JP2016024599A - Information management server, information management method, and information management program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an information management server, an information management method, and an information management program capable of accurately grasping the latent needs of merchandise without imposing any labor.SOLUTION: An information management server 20 includes: sensor data acquisition means 21 for acquiring sensor data output from sensors 11 to 16 for detecting the state of a user; history data acquisition means 22 for acquiring history data in which the action of the user on the Internet for merchandise is stored; contribution value acquisition means (coefficient setting means 26) for acquiring a contribution value that the state of the user gives to a user's degree of interest in the merchandise on the basis of the history data; and interest degree calculation means 23 for calculating the degree of interest on the basis of the sensor data and the contribution value.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an information management server, an information management method, and an information management program for managing information.

In recent years, it has been required to grasp the potential needs of users for products. It is conceivable to use a purchase history in order to obtain the latent needs. However, since the price of the product greatly affects the purchase history, the latent needs cannot be obtained accurately based on the purchase history.
For this reason, conventionally, for example, a questionnaire regarding products on the web has been conducted to find potential needs (see, for example, Patent Document 1).

Japanese Patent Laid-Open No. 10-171880

  However, there is a problem that the questionnaire takes time and effort for counting. Further, in the questionnaire on the web, the aggregated result also changes depending on the preconception of the user, not the evaluation when the actual product is seen or touched, so that it is not always possible to grasp the potential needs with high accuracy.

  An object of the present invention is to provide an information management server, an information management method, and an information management program capable of accurately grasping the latent needs of products without taking time and effort.

  The information management server of the present invention includes a sensor data acquisition unit that acquires a plurality of types of sensor data output from a sensor that detects a user's state, and a history that acquires history data that accumulates user actions on the Internet for products. Based on the data acquisition means, the contribution value acquisition means for acquiring the contribution value that the user's state gives to the evaluation of the user's product based on the history data, the plurality of types of sensor data, and the contribution value And an evaluation calculation means for calculating the evaluation.

  According to the present invention, a user's evaluation for a product is calculated using sensor data and history data on the Internet. For example, based on the sensor data, it is possible to determine the user's behavior and reaction such as the user standing in front of the product or picking up the product, that is, real data that is the actual user's behavior or reaction. Further, by acquiring history data that accumulates user actions on the Internet, it becomes possible to determine user actions, impressions, and the like on the Internet. Therefore, a highly accurate evaluation can be calculated by calculating the evaluation based on these sensor data and history data on the Internet by the evaluation calculation means. At this time, since sensor data and data on the Internet are used compared to a complicated method using a questionnaire or the like, the evaluation can be calculated without taking time and effort.

It is a figure which shows the structure of the information management system which concerns on 1st Embodiment of this invention. It is a flowchart which shows the calculation process of the positive degree in the said embodiment. It is a radar chart which shows an example of the total result which totaled each positive degree in the said embodiment. It is a flowchart which shows the calculation process of the coefficient in the said embodiment. It is a figure which shows the structure of the information management server which concerns on 3rd Embodiment of this invention. It is a figure which shows the structure of the information management server which concerns on the modification 2 of this invention.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
[First Embodiment]
FIG. 1 is a diagram illustrating a configuration of an information management system according to the first embodiment.
As shown in FIG. 1, the information management system 1 includes a sensor group 10 that can be connected to a network 2, an information management server 20, a product management server 30, and a history data management server 40.

[Sensor group]
The sensor group 10 is arranged in a store where merchandise is displayed.
The sensor group 10 includes an infrared sensor 11, an acceleration sensor 12, a voice recognition sensor 13, a thermo sensor 14, an electroencephalogram sensor 15, and a camera 16. The infrared sensor 11, the voice recognition sensor 13, the thermo sensor 14, and the camera 16 are installed around the target product. The acceleration sensor 12 is attached to a target product, and the electroencephalogram sensor 15 is attached to a user in the store.
Each sensor 11-16 is controlled by the information management server 20, detects the state of the user who has approached the target product, and transmits sensor data corresponding to the detection result to the information management server 20 via the network 2 (output) )
As shown in Table 1, each sensor 11 to 16, it transmits the sensor data _V 1 ~V _7.

Infrared sensor 11, using infrared, measures the time (Tachidomari time) the user has stopped in front of the product, and transmits the measured Tachidomari time as sensor data V ₁ to the information management server 20.

The acceleration sensor 12 measures the acceleration of the goods, to the information management server 20 the acceleration measured as the sensor data V _2.

The speech recognition sensor 13 includes a speech recognition unit that recognizes speech and a speech determination unit that determines recognized speech. Voice recognition sensor 13 extracts and negative keywords affirmative keywords uttered by the user, transmitted from the number of positive keywords, a value obtained by subtracting the number of negative keywords, as the sensor data V ₃ to the information management server 20 To do. Incidentally, if the user does not issue a positive keywords and negative keywords, sensor data V ₃ is "0".
In addition, the voice recognition sensor 13 uses, for example, individual sensor data for voice presence / absence data (with voice: “1”, without voice: “0”), the number of positive keywords, and the number of negative keywords. May be transmitted to the information management server 20.

The thermosensor 14 includes a heat distribution detection unit that detects a heat distribution using infrared rays emitted from a user, and a heat distribution determination unit that determines the detected heat distribution.
Thermosensor 14, depending on the difference between the average value and the common value of the normal body temperature of heat distribution, to determine the excitement level (0,1,2 ...), information management excitement level determined as sensor data V ₄ Send to server 20.

The electroencephalogram sensor 15 includes an electroencephalogram detection unit that detects a user's electroencephalogram, and an electroencephalogram determination unit that determines the detected electroencephalogram.
EEG sensor 15, based on the brain wave to determine the excitement level (0, 1, 2), to the information management server 20 the excitement level determined as sensor data V _5.

The camera 16 includes a photographing unit that photographs a user and an image determination unit that determines a captured image.
The camera 16 analyzes the image of the user's face and digitizes the user's facial expression (if the facial expression changes to a positive facial expression such as a smile: “1”, changes to a negative facial expression such as a grumpy face) If: "- 1", if you did not change the expression: "0"), to send as sensor data V ₆ to the information management server 20.
The camera 16 analyzes the image obtained by photographing the eye of a user, the time the user's gaze has been directed to the product of (fixed time) is measured, the information management server gaze time measured as the sensor data V ₇ 20 to send.
For example, the camera 16 may analyze an image obtained by photographing the user's face and transmit the number of times the same user has approached the product to the information management server 20 as sensor data.
In addition, although the image image | photographed with the camera 16 is memorize | stored in the memory | storage device of the camera 16, a user's image is replaced by the similar character image at this time, for example. Thereby, a user's privacy can be protected.

  Note that the voice determination unit of the voice recognition sensor 13, the heat distribution determination unit of the thermosensor 14, the brain wave determination unit of the electroencephalogram sensor 15, and the image determination unit of the camera 16 may be provided in the information management server 20.

[Product Management Server]
The merchandise management server 30 is configured by a computer and includes a storage device such as an HDD (Hard Disk Drive) and an arithmetic circuit such as a CPU (Central Processing Unit).
In the product management server 30, management data for each product to be sold is registered. The management data includes a product code for identifying a product, a sales status (whether or not sold), a sales time, a store name, and the like.
When a product sales process is performed by a POS (Point Of Sales system) terminal (not shown) that can be connected to the network 2 installed in the store, the product management server 30 sells management data corresponding to the product. Update the sales time.

[History data management server]
The history data management server 40 is configured by a computer, and includes a storage device such as an HDD and an arithmetic circuit such as a CPU.
The history data management server 40 manages history data in which user actions on the Internet for products are accumulated. The history data includes the degree of buzz (topic frequency) for a product on the web, the writing (review) for the product, the number of searches for the product, and the like.

[Information Management Server]
The information management server 20 is configured by a computer and includes a storage device such as an HDD and an arithmetic circuit such as a CPU. As shown in FIG. 1, the arithmetic circuit of the information management server 20 reads and executes the information management program stored in the storage device, so that the sensor data acquisition unit 21, the history data acquisition unit 22, the interest degree calculation unit 23, It functions as a coefficient storage unit 24, an interest degree storage unit 25, and a coefficient setting unit 26.

The sensor data acquisition unit 21 controls the sensors 11 to 16 and acquires sensor data V _{1 to} V ₇ from the sensors 11 to 16.
The history data acquisition unit 22 acquires history data from the history data management server 40.

The interest level calculation means 23 calculates the user's interest level for the product. Here, the interest level calculation means 23 constitutes the evaluation calculation means of the present invention, and the degree of interest corresponds to the evaluation of the present invention.
In the present embodiment, as the interest level, for example, a stop positive degree Sps (T) indicating an interest level from the viewpoint of stopping, and a positive degree Sps (H) taken by a hand indicating an interest level from the viewpoint of being taken by the hand. ), Electroencephalogram positive degree Sps (N) indicating the degree of interest in terms of brain waves, line-of-sight positive degree Sps (S) indicating the degree of interest in terms of line of sight, and voice positive indicating the degree of interest in terms of speech The degree Sps (O) is exemplified.

Here, each positive degree Sps (X) (X = T, H, N, S, O) can be obtained by the following equation (1). In Expression (1), V _n represents a value of sensor data, W _n (X) represents a weighting coefficient corresponding to Sps (X) and V _n , and n is 1 to 7 Indicates the variable.

That is, Sps (X) is the total sum of values obtained by multiplying each of the sensor data V _{1 to} V ₇ by the corresponding weighting coefficient.
Here, the weighting factors W ₁ (X) to W ₇ (X) corresponding to each of the sensor data V _{1 to} V ₇ are set according to the contribution degree of the corresponding sensor data to Sps (X), and the coefficients It is stored in the storage means 24. Here, the weighting factors W ₁ (X) to W ₇ (X) correspond to contribution values of the present invention.
For example, the weighting factors W ₁ (T) to W ₇ (T) corresponding to the stationary positive degree Sps (T) are independent of the other positive degrees Sps (X) (X = H, N, S, O). Thus, it is set corresponding to each of the sensor data V _{1 to} V ₇ .
Here, the stationary positive degree Sps (T) will be described below.
Table 2 shows an example of the weight coefficients W ₁ (T) to W ₇ (T) stored in the coefficient storage unit 24.

As shown in Table 2, since the sensor data V ₁ indicating the user's stop time has a high contribution to the stop positive degree Sps (T), the weighting factor W ₁ (T) corresponding to the sensor data V ₁ is It is set to “10”, which is a relatively high value.
In addition, since the difference between whether the user stopped to see the product or just stopped affects Sps (T), here, it is possible to estimate that the user stopped with interest. The data weighting factor is also set relatively high. As such sensor data, there is sensor data V ₇ indicating a gaze time for the user's product. For this reason, the weighting factor W ₇ (T) corresponding to the sensor data V ₇ is set to “5”.
Note that the weighting factors W ₂ (T) to V ₆ (T) corresponding to the sensor data V _{2 to} V ₆ have a relatively low contribution to Sps (T), so here the values are less than “1”. Is set.
That is, in the example of Table 2, the stationary positive Sps (T) is “Sps (T) = 10 × V ₁ + 0.1 × V ₂ + 0.5 × V ₃ + 0.1 × V ₄ + 0.1 × V ₅ + 0.1 × V ₆ + 5 × V ₇ ”.

  Also, the interest level calculation means 23 adds the calculated positive degrees to calculate the total positive degree Sps (A). That is, the total positive degree Sps (A) can be obtained by the following equation (2). In Expression (2), X is “X = T, H, N, S, O”.

Here, the overall positive degree Sps (A) corresponds to the overall evaluation of the present invention.
Sps (A) may be obtained by weighting and adding each positive degree. For example, as shown in the following formula (3), each positive degree may be added by applying a preset coefficient. In Expression (3), α _X is a coefficient, and X is “X = T, H, N, S, O”.

That is, the coefficient α _X also corresponds to the contribution value of the present invention.
Further, the interest level calculation means 23 calculates the calculated positive levels Sps (T), Sps (H), Sps (N), Sps (S), Sps (O), and the total positive level Sps (A). The interest is accumulated in the interest accumulation means 25 in association with the product.

The coefficient setting unit 26 sets each coefficient stored in the coefficient storage unit 24.
The coefficient setting unit 26 calculates and sets the weight coefficient W _n (X) and the coefficient α _X based on the values of the sensor data V _{1 to} V ₇ and the history data acquired from the history data management server 40. That is, the coefficient setting unit 26 constitutes a contribution value acquisition unit of the present invention.
Specifically, the coefficient setting unit 26 sets the correct score of Sps (A) based on the buzz degree for the product, the writing to the product, the number of searches for the product, and the like included in the history data.
Then, the coefficient setting means 26 uses the coefficients W _n (X) and α _X as unknowns in the formula obtained by substituting the formula (1) into the formula (2) or the formula (3), and acquires the acquired sensor data V A correct answer formula is created by substituting the values of _{1 to} V ₇ and the set correct answer score into Sps (A). Then, the coefficients W _n (X) and α _X are calculated by solving a large number of correct answer formulas.

[Positiveness calculation process]
Next, a positive degree calculation process by the information management system 1 will be described.
FIG. 2 is a flowchart showing a positive degree calculation process.
When the user approaches the target product in the store, each of the sensors 11 to 16 detects the state of the user and transmits sensor data V _{1 to} V ₇ corresponding to the detection result to the information management server 20.
The sensor data acquisition unit in the information management server 20 21 acquires sensor data _V 1 ~V ₇ transmitted from the sensors 11 to 16 (step S11).

Next, the degree-of-interest calculation unit 23 calculates each positive degree Sps (Tp) according to the above equation (1) based on the acquired sensor data V _{1 to} V ₇ and the weighting coefficient stored in the coefficient storage unit 24. ), Sps (H), Sps (N), Sps (S) and Sps (O) are calculated (step S12).

  Next, the degree-of-interest calculation means 23 calculates the total positive degree Sps (A) by the above formula (2) or (3) (step S13). Then, the interest level calculating means 23 stores the calculated Sps (T), Sps (H), Sps (N), Sps (S), Sps (O), Sps (A) in association with the product and stores the degree of interest. The data is stored in the means 25 and the process is terminated (step S14).

Then, by summing up the positive degree accumulated in the interest accumulation means 25 for each product, that is, by summing up the positive degrees of a plurality of accumulated users for each product, the potential needs of the user for the product can be met. I can grasp.
FIG. 3 is a radar chart showing an example of a totaling result obtained by totaling each positive degree accumulated in the interest degree accumulating means 25.
In the radar chart, the average value for each user group of each positive degree Sps (T), Sps (H), Sps (N), Sps (S), Sps (O) for the target product is shown. The user group is divided according to, for example, the gender and age of the user. For each user group, for example, the information management server 20 analyzes an image captured by the camera 16 to acquire user attributes such as the sex and age of the user, and calculates the acquired user attributes. It can be realized by accumulating with the degree.

  In addition, as one of the indexes indicating the potential needs for the product, a value obtained by adding the average values of each positive degree may be used, or when the average value of each positive degree is shown in a radar chart as shown in FIG. You may use the value of the area of the pentagon enclosed by the outer periphery. Moreover, you may use the average value of total positive degree Sps (A).

[Coefficient calculation process]
Next, calculation processing of each coefficient W _n (X), α _X by the information management system 1 will be described.
FIG. 4 is a flowchart showing a calculation process of each coefficient W _n (X), α _X.

The sensor data acquisition means 21 of the information management server 20 acquires a plurality of sets of sensor data V _{1 to} V ₇ transmitted from the sensors 11 to 16 (step S21). This process may be a process common to step S11 of the positive degree calculation process.

  Next, the history data acquisition unit 22 acquires history data from the history data management server 40 (step S22).

Next, the coefficient setting means 26 sets the above-mentioned correct answer score based on the acquired history data (step S23). Then, the coefficient setting unit 26 creates a plurality of the above-described correct formulas based on the acquired set of sensor data V _{1 to} V ₇ and the set correct score (step S24), and the plurality of correct formulas thus created. Is calculated to calculate each coefficient W _n (X), α _X (step S25).

[Effects of First Embodiment]
In the present embodiment, sensor data acquisition means 21 for acquiring sensor data V _{1 to} V ₇ , history data acquisition means 22 for acquiring history data, and a coefficient for acquiring a weighting coefficient (contribution value) based on the history data. Based on the setting means 26, the sensor data V _{1 to} V ₇ , and the weighting coefficient, the positive degree Sps (T), Sps (H), Sps (N), Sps (S), Sps (O) for the product is determined. Interest degree calculating means 23 for calculating.
According to this, using the sensor data V _{1 to} V ₇ and the history data on the Internet, the positive degree for the product is calculated. For this reason, based on the sensor data V _{1 to} V ₇ , the user's behavior and reaction such as the user standing in front of the product or the product being picked up by hand, that is, real data that is the actual user's behavior or reaction is obtained. Judgment is possible. Further, by acquiring history data that accumulates user actions on the Internet, it becomes possible to determine user actions, impressions, and the like on the Internet. Therefore, by calculating the positive degree based on the sensor data and the history data on the Internet, the degree of positiveness can be calculated with high accuracy. At this time, since sensor data and data on the Internet are used as compared with a complicated method using a questionnaire or the like, the positive degree can be calculated without taking time and effort.

In the present embodiment, the degree-of-interest calculation means 23 calculates the degree of positivity based on a value obtained by multiplying each of the sensor data V _{1 to} V ₇ by a weighting coefficient.
According to this, the weighting factor, the contribution to the positive of each sensor data V ₁ ~V _7, since that can be set for each sensor data V ₁ ~V _7, can be more accurately calculate the positive degree.

In the present embodiment, the degree-of-interest calculation means 23 calculates the positive degrees Sps (T), Sps (H), Sps (N), Sps (S), Sps (O) from different viewpoints, and these positive degrees. Based on this, the total positive degree Sps (A) is calculated.
The total positive degree Sps (A) reflects the positive degree of various viewpoints of the user with respect to the product. Therefore, by obtaining Sps (A), the potential needs of the product can be accurately grasped.

[Second Embodiment]
Next, a second embodiment of the present invention will be described. In the information management system of the second embodiment, the interest degree calculation means 23 uses the sensor data V _{1 for} each positive degree Sps (T), Sps (H), Sps (N), Sps (S), Sps (O). in addition to ~V _7, it is calculated also based on historical data acquired from the history data management server 40. Since other configurations are the same as those of the information management system 1 of the first embodiment, the description thereof is omitted.

That is, in the second embodiment, when the sensor data acquisition unit 21 acquires the sensor data V _{1 to} V ₇ , the history data acquisition unit 22 determines from the history data management server 40 that the sensor data V _{1 to} V ₇ are to be determined. historical data on the Internet users became acquires the sensor data V _8. The history data here includes writing on the product, the number of searches for the product, and the like.

The identification on the Internet users became determination target sensor data V ₁ ~V ₇ is performed, for example, in the following manner.
That is, in the store, the user who is the determination target of the sensor data V _{1 to} V ₇ is tracked using a camera or the like, and it is confirmed that the merchandise sales process has been performed by the POS terminal or the like in the cash register. And ID (Identifiers, identifiers), such as a point card and a credit card which the user used in the sales process at that time, is acquired. Here, if the acquired ID and the ID of the user on the Internet are linked in advance, the user who is the determination target of the sensor data V _{1 to} V ₇ can be identified on the Internet based on the acquired ID. .

Then, the interest level calculating means 23 calculates each positive degree Sps (T), Sps (H), Sps (N), Sps (S), Sps (O) based on the sensor data V _{1 to} V _8. .
That is, Sps (X) is the sum of values obtained by multiplying each of the sensor data V _{1 to} V ₈ by the corresponding weighting factor.

[Effects of Second Embodiment]
In the present embodiment, the sensor data acquisition unit 21 acquires history data that accumulates user actions on the Internet as sensor data V ₈ , and the interest degree calculation unit 23 is based on the sensor data V _{1 to} V ₈ . To calculate the positive degree.
According to this, not only real data but also history data on the Internet can be used as an index for calculating a positive degree.

[Third Embodiment]
In the first embodiment, in the coefficient calculation process, the coefficient setting means 26 sets the above-described correct score based only on the history data acquired from the history data management server 40. In the third embodiment, The correct score reflects the correlation of evaluations between products.

FIG. 5 is a diagram illustrating a configuration of the information management server 20A according to the third embodiment.
The information management server 20A includes a correlation data acquisition unit 27 in addition to the components of the information management server 20 of the first embodiment.
In addition, since the other structure in 3rd Embodiment is the same as that of the information management system 1 of 1st Embodiment, description is abbreviate | omitted.

The correlation data acquisition unit 27 acquires correlation data indicating the correlation of evaluation between products.
The correlation data includes data indicating the relationship between products based on the user's behavior for each product. For example, when user behavior is obtained from a plurality of users, such as when the product “A” and the product “B” are picked up and compared and then the product “A” is selected and bought, the product “B” ”, Data such as increasing a weighting factor for the degree of interest of the product“ A ”by a predetermined value is recorded.
In this case, the correlation data acquisition unit 27 analyzes the user's behavior among a plurality of products based on a plurality of sensor data and the like. For example, the infrared sensor 11 analyzes whether or not the product “A” or “B” has stopped, analyzes whether the user who has stopped by analyzing the image captured by the camera 16 is the same user, and the acceleration sensor 12. Thus, by determining whether or not the products “A” and “B” have been picked up, it is determined that the user has picked up both the products “A” and “B”. Furthermore, by acquiring management data (sales status) of the products “A” and “B” from the product management server 30, it can be determined that only the product “A” has been bought.
By correlating the user behavior of each product of a plurality of users with the correlation data acquisition means 27, correlation between a large number of products can be obtained.

In this embodiment, the coefficient setting unit 26 reflects the correlation data acquired by the correlation data acquisition unit 27 when setting the correct score in step S23 of FIG.
For example, as in the above-described example, when there are many users who have purchased and compared the product “A” with the product “B” and then purchased the product “A”, the coefficient setting unit 26 Adjusts the correct score of the products “A” and “B” such that the product “A” has a higher score than the product “B”.
Thereby, each coefficient W _n (X), α _X is calculated based on the correlation data. Then, each calculated coefficient W _n (X), α _X is used for calculation of each positive degree.

[Effects of Third Embodiment]
In the present embodiment, since the coefficients W _n (X) and α _x are calculated based on correlation data indicating the correlation of evaluation between products, the correlation of evaluation between products can be reflected in the positive degree. And the positive degree can be calculated with higher accuracy.
Further, in the case where the correlation data between the products is not used, for example, even when many products “A” are purchased for the products “A” and “B”, the user's evaluation is good for both. In some cases, both degrees of interest are calculated as substantially the same value. On the other hand, in this embodiment, by analyzing the degree of interest of the products “A” and “B” in more detail based on the correlation between the products, for example, both users are highly evaluated and the number of purchases is also large. Even if this is the case, these products can be further analyzed and differentiated. Thereby, it is possible to carry out a more accurate analysis of the user's latent needs.

[Modification]
In addition, this invention is not limited to embodiment mentioned above, In the range which can achieve the objective of this invention, the deformation | transformation shown below is also included.

[Modification 1]
In each of the above embodiments, each positive degree is calculated based on all of the sensor data V _{1 to} V ₇ , but the present invention is not limited to this. That is, each positive degree may be calculated based on sensor data selected in advance from the sensor data V _{1 to} V ₇ according to the type.
For example, when the sensor data related to the positive degree Sps (T) is the sensor data V ₁ and V ₇ , the positive degree Sps (T) is expressed as “Sps (T) = W ₁ (T) × V ₁ + W _7. (T) × V ₇ ”.
According to this, since the amount of data can be reduced as compared with the case where each positive degree is calculated based on all the sensor data V _{1 to} V ₇ , the load on the calculation process can be reduced.

[Modification 2]
In each of the above-described embodiments, the interest degree calculation unit 23 may reflect the user attributes (gender, age, etc.) of the user who is the determination target of the sensor data in the calculation of each positive degree.
For example, each sensor data and each weight coefficient may be set for each user attribute. That is, each sensor 11 to 16, transmits the sensor data _V 1 corresponding to the user attribute, u _{~V 7, u} to the information management server 20, the interest degree calculating unit 23, sensor data _{V 1,} u ~V _{7 , U} may be multiplied by a corresponding user attribute weight coefficient to calculate each positive degree. In this case, each positive degree Sps (X) is obtained by the following equation (4). In Expression (4), V _{n, u} represents a value of sensor data corresponding to the user attribute, W _{n, u} (X) corresponds to Sps (X), and corresponds to V _{n, u} . Indicates the weighting factor.

User attributes such as gender and age can be acquired by, for example, the information management server 20 analyzing an image captured by the camera 16 or analyzing a voice recognized by the voice recognition sensor 13. That is, the information management server 20B of Modification 2 includes user attribute acquisition means 28 for acquiring user attributes as shown in FIG.
The user attribute may be a result of clustering users according to what the user is interested in. Products that the user is interested in can be acquired by tracking the user with the camera 16, for example.
The user attribute may be a result of clustering users based on an action history on the Internet, for example. As described above, the user who is the sensor data determination target can be identified on the basis of the ID of a point card or a credit card used by the user during the sales process.

In addition, as another method for reflecting the user attribute in the calculation of each positive degree, the value of the user attribute data is multiplied by the sum of values obtained by multiplying each of the sensor data V _{1 to} V _{7 by} the weight coefficient. Values may be added. In this case, each positive degree Sps (X) may be obtained by the following equation (5). In Equation (5), u _m represents the value of the user attribute data, W _m (X) represents Sps (X), and represents a weighting factor corresponding to u _m , where m is 1 to M is a variable, and M is the number of types of user attribute data.

For example, the first user attribute data u ₁ is data indicating whether or not the user is male, and is “1” when the user is “male” and “0” when the user is different. The second user attribute data u ₂ is data indicating whether or not the user is female, and is “1” when the user is “female” and “0” when the user is different.

As described above, in the second modification, the interest level calculation unit 23 calculates the positive degree based on the sensor data V _{1 to} V ₇ , the weighting factor, and the user attribute.
According to this, since the user attribute is also reflected in the positive degree, the accuracy of the positive degree can be further increased.
Note that, when the number of types of user attributes is relatively small, the accuracy of the positive degree is higher when calculated by the above formula (4) than when calculated by the above formula (5). On the other hand, when the degree of positiveness is relatively large, the accuracy is higher when it is obtained by the above equation (5) than when it is obtained by the above equation (4).

[Modification 3]
In each of the above-described embodiments, the interest level calculation unit 23 may reflect information on the companion configuration (one person, a male / female pair, a family, etc.) of a user who is a sensor data determination target in the calculation of each positive degree.
For example, each sensor data and each weight coefficient may be set for each companion composition. That is, each of the sensors 11 to 16 transmits sensor data corresponding to each accompanying person configuration to the information management server 20. As a companion composition, for example, when the number of combinations of the same users is a predetermined number or more based on, for example, each photographed image in the vicinity of a plurality of products by image analysis by the camera 16, the combination of these same users is regarded as a companion. judge. Moreover, according to the user attribute of each user, etc., a user's sex, an age group, etc. may be determined, and it may be determined whether a companion composition is a family or another person.
Then, the degree-of-interest calculation means 23 may calculate each positive degree by multiplying each of the sensor data by a weighting factor of a corresponding companion composition.
Or, a sensor in each of the data V ₁ ~V ₇ to the sum of values obtained by multiplying the weighting factors, may be added to multiplied by the weighting factor of the value of the bank's configuration data values.

[Modification 4]
In each of the above embodiments, each coefficient W _n (X), α _X is generated based on history data on the Internet, but the present invention is not limited to this.
For example, the coefficient setting means 26 uses the above-mentioned correct answer score for obtaining each coefficient as a result of a questionnaire or interview result regarding evaluation of a product for a user who has actually visited the store, or a product acquired by a POS terminal or the like. You may set based on the sales data etc.

[Modification 5]
In each above-mentioned embodiment, although information management system 1 is provided with sensors 11-16, the present invention is not limited to this. That is, other types of sensors may be provided, and the number of sensors may be two or more.

[Modification 6]
In the above embodiments, when a person close to the product is clerks, each sensor 11 to 16, may not transmit the sensor data V ₁ ~V ₇ to the information management server 20.
Whether the person approaching the product is a store clerk is determined by, for example, analyzing an image taken by the camera 16 or reading an ID tag worn by the store clerk with a reader or the like.
According to this, it is possible to prevent noise data from being transmitted to the information management server 20, and to increase the accuracy of each positive degree.

[Modification 7]
In each of the embodiments described above, when the user takes the following actions (1) to (5), the coefficient setting unit 26 calculates the value of the coefficient α _X used for calculating the total positive degree Sps (A). It may be changed.

(1) I picked up the product but did not buy it.
It can be determined using the acceleration sensor 12 that the product has been picked up.
The fact that the product has not been purchased may be determined, for example, by acquiring management data (sales status) of the target product from the product management server 30, or the user is tracked by the camera 16 without passing through the cash register. You may determine by confirming having left the store.
In this case, for example, the coefficient setting unit 26 lowers the values of the coefficients α _{T and} α _H corresponding to the stationary positive degree Sps (T) and the positive degree Sps (H) taken by the hand.

(2) I did not buy a bar code attached to a product, even though I read it using a mobile terminal using an application.
The fact that the price has been searched by the barcode is determined by, for example, acquiring the search history of the application of the mobile terminal.
In this case, the coefficient setting unit 26, for example, to lower the value of the coefficient alpha _T corresponding to the positive degree Sps (T) Tachidomari.

(3) Although I stopped in front of the product, I left without picking it up.
Whether the vehicle has stopped in front of the product may be determined using the infrared sensor 11 or may be determined by analyzing an image captured by the camera 16 or may be equipped with a pressure sensor on the floor. You may determine using a mat | matte.
In this case, the coefficient setting unit 26, for example, to lower the value of the coefficient alpha _T corresponding to the positive degree Sps (T) Tachidomari.

(4) Even though I had picked up the product for a long time, I didn't buy it.
The fact that the product has been picked up for a long time is determined by analyzing an image taken by the camera 16, for example.
In this case, for example, the coefficient setting means 26 reduces the value of the coefficient α _H taken by hand and corresponding to the positive degree Sps (H).

(5) I picked up the product and bought it immediately.
This behavior is determined by analyzing an image taken by the camera 16, for example.
In this case, for example, the coefficient setting unit 26 increases the value of the coefficient α _H that is taken by hand and corresponds to the positive degree Sps (H).

[Modification 8]
In each of the above-described embodiments, the interest level calculation means 23 calculates and calculates each positive degree Sps (T), Sps (H), Sps (N), Sps (S), Sps (O) based on the sensor data. The total positive degree Sps (A) is calculated based on each positive degree, but the present invention is not limited to this.
That is, the interest level calculation means 23 may directly calculate the total positive degree Sps (A) based on the sensor data.

[Modification 9]
In the above embodiments, the sensor data V _{1 to} V ₇ are generated by processing the detection data (raw data) detected by the sensors 11 to 16 as scalar values, but the present invention is not limited to this.
For example, the sensor data V _{1 to} V ₇ may be generated by processing the detection data detected by each of the sensors 11 to 16 into a vector value indicating a time-series wave, a feature amount, or the like in a certain time.
For example, the sensor data V ₂ may be a scalar which is the sum of the amount of motion within a predetermined time of the product obtained from the acceleration sensor 12, even as a value expressed in vector the motion of each unit time in a predetermined time period Good.

[Modification 10]
In each of the above-described embodiments, the interest degree calculating means 23 constituting the evaluation calculating means of the present invention indicates the degree of interest in the goods as positive indices Sps (T), Sps (H), Sps ( N), Sps (S), and Sps (O) have been calculated, but the present invention is not limited to this.
For example, the evaluation calculation means may calculate a commitment level for the product as the index. Specifically, the evaluation calculation means calculates the positive degree Sps (T), Sps (H), Sps (N ), Sps (S), Sps (O).
For example, “cute” and “cool” can be exemplified as the commitment.
For example, when the preference is “cute” or “cool”, the sensor data representing the user's emotion has a high contribution to each preference, and therefore, for example, the voice recognition sensor 13 corresponding to such sensor data. The respective weighting factors for the sensor data V ₃ according to the detection result, the sensor data V ₄ according to the detection result of the thermosensor 14, and the sensor data V ₅ according to the detection result of the electroencephalogram sensor 15 are set to be relatively high. Is expected.
Further, the evaluation calculation means may calculate a recommendation degree for a product as the index.
The evaluation calculation means may calculate the total value of the product by calculating the sum of the positive degree, the sticking degree, the recommendation degree, etc. as the index.

[Modification 11]
In the first embodiment and the modification, the information management server 20 recommends a product to the user according to the positive degree, the sticking degree, the recommendation degree, the total value, and the like calculated by the evaluation calculating unit. Also good.
For example, when the evaluation of the index is relatively high, the information management server 20 adds a product similar to the target product or a product that is often compared with the target product to the history data acquired by the history data acquisition unit 22. It may be recommended by extracting the product based on the data and transmitting the extracted product data to a portable information terminal owned by the user.

  DESCRIPTION OF SYMBOLS 1 ... Information management system, 10 ... Sensor group, 11 ... Infrared sensor, 12 ... Acceleration sensor, 13 ... Voice recognition sensor, 14 ... Thermo sensor, 15 ... Electroencephalogram sensor, 16 ... Camera, 20 ... Information management server, 21 ... Sensor Data acquisition means, 22 ... history data acquisition means, 23 ... interest level calculation means, 24 ... coefficient storage means, 25 ... interest degree storage means, 26 ... coefficient setting means, 27 ... correlation data acquisition means, 28 ... user attribute acquisition means 40: History data management server.

Claims (9)

Sensor data acquisition means for acquiring a plurality of types of sensor data output from a sensor for detecting a user's state;
History data acquisition means for acquiring history data accumulating user actions on the Internet for the product;
Based on the history data, contribution value acquisition means for acquiring a contribution value that the user's state gives to the evaluation of the user's product;
An information management server comprising: an evaluation calculation unit that calculates the evaluation based on the plurality of types of sensor data and the contribution value. In the information management server according to claim 1,
The contribution value acquisition means acquires a weighting coefficient associated with each type of the plurality of types of sensor data as the contribution value,
The information management server, wherein the evaluation calculation unit calculates the evaluation based on a value obtained by multiplying each of the plurality of types of sensor data by a corresponding weighting coefficient. In the information management server according to claim 1 or 2,
The evaluation calculation means calculates a plurality of evaluations having different viewpoints based on the plurality of types of sensor data and the contribution value, and calculates a comprehensive evaluation based on the plurality of evaluations. Information management server. In the information management server according to claim 3,
Each of the plurality of evaluations is calculated based on sensor data selected in advance from the plurality of types of sensor data in accordance with the type thereof. The information management server. In the information management server according to any one of claims 1 to 4,
Comprising user attribute acquisition means for acquiring the user attribute of the user;
The information management server, wherein the evaluation calculation means calculates the evaluation based on the plurality of types of sensor data, the contribution value, and the user attribute. Sensor data acquisition means for acquiring a plurality of types of sensor data output from a sensor for detecting a user's state;
Contribution value acquisition means for acquiring a contribution value that the user's state gives to the evaluation of the user's product;
An information management server comprising: an evaluation calculation unit that calculates an evaluation for a product based on the plurality of types of sensor data and the contribution value. In the information management server according to claim 6,
The information management server, wherein the sensor data acquisition means acquires history data that accumulates user behavior on the Internet for a product as the sensor data. An information management method executed by a computer,
Acquire multiple types of sensor data output from the sensor that detects the user's condition,
Obtain historical data that accumulates user actions on the product on the Internet,
Based on the history data, obtain a contribution value that the user's state gives to the evaluation of the user's product,
The information management method, wherein the evaluation is calculated based on the plurality of types of sensor data and the contribution value. An information management program for causing a computer to function as the information management server according to any one of claims 1 to 7.

Images