JP2008191842A - Evaluation research device and evaluation research method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an evaluation research device and an evaluation research method, capable of estimating high-reliability evaluation even in a stage wherein an actual product is not circulated in the world. <P>SOLUTION: This evaluation research device 1 has: a visitor action classification part 11 for classifying a plurality of action patterns of a research target person; a correlation research part 13 for researching correlation with the evaluation of the research target person to a research object; an evaluator action classification part 14 for classifying action of an evaluation target person into a plurality of action patterns; and a purchase probability estimation part 15 for estimating the evaluation of the evaluation target person to the evaluation object from the action pattern of the evaluation target person based on the correlation. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to an evaluation investigation apparatus and an evaluation investigation method for investigating the evaluation of an evaluation object based on the behavior of an evaluation object person.

  2. Description of the Related Art Conventionally, as a system for investigating product preferences, a system that counts approaches to products and evaluates consumers' unconscious preferences is known. For example, Patent Document 1 discloses customer preference information without the consciousness of responding to a questionnaire by the action of selecting the details of the product including the design by itself in order to obtain a screen saver that suits the user's preference. A preferred accompany survey system is disclosed.

In addition, in Patent Document 2, a wireless device is installed on a hanger in a boutique or the like, and by identifying a product picked up by a customer, an unconscious customer preference is analyzed from the product selection process of the customer.
A system for investigating customer preferences is disclosed.
JP 2003-76830 A JP 2004-295206 A

  However, the technique described in Patent Document 1 only selects what the user prefers from the choices, and cannot determine whether each operator can become a potential customer. In addition, since the operator uses the motivation to obtain a screen saver for information acquisition, each operator uses each item based on the satisfaction of the appearance on the screen, not the selected eye when purchasing the actual product. There is a possibility that an option may be selected, and there is a possibility that a deviation from an actual consumer's product selection behavior or potential demand may occur.

  In addition, the technique described in Patent Document 2 can solve the above-described problem by analyzing the customer's preference from the behavior in which the customer actually determines the product. However, since it is based on the behavior that the customer picks up a product that has already been released, the customer's preference for attributes such as colors and materials that the product has can be evaluated, but has not yet been released. The product itself cannot be evaluated. This is because it is impossible to reproduce the situation where the customer conducts an unconscious investigation while actually determining the product.

  Evaluating products including designs is a preliminary study to examine the direction of the product to be developed, and in the case of a design, etc., or measuring the market competitiveness of the product under development. Or it is a case of selecting a product to be commercialized from a plurality of proposals, and in any case, what is distributed as a product in the world at the time of evaluation is often not subject to evaluation.

  Therefore, in the evaluation meeting, the evaluator does not evaluate to actually purchase the target product. In addition, since the evaluator did not gather at the evaluation meeting to purchase the product to be evaluated, it was the same as when the actual purchase reviewer of the product considered the product purchase after the product was released. It is not always possible to evaluate.

  Evaluation results of evaluators who are not likely to consider purchasing the target product, evaluation results of evaluators who do not evaluate the product as a purchase target or purchase target, or the product is not seriously evaluated Analyzing and judging product evaluation results using the evaluator's evaluation results may cause serious problems in the product development and commercialization processes.

  Therefore, if an actual product is not yet distributed in the world and a true evaluation of the product can be obtained, it will be a great plus for product development and commercialization.

  An evaluation investigation device according to the present invention is an evaluation investigation device that investigates the evaluation of an evaluation object based on the behavior of the evaluation subject, and the first behavior classification that classifies the behavior of the evaluation subject into a plurality of behavior patterns. The evaluation target of the evaluation subject from the behavior pattern of the evaluation subject based on the result of investigating the correlation between the behavior pattern of the research subject and the evaluation of the survey subject with respect to the survey target An evaluation estimation unit for estimating an evaluation of an object.

  In the present invention, it is possible to estimate the evaluation subject's evaluation from the correlation between the behavior pattern of the survey subject who has been investigated in advance and the evaluation. In other words, it was possible to attach importance to the evaluation of evaluation subjects who took the same behavior pattern as those of those who evaluated highly among the survey subjects, leading to negotiations and purchases, etc. It becomes possible to place importance on the evaluation of a person who takes the same evaluation behavior as that of a person's showroom or the like.

  In addition, a correlation database that stores a correlation between a plurality of behavior patterns of the survey target person and the evaluation of the survey target person with respect to the survey target object, the evaluation estimation unit, based on the data of the correlation database, The evaluation of the evaluation object can be calculated by weighting the evaluation of the evaluation object with respect to the evaluation object so that the evaluation of the same action pattern as the action pattern with a high evaluation of the investigation object is large. Thereby, according to a correlation, the weight of evaluation of the evaluation subject who is highly likely to make an appropriate evaluation can be increased.

  Furthermore, it has a correlation database that stores the behavior pattern of the survey subject who has purchased the survey target, or has purchased the survey target and has indicated purchase intention, and the evaluation estimation unit includes the correlation Based on the data in the database, the purchase probability for the evaluation object of the evaluation object person can be calculated. Thereby, a purchase probability can be estimated from an evaluation subject's action pattern.

  Furthermore, it can further have an evaluation result analysis part which extracts the evaluation object person whose evaluation estimated by the said evaluation estimation part is more than predetermined value from the said evaluation object person. By extracting a specific evaluation target person, it is possible to perform an evaluation analysis using only the evaluation result, and to obtain a more reliable analysis result.

  A second behavior classifying unit for classifying the behavior of the survey target person into a plurality of behavior patterns; a correlation survey unit for investigating a correlation between the purchase behavior of the survey target person and the behavior pattern of the survey target person; Can further be included.

  And a behavior characteristic database that divides the periphery of the evaluation object into a plurality of areas in accordance with directions and distances, and stores a time during which the evaluation object person stayed in the area. Can classify the behavior of the person to be evaluated based on the data of the behavior characteristic database. Thereby, for example, it is possible to grasp whether the evaluation target person has stayed at a long position in the front side, the side surface, or the rear side of the product.

  Further, the evaluation object is a vehicle, and has an action characteristic database for storing an investigation result as to whether or not the evaluation object person has opened and closed a door of the vehicle and / or boarded the vehicle. The first behavior classifying unit can classify the behavior of the evaluation subject based on the data of the behavior characteristic database. Thereby, it is possible to examine the correlation between the behavior when the door is opened and closed and the passenger takes a positive behavior with respect to the evaluation object and the evaluation in that case. Further, contact history information and operation history information in the vehicle can also be used for examining the correlation of evaluation.

  Furthermore, the second behavior classifying unit can classify the behavior of the survey subject by the same method as the first behavior classifying unit. Similar to the behavior pattern of the evaluation target person, the correlation between the behavior pattern of the survey target person and the evaluation target person can be investigated by classifying the behavior pattern of the survey target person.

  An evaluation investigation method according to the present invention is an evaluation investigation method for investigating evaluation of an evaluation object based on an evaluation subject's behavior, and is a first behavior classification for classifying the evaluation subject's behavior into a plurality of behavior patterns. Based on the result of investigating the correlation between the process, the plurality of behavior patterns of the survey subject, and the evaluation of the survey subject with respect to the survey target, the evaluation subject of the evaluation subject from the behavior pattern of the assessment subject And an evaluation estimation step for estimating the evaluation of the object.

  According to the present invention, it is possible to provide an evaluation investigation device and an evaluation investigation method capable of estimating a highly reliable evaluation even when an actual product is not distributed in the world.

  Hereinafter, specific embodiments to which the present invention is applied will be described in detail with reference to the drawings. In this embodiment, the present invention is applied to an evaluation investigation apparatus and method for examining the evaluation of a product based on the behavior of an evaluation target person.

  In the present embodiment, a plurality of behavior patterns of the survey target person are stored in advance in a specific region where the product (evaluation target) is placed, and each behavior pattern is digitized. In addition, actions outside the area, for example, actions such as whether or not the product has been purchased at the time of the product negotiation are quantified, and a correlation between the action inside the area and the action outside the area is obtained. In other words, it is examined in advance what action pattern in the region has a high probability of purchasing the product. Then, by examining the behavior pattern in the evaluation target person's area and weighting the behavior pattern based on the correlation obtained above, the evaluation of the evaluation target product for the product, for example, purchase probability, satisfaction with the product , Calculate the satisfaction of various equipment equipped on the product.

  By this method, for example, evaluation subjects with a high purchase probability can be extracted, and in design evaluation of a new model car, design evaluation is performed from a viewpoint close to the person who purchases the target vehicle in order to improve the evaluation accuracy. A person can be selected and the evaluation result reflected in the product.

  In the following embodiments, the evaluation object will be described as a vehicle before the release, the evaluation target person as a visitor who visited the design evaluation hall, and the survey target person as a visitor who visited a store or an exhibition showroom. The target object, the evaluation subject, and the survey subject are not limited to this.

  FIG. 1 is a diagram showing an evaluation survey apparatus according to the present embodiment. The evaluation investigation device 1 includes a calculation unit 10 and a storage unit 20 in which calculation results are stored. The calculation unit 10 includes a visitor behavior classification unit 11 as a second behavior classification unit that classifies the behavior of the survey target person into a plurality of behavior patterns, a purchase behavior classification unit 12 that classifies the purchase behavior of the survey target person, and the survey target Correlation research unit 13 that investigates the correlation between the purchase behavior of the user and the behavior pattern of the survey subject, and the evaluator behavior classification unit 14 as a first behavior classification unit that classifies the behavior of the assessment subject into a plurality of behavior patterns Based on the results of investigating the correlation between the survey subject's multiple behavior patterns and the assessment subject's assessment of the assessment subject, the assessment subject's assessment of the assessment subject's assessment target is estimated from the assessment subject's behavior pattern A purchase probability estimation unit 15 serving as an evaluation estimation unit, and an evaluation result analysis unit 16 that extracts a specific evaluation target person from a plurality of evaluation target persons based on the purchase probability estimated by the purchase probability estimation unit.In addition, the storage unit 20 includes a visitor behavior characteristic DB 21, a visitor purchase behavior DB 22, a correlation value DB 23, and an evaluator behavior characteristic DB 24.

  The visitor behavior classification unit 11 identifies and digitizes the behavior of the visitors who have visited the showroom, for example. For example, the total number of visitors walking, the average staying time in the showroom, whether or not the door of the vehicle being surveyed is open or closed, whether or not there is a ride, whether or not there is a ride, contact with each operation device in the vehicle, operation history information, etc. Investigate action patterns and digitize them so that each action pattern can be identified. In addition, as shown in FIG. 2, for example, the area around which the product 100 is placed is divided into a plurality of areas A01 to A16 according to the direction and distance, and an action pattern such as how much time is spent in which area is displayed. Investigate. In this example, for example, if the user stays in the areas A01, A02, A08, A09, A10, and A16 for a long time, it can be estimated that the front portion of the product 100 is interested.

  Here, for example, an ID tag can be attached to a clipboard or a pen, a sensor can be attached to the floor of the hall, or a video camera can be used to investigate and recognize a visitor's behavior pattern for their image processing. Alternatively, for example, a touch panel type questionnaire device may be installed at the venue, an ID is attached to each visitor, and the ID is input to answer the questionnaire. Then, these behavior patterns are digitized, tabulated for each visitor ID, and recorded in the visitor behavior characteristic DB 21.

  The purchase behavior classifying unit 12 asks the staff whether the business has been negotiated, takes an action indicating purchase intention such as requesting an evaluation table or a catalog, whether the business has been concluded (whether the business has been concluded) The purchase behavior of visitors such as (No) is digitized by attaching an identification number. And it stores in visitor purchase action DB22.

  The correlation investigation unit 13 obtains a correlation between the visitor behavior pattern stored in the visitor behavior characteristic DB 21 and the purchase behavior stored in the visitor purchase behavior DB 22, and contributes to the purchase behavior of each behavior characteristic. The degree (weight) is calculated. In other words, this correlation indicates what kind of behavior pattern the visitor is performing on the premise of purchasing the product. Further, as a product in the target area of interest, the purchase may be discarded as a result of seriously determining the product, so it is necessary to use not only the purchase result but also the history information of the negotiation.

  The evaluator behavior classification unit 14 identifies and digitizes each evaluator's behavior at the product evaluation venue. Similar to the visitor behavior classification unit 11, the total number of walks of visitors, staying time, presence / absence of opening / closing of doors of vehicles as investigation objects, presence / absence of boarding, presence / absence of boarding, operation devices in the vehicle Behavior patterns such as contact with the user, operation history information, and how long the product is placed in each of the plurality of areas divided according to direction and distance. In the investigation, for example, as described above, the information of the ID tag is used, the behavior of the visitor is photographed, and image processing is performed. Then, these behavior patterns are digitized and recorded in the evaluator behavior characteristic DB 24.

  The purchase probability estimation unit 15 calculates the purchase probability of each evaluator by weighting the digitized evaluator's action pattern according to the contribution calculated by the correlation research unit 13. This purchase probability indicates the seriousness of whether or not there is an intention to purchase a product seriously.

  The evaluation result analysis unit 16 extracts an appropriate evaluator based on the calculated purchase probability and analyzes the evaluation result. For example, only evaluators whose purchase probabilities are equal to or higher than a predetermined threshold are extracted, and the evaluations of products by the evaluators are totaled. As a result, among the evaluators at the product evaluation venue, it is possible to select a person who is close to the customer's point of view who is likely to actually purchase the product, and to reflect the evaluation of the visitor's product on the product. Thus, the accuracy of the evaluation result can be improved.

  FIG. 3 shows an example of the hardware configuration of the evaluation investigation device according to the present embodiment. As shown in FIG. 3, the evaluation investigation device 1 includes a CPU (Central Processing Unit) 101, a ROM (Read Only Memory) 102, and a RAM (Random Access Memory) 103, which are connected to each other via a bus 104. Computer. An input / output interface 105 is also connected to the bus 104.

  The input / output interface 105 includes an input unit 106 including a keyboard and a mouse, a display including a CRT and an LCD, an output unit 107 including headphones and speakers, a storage unit 108 including a hard disk, a modem, and a terminal adapter. A communication unit 109 configured by the above is connected. Total number of visitors or evaluators walking from the input unit 106, average staying time in the showroom, opening / closing or contact of the door of the vehicle, which is the subject of investigation, presence / absence of boarding, contact to each operation device in the vehicle, operation History information, the time spent in the areas A01 to A16, purchase behavior, and the like are input, and the output unit 107 outputs an evaluation result such as a purchase rate for each visitor.

  The CPU 101 executes various processes according to various programs constituting the software module stored in the ROM 102 or various programs constituting the software module loaded from the storage unit 108 to the RAM 103. For example, each process of the visitor behavior classification unit 11, the purchase behavior classification unit 12, the correlation survey unit 13, the evaluator behavior classification unit 14, the purchase probability estimation unit 15, and the evaluation result analysis unit 16 in the evaluation survey according to the present embodiment. , For example, digitizing visitor behavior patterns, digitizing visitor purchase behavior, obtaining correlations between behavior patterns and purchase behavior, quantifying evaluator behavior patterns, and from evaluator behavior patterns Each processing such as estimation of purchase probability is performed. The RAM 103 also appropriately stores data necessary for the CPU 101 to execute various processes. For example, each data of visitor action characteristic DB21, visitor purchase action DB22, correlation DB23, and evaluator action characteristic DB24 is stored.

  The communication unit 109 performs communication processing via the Internet (not shown), transmits data provided from the CPU 101, and outputs data received from a communication partner to the CPU 101, the RAM 103, and the storage unit 108. The storage unit 108 exchanges with the CPU 101 to save and erase information. The communication unit 109 also performs communication processing of analog signals or digital signals with other devices.

  A drive 110 is connected to the input / output interface 105 as necessary, and a magnetic disk 111, an optical disk 112, a flexibil disk 113, a semiconductor memory 114, or the like is appropriately mounted, and a computer program read from them is loaded. It is installed in the storage unit 108 as necessary.

  Next, a method for conducting an evaluation survey based on the product observation behavior of the showroom visitors will be described in detail. FIG. 4 is a flowchart showing the evaluation investigation method according to the present embodiment. As shown in FIG. 4, first, a visitor's action pattern is input into the visitor action classification | category part 11 from the ID tag attached to the clipboard or pen, the sensor attached to the floor. Based on this input, the visitor behavior classification unit 11 digitizes the visitor's behavior pattern. Alternatively, a video by a video camera that captures the venue is input and image processing is performed to identify the location and behavior of the visitors. (Step S1). At this time, as shown in FIG. 5, the behavior pattern of each visitor is associated with an evaluator ID assigned to each visitor and is stored in a visitor behavior characteristic DB 21 provided in a storage unit such as a RAM or HDD. To be recorded.

  Next, the purchase behavior of each visitor is classified (step S2). The purchase behavior is, for example, the start of a business negotiation or a contract. For example, the purchase behavior classifying unit 12 determines whether or not the business negotiation has been performed, whether or not the staff has been questioned, whether or not an action indicating purchase intention such as charging a price list or a catalog has been taken, whether the business negotiation has been established, or the like. And is recorded for each visitor ID in a visitor purchase behavior DB 22 provided in a storage unit such as a RAM or HDD.

  Then, the correlation investigation unit 13 analyzes the correlation between the visitor's product observation behavior (behavior pattern) and the subsequent purchase behavior, and the contribution of each behavior characteristic to the subsequent purchase behavior (start of negotiations, contracts, etc.). The size is digitized and obtained as a weight (see step S3, FIG. 5). That is, the relationship between the degree of interest or the purchase probability for the survey object is obtained from each behavior pattern. Here, for example, a visitor who has boarded a product is N times as likely as a visitor who did not get on, and the probability that a contract will be established, or if the stay time in an area close to the product is more than a few minutes From the correlation between the action pattern and the purchase action, the weight for the purchase action is obtained for each action pattern. In this case, a larger weight is set for an action pattern with a higher purchase probability. This weight is recorded in the correlation value DB 23 provided in a storage unit such as a RAM or HDD.

  Based on the above analysis results, a product evaluation survey will be conducted based on the behavior patterns of visitors who visited the product evaluation venue. It should be noted that it is preferable that the products displayed in the showroom for acquiring the data recorded in the correlation value DB 23 and the products displayed in the product evaluation venue have some relationship. For example, in the case of a vehicle, it is preferable that there is a relevance such as for young women and for young men, for sports types, and family types.

  In this embodiment, the visitor behavior classification unit 11, the purchase behavior classification unit 12, and the correlation survey unit 13 have been described. However, the correlation between the visitor behavior pattern and the purchase behavior is separately obtained in advance. The following purchase probability may be estimated using the data. Moreover, since the visitor action classification | category part 11 and the evaluator action classification | category part 14 perform the same process, it is also possible to comprise with one module.

  Next, the evaluation behavior (position and behavior) of the evaluator at the product evaluation venue is specified and digitized (step S4). The action pattern collection method here can be the same as in step S1. The obtained behavior pattern is recorded in the evaluator behavior characteristic DB 24 provided in the storage unit including the RAM and the HDD.

  Next, based on the weight obtained in step S3, the seriousness degree (proximity to the customer) of the evaluator's target product is quantified from the behavior pattern of each evaluator as shown in FIG. 6 (step S5). ). This seriousness indicates, for example, a purchase probability for a product. Specifically, the purchase probability estimation unit 15 reads the behavior pattern stored in the evaluator behavior characteristic DB 24 of the storage unit, and calculates the purchase probability of the behavior pattern with reference to the correlation DB 23. Then, the evaluation result analysis unit 16 extracts a more appropriate evaluator based on the obtained purchase probability (seriousness data), and analyzes the evaluation result (step S6). For example, if the analysis of the evaluation is performed using the evaluation of the product of a visitor having a purchase probability of 50% or more, an evaluation from a viewpoint closer to the product purchaser can be obtained.

  In this way, out of those who participated in the design evaluation, by selecting an evaluator who took an action close to the person who led to purchase or purchase consideration at the store or showroom, and totaled the evaluation results, The accuracy can be improved.

  In this embodiment, the behavior of visitors in showrooms such as dealers and exhibition halls is measured without making the person aware of it, and the correlation with the subsequent behavior of each visitor (starting or closing of a business negotiation) is analyzed. In this way, the observation behavior of people who can be customers is identified and quantified. In the design evaluation of new products, the observation behavior of the evaluator measured by the same mechanism is digitized using these data, the person close to the customer's viewpoint is identified, and the evaluator to be adopted is selected and evaluated. The accuracy of the evaluation result can be increased by weighting the result.

  It should be noted that the present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the scope of the present invention.

  For example, in the above-described embodiment, the hardware configuration has been described. However, the present invention is not limited to this, and any processing is realized by causing a CPU (Central Processing Unit) to execute a computer program. Is also possible. In this case, the computer program can be provided by being recorded on a recording medium, or can be provided by being transmitted via the Internet or another transmission medium.

It is a figure which shows the evaluation investigation apparatus concerning embodiment of this invention. It is a figure which shows an example of the hardware constitutions of the evaluation investigation apparatus concerning embodiment of this invention. It is a figure which shows the surrounding area in which the goods concerning embodiment of this invention were mounted. It is a flowchart which shows the evaluation investigation method concerning embodiment of this invention. It is a figure which shows a behavior pattern, purchase behavior, and the correlation (contribution) calculated | required from them. It is the figure which digitized the seriousness (closeness to a customer) with respect to the object goods of an evaluator.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Calculation part 11 Visitor action classification part 12 Purchasing action classification part 13 Correlation investigation part 14 Evaluator action classification part 15 Purchase probability estimation part 16 Evaluation result analysis part 20 Storage part 21 Visitor action characteristic DB
22 Visitor Purchasing Behavior DB
23 Correlation value DB
24 Evaluator behavior characteristics DB
100 products

Claims (18)

An evaluation investigation device for investigating the evaluation of an evaluation object based on the behavior of the evaluation object,
A first behavior classifying unit that classifies the behavior of the person to be evaluated into a plurality of behavior patterns;
Based on the result of investigating the correlation between the plurality of behavior patterns of the survey target person and the evaluation of the survey target person with respect to the survey target object, the evaluation target person's evaluation of the evaluation target object from the behavior pattern of the evaluation target person An evaluation investigation device comprising: an evaluation estimation unit that estimates A correlation database for storing a correlation between the plurality of behavior patterns of the survey target and the evaluation of the survey target with respect to the survey target;
The evaluation estimation unit weights evaluation of the evaluation target with respect to the evaluation target so that the evaluation of the same behavior pattern as the behavior pattern with high evaluation of the survey target is increased based on the data of the correlation database. The evaluation investigation apparatus according to claim 1, wherein the evaluation of the evaluation object is calculated. Having a correlation database storing the behavior pattern of the survey subject who has purchased the survey target;
The evaluation investigation device according to claim 1, wherein the evaluation estimation unit calculates a purchase probability for the evaluation object of the evaluation object person based on data in the correlation database. Having a correlation database for storing the behavior pattern of the survey subject who has purchased the survey object and indicated a purchase intention;
The evaluation investigation device according to claim 1, wherein the evaluation estimation unit calculates a purchase probability for the evaluation object of the evaluation object person based on data in the correlation database. 5. The apparatus according to claim 1, further comprising an evaluation result analysis unit that extracts an evaluation target person whose evaluation estimated by the evaluation estimation unit is a predetermined value or more from the evaluation target person. Evaluation survey equipment. A second behavior classifying unit for classifying the behavior of the survey target person into a plurality of behavior patterns;
The evaluation investigation device according to claim 1, further comprising a correlation investigation unit that investigates a correlation between the purchase behavior of the survey subject and the behavior pattern of the survey subject. A behavior characteristic database that divides the periphery of the evaluation object into a plurality of areas according to directions and distances, and stores the time during which the evaluation object person stayed in the area;
The evaluation survey device according to any one of claims 1 to 6, wherein the first behavior classifying unit classifies the behavior of the evaluation target person based on data of the behavior characteristic database. The evaluation object is a vehicle, and has an action characteristic database that stores an investigation result of whether or not the evaluation object person opened and closed the door of the vehicle and / or got on the vehicle,
The evaluation survey apparatus according to any one of claims 1 to 7, wherein the first behavior classifying unit classifies the behavior of the evaluation target person based on data of the behavior characteristic database. The evaluation survey device according to any one of claims 6 to 8, wherein the second behavior classifying unit classifies the behavior of the survey target person by the same method as the first behavior classifying unit. . An evaluation survey method for investigating the evaluation of an evaluation object based on the behavior of the evaluation object,
A first behavior classification step for classifying the behavior of the person to be evaluated into a plurality of behavior patterns;
Based on the result of investigating the correlation between the plurality of behavior patterns of the survey target person and the evaluation of the survey target person with respect to the survey target object, the evaluation target person's evaluation of the evaluation target object from the behavior pattern of the evaluation target person An evaluation investigation method comprising: an evaluation estimation step for estimating The evaluation estimation unit is based on data in a correlation database that stores correlations between a plurality of behavior patterns of the survey subject and the assessment of the survey subject with respect to the survey target, and behaviors that are highly evaluated by the survey subject The evaluation of the evaluation object according to claim 10, wherein the evaluation of the evaluation object is calculated by weighting the evaluation of the evaluation object with respect to the evaluation object so that the evaluation of the same behavior pattern as the pattern is increased. Method. In the evaluation estimation step, a purchase probability for the evaluation object of the evaluation object is calculated based on data of a correlation database storing the behavior pattern of the investigation object who has purchased the investigation object. The evaluation investigation method according to claim 10 or 11. In the evaluation estimation step, the purchase probability for the evaluation object of the evaluation object is based on data of a correlation database that stores the behavior pattern of the investigation object person who has purchased the investigation object and indicated the purchase intention. The evaluation investigation method according to claim 10 or 11, wherein the evaluation investigation method is calculated. The evaluation result analyzing step of extracting an evaluation target person whose evaluation estimated by the evaluation estimation unit is equal to or higher than a predetermined value from the evaluation target person. 14. Evaluation survey method. A second behavior classification step of classifying the behavior of the survey subject into a plurality of behavior patterns;
The evaluation investigation method according to claim 10, further comprising a correlation investigation step of investigating a correlation between the purchase behavior of the survey subject and the behavior pattern of the survey subject. In the first behavior classification step, based on data of a behavior characteristic database that divides the periphery of the evaluation target object into a plurality of areas according to directions and distances, and stores the time during which the evaluation target person stayed in the area. The evaluation investigation method according to any one of claims 10 to 15, wherein the behavior of the evaluation subject is classified. In the first behavior classification step, the evaluation is performed based on data of a behavior characteristic database that stores an investigation result of whether or not the person to be evaluated has opened / closed the door of the vehicle and / or boarded the vehicle. The evaluation survey method according to any one of claims 10 to 16, wherein the behavior of the subject is classified. The evaluation survey according to any one of claims 15 to 17, wherein, in the second behavior classification step, the behavior of the survey target person is classified by the same method as the first behavior classification unit. apparatus.

Images