JP5003536B2 - Service support system - Google Patents

Description

  The present invention relates to a service support system that analyzes an action of a user existing in a predetermined area and presents an appropriate service according to a situation.

  For example, the service support system analyzes the situation of customers staying in the store, notifies the store clerk of the current position of the person who moves unusually, or provides a guidance message to the corresponding customer itself System.

FIG. 2 is a configuration diagram of a conventional service selection support system described in Patent Document 1 below.
This service selection support system includes a sensor information acquisition unit 10, an interpretation unit search unit 11, an interpretation unit database 12, a user context interpretation execution unit 13, a user context condition generation unit 14, a service database search unit 15, a service database 16, and a service use A history database 17 and a human relationship database 18 are included.

  The sensor information acquisition unit 10 acquires sensor information obtained by sensors arranged in the environment. The interpretation means retrieval unit 11 retrieves interpretable user contexts from the interpretation means database 12 in which information necessary for interpretation is described. The user context interpretation execution unit 13 interprets interpretable user contexts according to interpretation means. The user context condition generation unit 14 generates a user context grasped by each interpretation unit and sensor information acquired by the sensor information acquisition unit 10 as user context conditions. The service database 16 describes service presentation conditions for each service when the service is presented to the client terminal.

  The service database search unit 15 determines each service presentation condition of the service database 16 based on the user context condition generated by the user context condition generation unit 14 and the search request input to the client terminal, and the search request and the context condition Is provided to the client terminal as a search result. The client terminal includes a service usage history database 17 and a human relationship database 18.

  With such a configuration, if the sensor information necessary for interpreting the user context can be acquired, the interpretation is performed according to the interpretation means, and this is used as a condition for service search. Thus, the user context can be interpreted according to the difference in environment and used as a service search condition, so that efficient service selection and use can be performed in that environment.

JP 2006-59094 A

  However, in the service selection support system, interpretation means and sensor information are held in a database, and user context interpretation and service search are performed. In such a service, a more detailed service for each user will be provided in real time. In this case, it is necessary to analyze the individual behavior specific to the user in real time. However, when individual actions of all users are analyzed in detail, the amount of data processing becomes enormous, and there is a problem that a time lag occurs before feedback of the analysis results.

  The service support system of the present invention reduces the amount of data processing by analyzing in detail only actions that deviate from the average user action as individual actions, and provides a detailed service for each user in real time without delay. It is aimed.

  According to the present invention, a service support system that provides an appropriate service according to a user's situation existing in a predetermined area is detected by the sensor information acquisition unit for detecting the user's situation and the sensor information acquisition unit. A common rule application unit that presents a service corresponding to the user according to the user status and a predetermined common determination criterion, an average user status detected by the sensor information acquisition unit, and each user's An information analysis / analysis unit that compares individual situations and generates individual determination criteria for individual users whose difference exceeds a preset range, and the individual user situations and the individual users An individual rule application unit that presents a service corresponding to the individual user in accordance with the individual determination criteria, and the individual rule application unit for the individual user Providing services, it is characterized by being configured by the service providing unit that provides the services that are presented by the common rule application unit to the user other than the individual by the user.

  In the present invention, an information analysis / analysis unit that compares an average situation of users with individual situations of each user and generates individual determination criteria for individual users whose difference exceeds an allowable range is generated. And an individual rule application unit that presents a service corresponding to an individual user according to the individual determination criteria. As a result, it is not necessary to analyze the individual behavior of all users in detail, but it is only necessary to analyze the individual behavior only for users whose situation exceeds the permissible range. Can be provided in real time.

  The above and other objects and novel features of the present invention will become more fully apparent when the following description of the preferred embodiment is read in conjunction with the accompanying drawings. However, the drawings are for explanation only, and do not limit the scope of the present invention.

  The present invention extracts average customer behavior by customer behavior analysis, and provides services of contents common to all customers for average customer behavior obtained by statistical methods, The behavior that deviates from the average customer behavior is analyzed in real time as individual behavior specific to the customer.

  In addition, customer behavior analysis is a process of comparing and examining what kind of process the customer picked up at the sales floor and what they decided to purchase in order to elucidate the reason why the product sold. This refers to collecting information such as whether it was targeted by tracking (behavior tracking survey), etc., and performing analysis such as extracting average customer behavior using statistical methods.

FIG. 1 is a configuration diagram of a service support system showing an embodiment of the present invention.
This service support system includes a sensor information collection unit 1, a rule determination unit 2, a common rule application unit 3, an information analysis / analysis unit 4, an individual rule application unit 5, and a service provision unit 6.

  The sensor information collection unit 1 collects, in real time, location information of all users (customers) existing in a predetermined area such as a department store sales area and manages it as a database. As a method for acquiring user position information, for example, there is a method in which a monitoring area is photographed with a video camera or the like, and the positions of individual persons are tracked by image recognition or the like at regular intervals (for example, at intervals of 1 second). In addition, when it is approved to wear an admission card or the like, there is a method in which a wireless chip is incorporated into the admission card or the like to measure the radio wave intensity, the arrival time difference of radio waves, and the like to obtain the user's position information.

  The rule determination unit 2 rules conditions for determining the type and content of user behavior based on the user location information collected by the sensor information collection unit 1 and behavior tracking data generated from the user location information. Is used to determine whether the information collected by the sensor information collection unit 1 matches the rule. Regarding the rules for determination, common rules that apply to all users and individual rules with different contents are provided for each user, and users who have individual rules perform rule determination processing according to the individual rules. .

  The common rule application unit 3 applies a common rule to a user who matches the conditions of the common rule by the rule determination unit 2 and selects an appropriate service content. The common rule is applied to all users. However, the individual rule is applied with priority over the common rule for the user for which the individual rule is set.

  The information analysis / analysis unit 4 detects an action slightly deviating from the aggregated average action as a user's individual action, performs analysis / analysis, and generates / updates an individual rule for each user in real time. is there.

  The individual rule application unit 5 applies an individual rule with priority over a common rule to a user who has an individual rule, and selects an appropriate service content.

  The service providing unit 6 applies the service content selected based on the applied individual rule to the user to which the individual rule is applied by the individual rule applying unit 5 and the average of the individual rule not applied. The service contents selected based on the common rule by the common rule application unit 3 are provided to the user.

  FIG. 3 is a flowchart of the process showing the operation of FIG. 1, and FIG. 4 is an explanatory diagram showing a process example of the rule determining unit 2 in FIG. Hereinafter, the operation of FIG. 1 will be described with reference to FIGS. 3 and 4.

  Here, it is assumed that the common rule is “determine that a user staying in area A for 5 minutes or more is staying and notify the store clerk of the current position of the staying user”. Regarding the individual rules, two rules, the individual rule 1 and the individual rule 2, will be described.

  The individual rule 1 states that “10% of the ratio of users determined to be staying is an error range, and a user whose stay time is slightly less than 5 minutes within this error range is subject to the individual rule and is determined to be staying. It shortens the staying time. The individual rule 2 is that the error range is 10% of the ratio of users determined to be staying, the user whose stay time is slightly less than 5 minutes within this error range, and the moving speed is faster than 4 km / h. The user is subject to the individual rule, and the staying time for judging the stay is shortened.

  When the process is started, first, sensor information is acquired by the sensor information collecting unit 1 of FIG. 1 in step S1 of FIG. Position information is used as the sensor information, and the position information of each user is acquired at a constant cycle. For example, as shown in FIG. 4, the position of each user existing in a 30 m square monitoring area is acquired as xy coordinate position data at 1 second intervals, and this is held together with the user ID (identification number). Note that the user ID is given for convenience in image recognition or the like except when it is obtained from a wireless chip or the like. The acquired information, except for some information that needs to be referred to later, retains only information for a certain past period required for behavior analysis, and unnecessary information is sequentially discarded.

In step S2, the rule determination unit 2 determines whether the user is staying.
The stay determination is performed by dividing the monitoring area into 15 m square areas A, B, C, and D, and detecting a user who stays in one area, for example, for 5 minutes or more. For example, as shown in FIG. 4, the position information of the user ID = 0 acquired at 1 second intervals is continuously 0 ≦ x <1500 (cm), 0 ≦ y <1500 for 5 minutes (300 seconds) or more. If the condition (cm) is satisfied, it is determined that the user stays in the area A.

  The rules for determination may be a common rule that applies to all users, or an individual rule with different contents for each user. For users with individual rules, the individual rule According to the rule determination process. As a result of the rule determination, if the rule is met, the process proceeds to step S3, and if not, the process proceeds to step S4.

  In step S3, the common rule is applied. For example, as a common rule, a rule of a determination threshold value is applied in which, when staying in the area A for 300 seconds or longer, the store clerk is notified of stay. The common rule is applied to all users, but an individual rule may be set for each user. For users who fall under the individual rule, the rule is applied with priority on the individual rule. After step S3, the process proceeds to step S8.

  In step S4, the information analysis / analysis unit 4 performs sensor information information analysis / analysis processing. For example, as shown in FIG. 4, it is constant based on the frequency of users determined to stay in area A (that is, the ratio of users who have stayed for 300 seconds or longer, in this example 3.81%). The error range (that is, 10% of the stay frequency, in this case 0.38%) is obtained, and a user whose stay time is slightly less than 300 seconds within this error range is detected. That is, information on the top 0.38% of users whose area A stay time is less than 300 seconds is detected.

  Furthermore, as shown in FIG. 4, based on the detected information, as an analysis of individual behavior, for example, the difference between the residence time grouped in units of 5 seconds and the determination threshold (300 seconds) (that is, the time error is 1 to 5). Second group and 6-10 second group) and the average moving speed based on the moving distance for 1 second (4 km / h), and the threshold value for stay determination is changed for each user according to the calculation result (ie, In the group having a time error of 1 to 5 seconds and the group having a time error of 6 to 10 seconds, only the group whose moving speed is faster than 4 km / h can be changed).

  In step S5, based on the analysis result of the information analysis / analysis unit 4, the individual rule application unit 5 determines whether or not to update the individual rule for each user. Here, according to individual rule 1, the top 0.38% of users who have stayed in area A for less than 300 seconds (ie, groups with a time error of 1 to 5 seconds and groups with 6 to 10 seconds) The process proceeds to step S6 as a generation / update target. Other users are not targeted for individual rule application, and the process ends.

  In step S6, the individual rule application unit 5 generates an individual rule for each user according to the difference between the staying time and the determination threshold. Further, in step S7, the individual rules are updated for each user.

  For example, as shown in FIG. 4, a user who is a target of generation / update of the individual rule 1 (a user of a group with a time error of 1 to 5 seconds and a user with a group of 6 to 10 seconds: user ID = 2, 5, 7 , 10,...), The determination threshold of a user with a time error of 1 to 5 seconds (user ID = 2, 10,...) Is changed from 300 seconds to 295 seconds, and a user with a time error of 6 to 10 seconds. The determination threshold of (user ID = 5, 7,...) Is changed from 300 seconds to 290 seconds.

  Alternatively, according to the individual rule 2, it is also possible to update only users with a high average moving speed and not update users with a low average movement speed. For example, as shown in FIG. 4, among the target users (user ID = 2, 5, 7, 10,...), Users (users) whose average moving speed is faster than 4 km / h and whose time error is 1 to 5 seconds. ID = 2,...) Is changed from 300 seconds to 295 seconds, and the determination threshold of a user (user ID = 7,...) Whose average moving speed is higher than 4 km / h and the time error is 6 to 10 seconds is 300. Change from seconds to 290 seconds. After step S7, the process proceeds to step S8.

  In step S8, the corresponding rule is applied to each user. For all users, the common rule of notifying the store clerk if the stay time in area A is 300 seconds or more is applied, but the user whose determination threshold is changed by the individual rule (for example, user ID 2 295 seconds), the changed individual rule is preferentially applied.

  In step S9, a service is provided to the user based on a preset rule. In this example, the store clerk is notified of the user ID of the user staying in the area A and the current position information. Thereby, the salesclerk who received the notification can perform appropriate guidance, assistance, or the like for the corresponding user.

  As described above, the service support system according to the present embodiment includes the information analysis / analysis unit 4 that analyzes and analyzes the behavior deviating from the average customer behavior as the individual behavior specific to the customer, and the individual rule according to the analyzed individual behavior. It has the individual rule application part 5 which applies. As a result, services are provided based on common rules for average user behavior, and individual rules are set for users who take behavior that deviates from average user behavior. Can be provided. In this case, since the frequency of users who take an action that deviates from the average user action is low, it becomes possible to significantly reduce the data processing amount compared to the case where the individual actions of all users are analyzed in detail, There is an advantage that a preference / prediction model and a behavior model are generated for each user in real time, and a detailed service can be provided in real time for each user.

In addition, this invention is not limited to the said Example, A various deformation | transformation is possible. Examples of this modification include the following.
(A) Although the user's position information is used as the sensor information collection unit 1, in addition to the position information according to the service content, temperature, humidity, rainfall, volume, altitude, acceleration, inclination angle, weight, pressure Various information such as vibration, shock, barcode, IC tag, fingerprint, vein pattern, gas leak, smoke, heat, radiation, chemical substance, etc. can be acquired and used.
(B) As a sensor for collecting information, in addition to a position sensor, a temperature sensor, a humidity sensor, a rainfall sensor, a sound sensor, an altitude sensor, an acceleration sensor, an inclination sensor, a weight sensor, a pressure sensor in addition to a position sensor. Vibration sensors, impact sensors, barcode readers, IC tag readers, fingerprint sensors, vein pattern sensors, gas leak sensors, smoke sensors, flame sensors, thermal sensors, radiation sensors, chemical substance sensors, and the like can be used as appropriate.
(C) The rule determination unit 2 individually determines the staying state of the user in the area A according to the exemplified common rule and the individual rules 1 and 2, but these rules are examples. For example, it is possible to determine the number of users existing in the area A, or to determine environmental conditions such as the temperature, humidity, and volume of the area A.
(D) In the information analysis / analysis unit 4, the error range slightly deviated from the aggregated average behavior is detected as 10% as an individual action of the user, but the setting of the error range is arbitrary. The error range may be set with an absolute value instead of a relative ratio. By increasing the error range, more analysis / individual rule generation is performed and more detailed service can be realized, but on the other hand, there is a trade-off relationship that real-time property is lowered.
(E) As an analysis method of individual user behavior, in addition to the average moving speed, it is also possible to analyze the degree of congestion by measuring the number of people in area A, and to analyze the movement route that enters and exits area A. . Furthermore, it is also possible to generate / update individual rules by analyzing in various ways in cooperation with user attribute information on the database using various sensor information and user IDs as keys.
(F) The service providing unit 6 notifies the long-staying user information to the clerk in charge of providing the service in the area instead of the user himself / herself, but the service content is not limited to this. For example, when a user carries a terminal, it is possible to output appropriate information (map, advertisement, product information, navigation, etc.) to the terminal. It is also possible to output to a terminal or the like installed in the area or the periphery.

It is a block diagram of the service assistance system which shows the Example of this invention. It is a block diagram of the conventional service selection assistance system. It is a flowchart of the process which shows the operation | movement of FIG. It is explanatory drawing which shows the process example of the rule determination part 2 in FIG.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Sensor information collection part 2 Rule determination part 3 Common rule application part 4 Information analysis and analysis part 5 Individual rule application part 6 Service provision part

Claims (8)

A service support system for providing an appropriate service according to a situation of a user existing in a predetermined area,
A sensor information acquisition unit for detecting the user's situation;
A common rule application unit that presents a service corresponding to the user according to the user status detected by the sensor information acquisition unit and a predetermined common criterion;
Information that compares the average situation of users detected by the sensor information acquisition unit with the individual situation of each user, and generates individual determination criteria for individual users whose difference exceeds a preset range Analysis / analysis department,
An individual rule application unit that presents a service corresponding to the individual user according to the status of the individual user and the individual determination criterion generated for the individual user;
Providing a service presented by the individual rule application unit to the individual user and providing a service presented by the common rule application unit to the user other than the individual user;
A service support system characterized by comprising.   The sensor information acquisition unit periodically acquires position information of each user recognized based on image information captured by the camera, position information of each user grasped wirelessly, or position information obtained by a combination thereof. The service support system according to claim 1.   The said common rule application part grasps | ascertains the said user's condition by hold | maintaining the positional information acquired periodically by the said sensor information acquisition part continuously based on progress of time. The service support system described.   The said individual rule application part grasps | ascertains the situation of the said individual user by hold | maintaining the positional information acquired periodically by the said sensor information acquisition part continuously based on progress of time. 2. The service support system according to 2.   The said common rule application part grasps | ascertains the condition of this user based on the sensor information acquired by the sensor with which the user himself was mounted | worn, or the sensor installed in the position around a user. The service support system described.   The information analysis / analysis unit generates the individual determination criterion only for an individual user whose difference exceeding the preset range among the individual users is smaller than a preset threshold value. The service support system according to any one of claims 1 to 5.   The information analysis / analysis unit obtains a ratio of generating the individual determination criteria based on a ratio of the users in the preset range among the individual users, and sets the preset range in the users. The service support system according to any one of claims 1 to 5, wherein the individual determination criterion is generated only for an individual user having a smaller difference from the user.   8. The service provider according to claim 1, wherein the service provider provides the service to a terminal carried by the service provider or the user who is in charge of the predetermined area or a terminal installed around the service provider. 8. Service support system described in 1.

Images