table 201 that stores information about the individual 60 wearing the sensing device 1, and a movement information table 202 that records activity amount information of each individual 60.
[0064] The user information table 201 is preset information, and includes in each entry a user ID 2011, a user name 2012, and a user type 2013. The user ID 2011 stores a unique identifier for identifying the individual 60 wearing the sensing device 1.
[0065] The user name 2012 stores the name or nickname of the individual 60 wearing the sensing device 1. The user type 2013 may record age, sex, and the like as user information, and also records various information of the individual 60 that is not limited to these.
[0066] The movement information table 202 stores the sensing data for each time received from the sensing device 1. The movement information table 202 includes a user ID 2021, a measurement date and time 2022, an acceleration x-axis 2023, an acceleration y-axis 2024, an acceleration z-axis 2025, a latitude 2026, and a longitude 2027 in one entry.
[0067] User ID 2021 stores a user ID for identifying individual 60, and corresponds to user ID 2011. Measurement date and time 2022 stores the date and time when sensing data was measured. Acceleration X-axis 2023, acceleration Y-axis 2024, and acceleration Z-axis 2025 store the measurement results of each axis of the three-axis acceleration sensor. Latitude 2026 and longitude 2027 store the measurement values of GNSS 35. Note that the movement information table 202 can store a body movement frequency (not shown) calculated from the acceleration data. The body movement frequency can be calculated by server 5, but may also be calculated by sensing device 1.
[0068] 4 is a diagram showing an example of data stored in the feature information 21. The feature calculation unit 11 calculates METs (Metabolic Equivalents) indicating exercise intensity as feature amounts for the sensor data 22 of all individuals 60 constituting the group. The calculation of the feature amounts is similar to that of Patent Document 1.
[0069] The feature information 21 stores features at each date and time 222 of all individuals 223 for which network analysis was performed. In this embodiment, METs are calculated as the feature from the sensor data 22 measured at a predetermined sampling frequency of 20 Hz, and values totaled every second are stored.
[0070] Fig. 5 is a flowchart showing the details of step S102 in Fig. 2 performed by the network analysis unit 12 of the embodiment 1. The network analysis unit 12 analyzes the relationships between the individuals constituting the population from the feature amounts calculated in step S101.
[0071] First, in step S1021, the network analysis unit 12 acquires the time interval of data used to analyze the network between individuals 60. A preset value may be used as the time interval of the data, so long as it is equal to or greater than the measurement interval (sampling period or a predetermined time interval) of the sensing device 1.
[0072] In step S1022, the network analysis unit 12 selects two individuals for network analysis. In this embodiment, two individuals are selected in sequence to analyze the presence or absence of correlation between the two individuals for all combinations of individuals that make up the population.
[0073] In step S1023, the feature amount calculated in step S101 of FIG. 2 is obtained from the feature amount information 21 for the individual 60 selected in step S1022 above.
[0074] In steps S1024, S1025, and S1026, the relationship between the two individuals is analyzed to determine whether or not a network exists. In this embodiment, the amount of information transmitted between the individuals 60 is calculated using transfer entropy to determine whether or not a network exists.
[0075] Transfer entropy is a type of information amount, and is a numerical value of the information flow between random variables as the average information amount. Unlike mutual information, which shows only the correlation between variables and does not include the flow of information, transfer entropy can specify the flow of information and is an index that measures the relationship between variables. In this embodiment, transfer entropy is used, but any index that measures the relationship between multiple variables, such as mutual information, correlation analysis, regression analysis, multivariate analysis, etc., and any method for calculating the index may be used.
[0076] In step S1024, the network analysis unit 12 calculates a probability distribution from the time series data of the feature of each individual 60. A method of approximating the probability distribution using a histogram is adopted, but a method of estimating by assuming a Gaussian distribution or a mixture of Gaussian distributions may also be adopted, and the method is not limited to these. When approximating using a histogram, the number of divisions of the histogram uses the Akaike Information Criterion or Sturges's formula, but is not limited to these.
[0077] In step S1025, the network analysis unit 12 calculates, from the transfer entropy, the amount of information transmission between the variables of the two individuals 60. The transfer entropy is calculated in the same manner as in Patent Document 1.
[0078] In step S1026, the network analysis unit 12 uses the value of the transfer entropy calculated in step S1025 to determine whether or not a network exists between two individuals. The network analysis unit 12 determines that information has been transmitted between individuals 60 whose transfer entropy is equal to or greater than a predetermined threshold, and that a network exists. In this embodiment 1, the threshold is a predetermined constant. The threshold may be a threshold determination method using a x2 distribution, a gamma distribution, or the like, and is a value that is preset depending on the sensing device 1 used, the environment in which the measurement is performed, and the like.
[0079] In step S1027, the network analysis unit 12 stores the presence or absence of the network determined in step S1026 in the synchronous network time series data 24.
[0080] In this embodiment, if a network exists between two individuals, a value according to the transfer entropy is set in the synchronization network time series data 24, and if not, the synchronization network time series data 24 is set to 1. The value according to the transfer entropy is determined by setting multiple thresholds and judging the network strength in several stages.
[0081] In step S1028, steps S1022 to S1027 are repeated until analysis of all combinations of individuals constituting the population is completed.
[0082] Through the above process, the magnitude of the relationship between the individuals 60 is calculated based on the magnitude of the transfer entropy, the individuals that make up the network are identified, and a network within the population is generated.
[0083] 6 is a diagram showing data stored in the synchronous network time series data 24 of Example 1. The synchronous network time series data 24 is a brute force diagram of the network at a certain time, and describes the presence or absence (strength or weakness) of the network at a certain time t for all individuals subjected to the network analysis.
[0084] The synchronous network time series data 24 is time series data generated at a predetermined time interval, and in the illustrated example, the synchronous network time series data 24 is generated at every predetermined time interval ΔT from time T1.
[0085] The first column of the synchronous network time series data 24 is the user ID 242 of the individual who gives information and influence, and the first row is the user ID 243 of the individual who gives information and is influenced. The value between two individuals shown by the row and column indicates the strength of the network on a five-point scale.
[0086] When the relationship between the individuals 60 is strong, "5" is set, and when the relationship between the individuals 60 is weak, "1" is set. In this embodiment, the strength of the relationship is described in five stages, but it may be expressed in any number of stages. The strength of the relationship between the individuals 60 can be expressed by degree centrality or closeness centrality, as disclosed in the above-mentioned Patent Document 1.
[0087] Furthermore, the synchronous network time series data 24 can be expressed not only by the strength of the relationship between the individuals 60, but also by the presence or absence of the relationship. For example, if there is a relationship between the individuals 60, a "1" may be set, and if there is no relationship between the individuals 60, a "0" may be set.
[0088] 7 is a diagram showing the relationship between the movement information table 202 of the sensor data 22 and the synchronous network time series data 24. In the illustrated example, a part of the synchronous network time series data 24 generated from the acceleration data acquired from the movement information table 202 is displayed as a graph, and the time-series change of the relationship between a plurality of individuals 60 is shown.
[0089] At time t1, the individual 60 with ID="1" has relationships with two individuals 60 with ID="2" and "4", and the individual 60 with ID="3" has a relationship with an individual 60 with ID="6". At time t2, when a predetermined time interval Δt (e.g., one minute) has elapsed, the individual 60 with ID="1" changes its related individual 60 from ID="4" to ID="5", and has relationships with two individuals 60 with ID="2" and "5". At time t3, the individual 60 with ID="1" has ID="5" as its only related individual 60.
[0090] The synchronous network time series data 24 is a network based on synchronization between individuals 60, and by comparing the synchronous networks between individuals 60 over time, it is possible to analyze sports involving intense group movements, such as soccer, rugby, or basketball.
[0091] Fig. 8 shows an example of extracting a clique from the synchronization network time series data 24. As shown in Fig. 7, the network analysis unit 12 can generate a graph from the synchronization network time series data 24 and extract changes in the connections of an individual 60 in time series.
[0092] The network analysis unit 12 extracts a predetermined number of cliques from the synchronous network time series data 24, and aggregates and outputs the cliques at each time in time series. In this embodiment, as shown in the figure, 1 clique consisting of one individual 60, 2 cliques consisting of two individuals 60, and 3 cliques consisting of three individuals 60 are extracted as evaluation targets.
[0093] The graph at time t1 in Fig. 7 can be decomposed into the relationships of 1 clique, 2 cliques, and 3 cliques shown in Fig. 8. In Fig. 8, the network analysis unit 12 extracts activated nodes from among the nodes constituting the graph as 1 clique. An activated node can be, for example, a node whose body movement frequency obtained from the movement information table 202 is equal to or greater than a predetermined threshold (e.g., 2 Hz) and which can be extracted by the network analysis unit 12 as 1 clique.
[0094] Next, the network analysis unit 12 connects the individuals 60 having a relationship with each other with one edge, and extracts a graph having the nodes at both ends of the edge as vertices as a 2-clique. In other words, a combination of nodes with degree 1 forms a 2-clique. For example, in the graph at time t1, IDs "1" and "2" form a 2-clique, and furthermore, IDs "1" and "4", IDs "2" and "4", and IDs "3" and "6" are extracted as 2-cliques. Note that in this embodiment, the direction of the edges is not evaluated and the graph is treated as an undirected graph.
[0095] When the node at the destination of the relationship in two cliques is related to a third node, and the node at the source of the relationship also has a relationship with the same third node, the network analysis unit 12 extracts these three nodes as a three-clique.
[0096] Each node in a 3-clique has degree 2, and the three nodes are related to each other. In the graph at time t1, in a 2-clique with IDs "1" and "2", if node "1" is the source of the relationship and node "2" is the destination of the relationship of "1", then the destination node "2" has a relationship with another (third) node "4", and node "1" of the source of the relationship has a relationship with the same other node "4". In this way, a configuration in which the edges connecting the three nodes are closed is called a 3-clique.
[0097] <Definition of collective consciousness> In a group of unrelated strangers who happen to be in the same place, if a network (a synchronous network or a communication network) is generated by the synchronization of body movements, it is thought to be a network with uniform random connections. This is because even if it appears synchronized, it is presumed to be by chance.
[0098] On the other hand, if a group is acting with some purpose, it is believed that the network structure will be different from that of a "group of complete strangers who have no relationship with each other and happen to be in the same place." Therefore, in this embodiment, the extent to which the network structure of a "group acting with a purpose" differs from a network that is uniformly and randomly connected is quantified and used as an index representing the "group consciousness."
[0099] In this embodiment, a collective consciousness index W that represents the "consciousness of a group" is defined using cliques (=attributes) that arise in a synchronous network, and the difference from a uniformly random group is visualized, making it possible to distinguish the group consciousness.
[0100] If each individual 60 is a node, and the nodes in the population are activated randomly and with equal probability, the probability that a node will be activated is p j The sum of k 1 teeth,
[0101]
number
[0102] and 0≦k 1 Let n be the number of nodes in a random network that are activated with equal probability. The information entropy of the state in which one node is activated with equal probability is H 1 If (max), it can be expressed by the following equation (2).
[0103]
number
[0104] If there is a network structure consisting of n nodes, the information entropy of the state in which one node is activated is defined as H 1 Then, it can be expressed by the following equation (3).
[0105]
number
[0106] The difference in information entropy between a random network and the activation of n nodes, I 1 teeth,
[0107] I 1 =H 1 (max)-H 1
[0108] It is expressed as the difference in information entropy I 1 indicates how far each node is from being activated randomly with equal probability, in other words, the information entropy H 1 This represents the extent to which a group is networked.
[0109] Difference I 1 is zero when all nodes are activated with probability 1, when no nodes are activated with probability 1, and when the activation probability of all nodes is equal. 1However, it is maximized when the activated node is fixed. When the node is fixed, for example, the relationship between the nodes does not change, which indicates a state of low diversity. The above is an example of a network structure in which one clique appears when one node is activated.
[0110] Next, we consider a two-clique network structure in which two nodes are activated and have a relationship. The probability that two nodes are activated and connected is p jq The sum of k 2 is expressed by the following equation (4).
[0111]
number
[0112] Here, j and q indicate the identifiers of the connected nodes. The result of calculating the probability that a node will be activated is stored in occurrence probability data 25 shown in FIG.
[0113] The occurrence probability data 25 is calculated using a plurality of time series networks. The number of time series networks that calculate the occurrence probability data 25 is adjusted as a parameter according to the measurement target. The time interval at which the occurrence probability data 25 is updated is also adjusted as a parameter according to the measurement target. Therefore, the occurrence probability data 25 is generated at regular time intervals that are determined to be optimal according to the measurement target, and components 251 and probabilities 252 are set for each of 1 clique, 2 cliques, and 3 cliques. For example, when a population is made up of 6 nodes as shown in the figure, the occurrence probability p j p 1 From p 6 The probability of two cliques appearing is p jq p 12 From p 56 The probability of occurrence of 3 cliques is p jqr p 123 From p 456 Here, j, q, and r indicate the identifiers of the connected nodes.
[0114] Let H be the information entropy of the state (2-clique) in which two activated nodes j and q are connected with equal probability at random. 2 (max) is the maximum value and is expressed by the following equation (5).
[0115]
number
[0116] Here, N indicates the number of combinations of connected nodes. If there is a network structure consisting of n nodes, the information entropy of the state in which two nodes are connected is expressed as H 2 Then, it can be expressed by the following equation (6).
[0117]
number
[0118] The difference between the information entropy H2(max) of a network in which two nodes j and q are randomly connected with equal probability and the information entropy I 2 teeth,
[0119] I 2 =H 2 (max)-H 2
[0120] It is expressed as the difference in information entropy I 2 represents how far apart the two cliques in a population are from being randomly connected with equal probability, i.e., how organized they are.
[0121] Information entropy difference I 2 becomes zero when all nodes are connected with probability 1, when no nodes are connected with probability 1, and when all paths have the same probability of appearing. 2 is maximized when the network structure is fixed.
[0122] Next, we consider a 3-clique network structure in which three nodes are activated and connected. The probability that three nodes are activated and connected is p jqr The sum of k 3 is expressed by the following equation (7).
[0123]
number
[0124] Here, j, q, and r are the identifiers of the nodes to be connected. The information entropy of the state in which three activated nodes j, q, and r are randomly connected with equal probability (3-clique) is defined as H 3 (max) is expressed by the following equation (8).
[0125]
number
[0126] If there is a network structure consisting of n nodes, the information entropy of the state in which three nodes j, q, and r are connected is defined as H 3 Then, it can be expressed by the following equation (9).
[0127]
number
[0128] The information entropy H of a network in which three nodes j, q, and r are randomly connected with equal probability 3 (max) and the information entropy H 3 Difference I 3 teeth,
[0129] I 3 =H 3 (max)-H 3
[0130] It is expressed as the difference in information entropy I 3represents how far the three cliques in a population are from being randomly connected with equal probability, i.e., how organized they are.
[0131] As with the two cliques above, the difference I 3 is zero when all 3-cliques appear with probability 1, when no 3-cliques appear with probability 1, and when all 3-cliques appear with equal probability. 3 is maximized when the network structure is fixed.
[0132] As described above, the difference in entropy based on the network in a random state can be used as an index of the consciousness of the group. In this embodiment, the number of cliques is i, and the information entropy H i (max) and information entropy H i Using the ratio of these, the collective consciousness index W is defined as follows:
[0133]
number
[0134] For example, in sports, the network within a team tries to approach perfect connection, but cannot be achieved due to the presence of opposing teams. Similarly, the opposing teams also try to approach perfect connection, so there is constant pressure to avoid a completely unconnected state. Since the network structure is not fixed but changed to accommodate the change, the element of the entropy ratio of the collective consciousness index W transitions between states that are neither 0 nor 1. Such a state is defined as the group having consciousness. In addition, the sum of the probability of occurrence of the collective consciousness index W, k i The larger the value, the higher the activity of the group.
[0135] Although the above example shows i=1 to 3 indicating the number of cliques, the number of cliques i may be 4 or more.
[0136] <Collective Consciousness Analysis Processing> 10 is a flow chart showing an example of processing carried out by the collective consciousness analysis unit 13. The collective consciousness analysis unit 13 acquires the synchronous network time series data 24 generated by the network analysis unit 12 (S211).
[0137] The collective consciousness analysis unit 13 extracts a predetermined number (i) of cliques from the synchronous network time series data 24 and calculates the occurrence probability p j is calculated, and occurrence probability data 25 is generated for each specified time interval (S212).
[0138] The group consciousness analysis unit 13 calculates the occurrence probability p from the occurrence probability data 25 for each predetermined number of cliques. j Calculate the sum of the occurrence probabilities k i is calculated for each specified time interval (S213). If the number of cliques is 1, the total occurrence probability k 1 If the number of cliques is 2, the total probability of occurrence k 2 If the number of cliques is 3, the total probability of occurrence k 3 Calculate.
[0139] Next, the group consciousness analysis unit 13 calculates the sum of the occurrence probabilities k for each clique number i. i Information entropy H for each time interval specified from i The entropy data 30 stores the information entropy H i is stored.
[0140] When the number of cliques is 1, the information entropy H 1 If the number of cliques is 2, the information entropy H 2 If the number of cliques is 3, the information entropy H 3 Calculate.
[0141] In addition, the collective consciousness analysis unit 13 calculates the information entropy Hi When the number of cliques is 1, the information entropy H 1 (max), and if the number of cliques is 2, calculate the information entropy H 2 (max), and if the number of cliques is 3, the information entropy H 3 Calculate (max).
[0142] The collective consciousness analysis unit 13 calculates the information entropy H i (max) and the information entropy H i The ratio H i / H i (max) ratio minus 1 (1-H i / H i (max) is calculated as an index representing diversity (hereinafter, diversity index) for each specified time interval (S222).
[0143] The collective consciousness analysis unit 13 calculates the sum of occurrence probabilities k for each type of clique i in a specified time interval. i The pair of the diversity index and the diversity index is plotted in a two-dimensional space as a collective consciousness index W to generate a plot diagram, which is stored in the collective consciousness graph 40 (S23). The generated plot diagram is displayed on the display of the input / output device 56 by the display unit 14. The server 5 may transmit the plot diagram to another computer connected via the network 4 to display it.
[0144] By the above process, the information entropy H i and the information entropy H of the random network i (max) is used as the index of diversity for each specified time interval. Also, the sum of the occurrence probabilities of cliques for each specified time interval, k iis calculated as activity, and a pair of diversity index and activity for each specified time interval is generated as a collective consciousness index W.
[0145] Then, the group consciousness analysis unit 13 plots pairs of activity and diversity index for each specified time interval in a two-dimensional space, thereby generating a plot diagram showing the consciousness of the group.
[0146] When generating plot diagrams for a plurality of clique numbers, the processes in steps S211 to S23 may be executed in parallel, or may be executed sequentially for each clique number.
[0147] <Plot diagram> FIG. 11 is a plot 400 showing group consciousness. The plot 400 has a diversity index (1-H i / H i (max)), and the horizontal axis is the activity (sum of occurrence probability k i ) plotted against the time series of synchronous network time series data 24.
[0148] The vertical axis characteristic is that the closer the diversity index is to 1, the more fixed the cliques that appear. On the other hand, the closer the diversity index is to 0, the more equal the probability of cliques appearing. The horizontal axis characteristic is the activity (sum of appearance probabilities k i As the activation level increases, the activity of the group becomes more active. In this embodiment, the synchronization of body movements is treated as unconscious communication.
[0149] The characteristics of each region of the plot 400 are as follows: In a region 401 where the diversity index is close to 1 and the activity level is close to 0, certain individuals 60 are active or all of the individuals 60 are inactive, activity (communication) within the group is low, and the diversity of relationships between the individuals 60 is low.
[0150] In an area 402 where the diversity index is close to 1 and the activity level is high, all the individuals 60 are active and activity (communication) within the group is vigorous, but the diversity of relationships between the individuals 60 is low.
[0151] In an area 403 where the diversity index is close to 0 and the activity level is also close to 0, certain individuals 60 are active or all individuals 60 are inactive, and activity (communication) within the group is low, but the diversity of relationships between the individuals 60 is high.
[0152] In an area 404 where the diversity index is close to 0 and the activity level is high, all the individuals 60 are active, activity (communication) within the group is vigorous, and the relationships between the individuals 60 are highly diverse.
[0153] Fig. 12 is a plot diagram 400 showing the consciousness of a group. From the characteristics of the areas 401 to 404 shown in Fig. 11, when the plot diagram 400 is created for a conscious group, values are often plotted in the illustrated area 410.
[0154] In FIG. 12, this is an area spanning areas 401 and 404, and the position at which it is plotted changes over time between a state in which all individuals 60 are inactive and have low diversity (area 401) and a state in which all individuals 60 are active and have high diversity (area 404).
[0155] 13 is a plot 400 of a population analysis of an athletic field. The illustrated example is a three-clique plot 400 showing the population consciousness of two athletic teams of individuals 60 over time.
[0156] In team sports, it is said that "when one team is pushed back by the opposing team, organizational movements become monotonous," and in the example shown, the values of the underdog team B are concentrated in the upper left area of the figure, i.e., area 401 in Figure 11, visualizing that all the players are inactive and communication (diversity) within the team is low. In a situation where one team is being pushed back by the opposing team, it is thought that the network structure has no choice but to become monotonous in order to maintain information transmission within the team.
[0157] Although an example of a plot diagram 400 with three cliques was shown above, a similar tendency can be seen in a plot diagram with two cliques, making it possible to visualize the consciousness of a group over time.
[0158] As described above, in this embodiment, the total occurrence probability k i and the information entropy H i By calculating at each specified time interval and pairing it with the collective consciousness index W, it is possible to quantify how far the analyzed group is from a random network.
[0159] Then, the group consciousness analysis unit 13 plots the group consciousness index W for each clique type i on a two-dimensional plane, making it possible to represent the activity (= group temperature) on the horizontal axis of the graph and the complexity (= diversity) of the network structure on the vertical axis. EXAMPLES
[0160] Fig. 14 shows a second embodiment, which is a diagram showing an example of a network analysis system applied to the field of education. In the first embodiment, an example is shown in which the group consciousness index W is applied to unconscious communication such as synchrony of body movements. In the second embodiment, an example is shown in which the group consciousness index W is applied to conscious communication such as face-to-face communication and conversation.
[0161] In the second embodiment, a sensing device 1A, which is the sensing device 1 of the first embodiment plus an infrared communication unit 36, is worn by an individual 60, and a face-to-face meeting (visit) between the individuals 60 is detected and transmitted to the server 5 for storage in the sensor data 22.
[0162] In the server 5, the network analysis unit 12 generates a meeting matrix 23 of the meeting individuals 60 for each specified time interval from the sensor data 22, and generates communication network time series data 24A from the meeting matrix 23.
[0163] The server 5 calculates the group consciousness index W consisting of a pair of a diversity index and activity level for each number of cliques from the communication network time series data 24A in the same manner as in the first embodiment, and generates a plot diagram.
[0164] This embodiment differs from the first embodiment in that the network analysis unit 12 generates communication network time series data 24A from a meeting matrix 23 instead of the synchronization network time series data 24 of the first embodiment, but the other configurations are the same as those of the first embodiment.
[0165] In addition to the contents of the first embodiment, the ID of the person (individual 60) detected by the infrared communication unit 36 is added to the sensor data 22. Calculation of the body movement frequency and the like are the same as in the first embodiment.
[0166] 15 and 16 are diagrams showing an example of the relationship between the meeting matrix 23 and the communication network time-series data 24A at times T1 and T2.
[0167] 15 and 16, ID231 and ID232 indicate identifiers (user IDs) of individuals 60 constituting a group, and the body movement frequency of each individual 60 is stored. Network analysis unit 12 can determine that an individual 60 whose body movement frequency is equal to or greater than a predetermined threshold value (e.g., 2 Hz) is in a speaking state (activated).
[0168] At time T1 in FIG. 15, an individual 60 with user ID=1 is in a speaking state, conversing with users ID=2 and 4 who are in a face-to-face state, and user ID=4 is speaking to user ID=2.
[0169] At time T2 in FIG. 16, an individual 60 with user ID=1 is in a speaking state, conversing with users ID=2 and 5 who are in a face-to-face state, and user ID=5 is speaking to user ID=4.
[0170] The network analysis unit 12 generates communication network time series data 24A at specified time intervals from the above-mentioned meeting matrix 23. Note that the generation of the communication network time series data 24A is not limited to the above, and any publicly known or well-known technology may be applied. For example, as disclosed in the above-mentioned Patent Document 2, the communication network time series data 24A can be created by a spring model, with each individual 60 as a node and the meeting time as an attractive force between the nodes.
[0171] 17 is a plot 400 of a group analysis in the field of educational activities. The illustrated example is a plot 400 of one clique that shows the collective consciousness of individuals 60 in two groups engaged in group learning over time.
[0172] The illustrated example is a plot diagram showing the results of group learning between two groups, Group A and Group B. Group A is distributed from the center to the right side of the plot diagram 400, and is always able to engage in conscious group activities.
[0173] In contrast, Group B distributes temporarily to the right side of plot 400, but the majority of the group is distributed in the lower left part of plot 400, making it visible that the conscious group activity is temporary.
[0174] As described above, according to the second embodiment, by applying the group consciousness index W to conscious communication such as face-to-face meetings and conversations, it becomes possible to quantify and visualize the consciousness of a group, and this can be reflected in group management, etc.
[0175] <Conclusion> As described above, the network analysis systems of the first and second embodiments can be configured as follows.
[0176] (1) A network analysis method in which a computer (server 5) including a processor (CPU 51) and a memory (54) analyzes a network between individuals (60) in a population, the method comprising: a sensing data collection step (information input unit 53) in which the computer (5) collects time-series information on the activities of the individuals (60) constituting the population and stores the information in sensor data (22); a network analysis step (network analysis unit 12) in which the computer (5) extracts the activated individuals (60) from the sensor data (22) as cliques and generates a network (synchronous network time-series data 24) composed of the cliques in time series; and a network analysis step (network analysis unit 12) in which the computer (5) calculates a sum k of the occurrence probabilities of the cliques. i and the information entropy of the clique H i and a group consciousness analysis step (group consciousness analysis unit 13) of calculating an index showing the activity level of the group and the diversity of the group as a group consciousness index (W).
[0177] With the above configuration, by quantifying the index showing the activity level and diversity of a group as the group consciousness index W, it becomes possible to visualize the changes over time in the network structure of a group that is acting with a purpose, making it possible to easily grasp the consciousness of the group.
[0178] (2) In the network analysis method according to (1) above, the group consciousness analysis step (13) calculates the sum (k i ) as the activity of the individual; and calculating the information entropy of the network in a random state of the individual (60) as a first information entropy (H i (max)) and calculating the information entropy of the network of the individual (60) at each of the predetermined times as a second information entropy (H i ) and calculating the first information entropy (H i (max)) and the second information entropy (H i) as a diversity index for each time point; and calculating a pair of the activity level and the diversity index as a collective consciousness index (W) for each specified time interval.
[0179] With the above configuration, it is possible to calculate a collective consciousness index W that quantifies the extent to which the network structure of a group acting with a purpose differs from a network in which connections are uniform and random.
[0180] (3) The network analysis method according to (2) above, wherein the computer (5) further includes a visualization step (display unit 14) for visualizing the collective consciousness index (W), the collective consciousness analysis step (13) further includes a step of generating a plot diagram (400) in which the collective consciousness index (W) for each specified time interval is plotted in a predetermined space, and the visualization step (14) outputs the plot diagram (400).
[0181] With the above configuration, it is possible to easily grasp how consciousness, which is made up of the activity level and diversity of a group, has changed by using a plot diagram in which pairs of activity level and diversity index are plotted.
[0182] (4) The network analysis method according to (1) above, wherein the group consciousness analysis step (13) calculates the group consciousness index (W) for each number (i) of cliques.
[0183] With the above configuration, the group consciousness index W can be easily understood by selecting the number of cliques (i) in the plot diagram 400 according to the characteristics of the group to be analyzed. For example, in group sports such as soccer or rugby, the group consciousness index W can be easily understood by a plot diagram 400 with two or three cliques. Also, in the fields of education and business, the group consciousness index W can be easily understood by a plot diagram 400 with one or two cliques.
[0184] (5) In the network analysis method according to (2) above, the group consciousness analysis step (13) is a method for calculating 1-2nd information entropy (H i ) / First information entropy (H i (max)) at each specified time interval as the diversity index.
[0185] With the above configuration, the information entropy (H i / H i (max)) and the information entropy H i It is then possible to calculate an index showing the diversity (structure) of a population.
[0186] (6) The network analysis method according to (1) above, wherein the sensing data collection step (53) collects sensor data (22) including acceleration data of the individuals (60) as time-series information on the activity of the individuals (60), and the network analysis step (12) calculates relationships between the individuals (60) from the acceleration data to generate a network (24) composed of the cliques in time series.
[0187] With the above configuration, when analyzing a group in a sporting field or the like, a group consciousness index W can be calculated according to the activity of the group by generating synchronous network time series data 24 based on acceleration data.
[0188] (7) The network analysis method according to (1) above, wherein the sensing data collection step (53) collects sensor data (22) including face-to-face information between the individuals (60) as time-series information on the activities of the individuals (60), and the network analysis step (12) calculates relationships between the individuals (face-to-face matrix 23) from the face-to-face information to generate a network (communication network time-series data 24A) composed of the cliques in time series.
[0189] With the above configuration, when analyzing a group in the field of education, business, or the like, a group consciousness index W can be calculated according to the activity content of the group by generating communication network time series data 24A based on face-to-face information.
[0190] The present invention is not limited to the above-mentioned embodiment, and various modifications are included. For example, the above-mentioned embodiment is described in detail to easily explain the present invention, and is not necessarily limited to including all the configurations described. In addition, it is possible to replace a part of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add the configuration of another embodiment to the configuration of one embodiment. In addition, any addition, deletion, or replacement of other configurations can be applied to a part of the configuration of each embodiment, either alone or in combination.
[0191] In addition, the above-mentioned configurations, functions, processing units, processing means, etc. may be realized in part or in whole by hardware, for example, by designing them as integrated circuits. In addition, the above-mentioned configurations and functions, etc. may be realized in software by a processor interpreting and executing a program that realizes each function. Information such as the program, table, file, etc. that realizes each function can be stored in a memory, a recording device such as a hard disk or SSD (Solid State Drive), or a recording medium such as an IC card, SD card, or DVD.
[0192] In addition, the control lines and information lines shown are those that are considered necessary for the explanation, and not all control lines and information lines in the product are necessarily shown. In reality, it can be considered that almost all components are connected to each other. [Explanation of symbols]
[0193] 1. Sensing device 5 Server 10 Analysis Department 11 Feature Calculation Unit 12 Network Analysis Department 13. Collective Consciousness Analysis Department 14 Display section 21 Feature information 22 Sensor Data 24 Synchronous network time series data 25 Probability of occurrence data 30 Entropy Data 40 Mass Consciousness Graph 51 CPU 52 Memory 55 Auxiliary storage device
Claims
1. A network analysis method in which a computer including a processor and a memory analyzes a network between individuals in a population, comprising: a sensing data collection step in which the computer collects time-series information on the activities of individuals constituting the population and stores the information in sensor data; a network analysis step in which the computer extracts the individuals activated from the sensor data as cliques and generates a network composed of the cliques in a time series; and a group consciousness analysis step in which the computer calculates the occurrence probability of the clique and the information entropy of the clique, and calculates an index indicating the activity of the group and the diversity of the group as a group consciousness index.
2. 2. The network analysis method of claim 1, comprising: The collective consciousness analysis step includes: calculating the sum of the occurrence probabilities of the cliques for each specified time interval as the activity of the individual; Calculating an information entropy of a network in a random state of the individuals as a first information entropy; calculating an information entropy of the network of individuals for each of the specified time intervals as a second information entropy; calculating a value based on a ratio between the first information entropy and the second information entropy as a diversity index for each designated time interval; calculating a pair of the activity level and the diversity index as a group consciousness index for each specified time interval; A network analysis method comprising:
3. 3. A network analysis method according to claim 2, comprising: The method further includes a visualization step of the computer visualizing the collective consciousness index, The collective consciousness analysis step includes: generating a plot of the collective consciousness index for each of the specified time intervals in a predetermined space; The visualization step includes: A network analysis method comprising the steps of: outputting the plot diagram.
4. 2. The network analysis method of claim 1, comprising: The collective consciousness analysis step includes: A network analysis method, comprising: calculating the collective consciousness index for each number of cliques.
5. 3. A network analysis method according to claim 2, comprising: The collective consciousness analysis step includes: 1-Second information entropy / First information entropy A network analysis method comprising: calculating the diversity index at each specified time interval.
6. 2. The network analysis method of claim 1, comprising: The sensing data collection step includes: collecting sensor data including acceleration data of the individual as time-series information regarding the activity of the individual; The network analysis step includes: A network analysis method comprising the steps of: calculating relationships between individuals from the acceleration data; and generating a network composed of the cliques in time series.
7. 2. The network analysis method of claim 1, comprising: The sensing data collection step includes: Collecting sensor data including face-to-face information between the individuals as time-series information regarding the activities of the individuals; The network analysis step includes: A network analysis method comprising the steps of: calculating relationships between individuals from the face-to-face information; and generating a network composed of the cliques in time series.
8. a server including a processor and a memory; A network analysis system including: The sensing device transmits time-series information regarding individuals performing activities in a group to the server; The server, a sensing data collection unit that collects time-series information on the individual received from the sensing device and stores the information in sensor data; a network analysis unit that extracts the activated individuals from the sensor data as cliques and generates a network composed of the cliques in a time series; a group consciousness analysis unit that calculates the occurrence probability of the clique and the information entropy of the clique, and calculates an index indicating the activity level of the group and the diversity of the group as a group consciousness index; A network analysis system comprising:
9. 9. The network analysis system of claim 8, The collective consciousness analysis unit: Calculating the sum of the occurrence probabilities of the cliques for each specified time interval as the activity of the individual; Calculating an information entropy of a network in a state in which the individuals are in a random state as a first information entropy; Calculating an information entropy of the network of the individuals for each of the specified time intervals as a second information entropy; calculating a value based on a ratio between the first information entropy and the second information entropy as a diversity index for each designated time interval; A network analysis system characterized in that a pair of the activity level and the diversity index is calculated as a group consciousness index for each specified time interval.
10. 10. The network analysis system of claim 9, A visualization unit that visualizes the collective consciousness index, The collective consciousness analysis unit: generating a plot of the collective consciousness index for each of the specified time intervals in a predetermined space; The visualization unit is A network analysis system that outputs the plot diagram.
11. 9. The network analysis system of claim 8, The collective consciousness analysis unit: A network analysis system, comprising: a network analysis unit that calculates the collective consciousness index for each number of cliques;
12. 10. The network analysis system of claim 9, The collective consciousness analysis unit: 1-Second information entropy / First information entropy A network analysis system comprising: a network analysis unit that calculates the diversity index at each specified time interval.
13. 9. The network analysis system of claim 8, The sensing data collection unit includes: collecting sensor data including acceleration data of the individual as time-series information regarding the activity of the individual; The network analysis unit A network analysis system characterized by calculating relationships between individuals from the acceleration data and generating a network composed of the cliques in time series.
14. 9. The network analysis system of claim 8, The sensing data collection unit includes: Collecting sensor data including face-to-face information between the individuals as time-series information regarding the activities of the individuals; The network analysis unit A network analysis system characterized by calculating relationships between individuals from the face-to-face information and generating a network composed of the cliques in time series.
15. A server including a processor and a memory, a sensing data collection unit that collects time-series information on individuals performing activities in a group and stores the information in sensor data; a network analysis unit that extracts the activated individuals from the sensor data as cliques and generates a network composed of the cliques in a time series; a group consciousness analysis unit that calculates the occurrence probability of the clique and the information entropy of the clique, and calculates an index indicating the activity level of the group and the diversity of the group as a group consciousness index; A server comprising:
Citation Information
Patent Citations
Sensor network system for measuring quality of communication, management computer and method
JP2012027589A
Causal network generating system and data structure for causal relationship
JP2015153133A
Analysis system and analysis method
JP2017000481A
Network analysis method, server and network analysis system
JP2018081406A
Contribution management server, and contribution management system
JP2021064018A
