JP2016177443A - Information processing device and method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To evaluate advertising effects including memorizing, in accordance with characteristics of a dialogic type of interaction.SOLUTION: An information processing device according to one embodiment of the present invention is an information processing device relating to interaction between a person and an artificial material. The information processing device includes: an object storing unit for storing an advertisement, having been output from an artificial material performing a dialogic type of interaction with a user to the user, in association with an object of the advertisement; and an evaluation unit for evaluating an advertisement effect of the advertisement, in the interaction with the user, on the basis of the advertisement stored in the object storing unit or an appearance of the object. The information processing device can be achieved by an information processing method or a computer program.SELECTED DRAWING: Figure 1

Description

  The present invention relates to human-artifact interaction.

  Conventionally, there is a proposal that a robot applies for an advertisement to a human (see, for example, Patent Document 1). On the other hand, conventional effect measurement techniques for electronic advertisements include display advertisements and text advertisements that are displayed on websites, etc., and use conversion rates such as click rate (CTR), purchase and document requests. Are known. Thus, as a background based on direct short-term events such as clicks based on the output advertisement and application screen transition, a conventional device such as a PC based on a user's manual operation is between a user and a server. This was due to the fact that it was merely a mechanical intervention and that it was technically difficult to confirm anything other than direct short-term events (for example, brand or product inscription).

JP 2014-017200 A

  However, artifacts such as robots based on interactive communication are the main actors that interact with users independently, and there is no technology to measure the effect of advertisements output from such artifacts in a form that is suitable for the characteristics of communication. It was.

  An object of the present invention is to evaluate an advertising effect including memorization according to a characteristic of interactive interaction.

  An information processing apparatus according to one aspect of the present invention includes an object storage unit that stores an advertisement output to the user from an artifact that performs interactive interaction with the user and the object of the advertisement in association with each other, The interaction with the user includes an evaluation unit that evaluates the advertisement effect of the advertisement based on the advertisement stored in the object storage unit or the appearance of the object.

  ADVANTAGE OF THE INVENTION According to this invention, according to the characteristic of interactive interaction, the advertising effect including memorization can be evaluated.

The functional block diagram which shows a structure about embodiment of this invention. The figure which shows the example of the data in embodiment of this invention. The figure which shows the example of the data in embodiment of this invention. The figure which shows the example of the data in embodiment of this invention. The flowchart which shows the example of the process in embodiment of this invention.

  Next, a mode for carrying out the present invention (referred to as “embodiment”) will be illustrated along the drawings. In addition, the assumptions common to the contents already described in the background art and problems are omitted as appropriate.

  The purpose of this embodiment is to realize a smooth interaction adapted to the familiarity with the user. Another object of this embodiment is to select an advertisement with a value according to the relationship between the user and the artifact. Another object of the present embodiment is to adapt the mode of outputting an advertisement from an artifact to the intimacy with the user. Another object of the present embodiment is to evaluate the advertising effect including the memorization according to the characteristics of interactive interaction.

[1. Configuration] FIG. 1
FIG. 1 shows the overall configuration of this embodiment. In this embodiment, an artifact P and a plurality of servers A and C are connected via a communication network N (Internet, mobile phone network, etc.), and the parent of each artifact P and the user of the artifact P is connected. The information processing apparatus is realized as a system that determines the density and outputs an advertisement according to the familiarity and evaluates the advertisement effect.

  The artifact P is an information processing apparatus that is typically a robot, and the robot may take a human form, an animal form, or any other form. The artifact P may be a smartphone, a tablet PC, or other portable information terminal, and may be a software agent that is a virtual character that is active on the information processing apparatus.

  The server C is a device that provides interaction with the user of the artifact P by network computing (so-called cloud) as, for example, a back-end system using artificial intelligence technology or the like.

  The artificial object P and the servers C and A each have a computer configuration, such as an arithmetic control unit 3 such as a CPU, a storage device 4 such as a main memory and an auxiliary storage device, and a communication device 5 (communication device) with the communication network N. Or a communication adapter) (not shown except for the artifact P).

  The artifact P further includes a sensory / effector group SM for acquiring information on the outside world and working on the outside world. The sensory / effector group SM is not shown separately, but a visual system (for example, a camera, equipment using infrared rays, laser beams, etc.), a hearing system such as a microphone, a posture / navigation system (for example, an accelerometer) , GPS, etc.), information output system (eg, speaker for speech, display screen, infrared output device, etc.), drive system (moving wheels or legs, arms, hands, head, torso, etc.) A mechanism for realizing gestures), a built-in rechargeable battery control system, and the like, and each system includes hardware, software, and the like.

  The artifact P in this embodiment is a computer configuration (arithmetic control unit 3, storage device 4, communication device) as a control device for controlling the robot body (exterior case, joint, drive system, etc.) and sensory / effector group SM. The control device may be separate from the artifact P.

  That is, in the artifact in the present invention, since it is not essential to incorporate an information processing function, like a puppet (puppet), it is composed only of a body, a drive system such as an actuator, and necessary sensors. The interaction, the recognition associated therewith, the calculation, the motion control, and the like may be performed from a control device provided outside to the artificial object P by transmission using wired or wireless signal transmission, electric machinery, air pressure, hydraulic pressure, or the like.

  In the artifact P and each of the servers C and A, the arithmetic control unit 3 executes the computer program stored in the storage device 4, thereby realizing each element (see FIG. 1) that realizes the following functional actions.

[2. Determination of intimacy
FIG. 5 is a flowchart showing an outline of information processing in the present embodiment. First, in this embodiment, the determination unit (the local determination unit 40 of the artifact P and the relative determination unit C3 of the cloud server C) determines the closeness between the user and the artifact P (step S2).

  As a premise thereof, the artifact P performs interactive interaction with the user. Interactive interaction is centered on conversation (speech to the other party and recognition of the other party's utterance), and is an interaction that involves input / output of gestures, contact with the partner, text, images, and sound as required. It is simply abbreviated as “interaction”.

  The process in which the artifact P recognizes the user and other external objects, and the interaction between the artifact P and the user based on them, includes various filtering, independent component analysis, SVM, contour extraction, and other image processing Technology, pattern recognition (eg, speech recognition, face recognition, etc.), natural language processing, knowledge information processing, reinforcement learning, Bayesian network, SOM (self-organizing map), neural network, Deep Learning (deep learning), and other various machines It can be realized by a known technique such as learning.

  The basic response of the interaction is the input / recognition unit 10 for recognizing external things and events such as user actions, mainly using sensory devices such as sensors in the sensory device / effector group SM, and the user And the action history of the artifact P and other memories (for example, the history storage unit 25), an inference response unit 20 that determines inference and response contents according to the situation, and an information output system of the sensory / effector group SM And an output control unit 30 for outputting information, action, and the like, mainly using a drive system and using feedback of an output result.

  When the artifact P is able to communicate with the cloud server C, the complex advanced inference judgment and the inference judgment based on collective intelligence based on the interaction history of a large number of artifacts P using the cloud server C, The storage of other storage that does not fit in the local storage (for example, history storage unit 25) on the artifact P, access to such storage, and the like are entrusted to the inference determination unit C1 of the cloud server C.

  The artifact P identifies a main user (so-called “master”) who is a partner of the interaction by collation of face and voice, login authentication by inputting a secret password or password, a place where the person is, and the like.

  Specifically, among the determination units, the local determination unit 40 (relative determination unit C3 will be described later) is the type and amount of user behavior detected when accompanying an artificial object P that interacts with the user. Is stored in the history storage unit 25 (step S1), based on this, the intimacy between the user and the artifact is determined and stored in the user information storage unit C4 of the cloud server C (for example, FIG. 2). Is updated (step S2). The companion is that the user and the artifact P exist in a range where interaction is possible.

  The changing unit C5 changes the content of the interaction with the artifact P according to the familiarity determined by the determining unit (the local determining unit 40 and the relative determining unit C3) (step S3). Examples of changes include changes in the information content provided to the user from the artifact P in the interaction, and whether the utterances to the user are in a polite tone or in a familiar tone It is.

  The user behavior that can be used as a basis for the intimacy determination is, for example, maintenance and management of the artifact P (for example, cleaning, care for charging, inspection, etc.). For cleaning, for example, a camera or a contact sensor detects an operation of stroking the surface of the artificial object P with a cleaning tool such as cloth. The care of charging includes, for example, transporting an artifact to the charging place, teaching the charging place, guiding to the charging place, connecting to a charging cable, replacing the battery, and the like.

  The inspection includes, for example, an action of lightly touching each part of the sensory sensation with a fingertip, an action of carefully removing parts so as not to break, disassembly, activation of a self-diagnosis function built into the artifact P, and the like.

  Another user behavior that can be used as a basis for the intimacy determination is a predetermined behavior representing the user's attachment to the artifact. Such predetermined actions include, for example, greeting the artifact, showing a positive expression such as a smile, calling with a name, wearing clothes, glasses, sunglasses, covers, protective seals, etc. In addition to skinship (boiled, touched, leaned, hugged, kissed, etc.), you may bring it to the bed or drive it in a car seat.

  In addition, the predetermined behavior indicating attachment is an action of installing a service via the Internet and a centralized password management software on the artifact P having a computer function, and an electronic lock of the entrance using a near field communication function such as NFC. An action of installing software that realizes an electronic key function for unlocking the screen may be used.

  Other user actions that can be used as a basis for the determination of intimacy are important predetermined actions (for example, meals, bathing, electronic commerce, investment transactions, safe open / close, sleep, etc.). Whether or not the user is taking each type of user behavior and the amount thereof are determined by detecting a behavior pattern predefined as a user behavior for each type from the sensor input. Further, depending on the type of user behavior, the user action is detected from information processing activities such as Internet access performed by the user using the function of the artifact P itself as a computer.

  For information processing activities that user actions are performed using a computer other than the artifact P, the user accesses the other computer using the access authority, ID, password, etc. granted in advance by the user. The presence / absence or amount of user behavior may be determined by monitoring or inquiring about the content of the information processing activity being performed.

  The amount of user behavior determined as described above is, for example, an accompanying range when the user behavior is detected. The companion is one or both of the state in which the user is accompanied by the artifact P, or the state in which the user and the artifact P are in an interactable range. , Cost burden, action type, and other factors.

  In addition, the determination of intimacy may not be completed with a single artifact P. For example, the relative determination unit C3, which is the determination unit of the cloud server C connected to the artifact P through the communication network N, accumulates and aggregates the user actions received as information about the plurality of artifacts P in the behavior storage unit C2, By performing relative evaluation and obtaining a predetermined evaluation value such as a deviation value, the closeness with the user corresponding to each individual artifact P may be determined.

  The determined familiarity is stored and updated in the user information storage unit C4 (FIG. 2). Further, the reception of the user behavior “about the artificial object P” may be received from the artificial object P itself or may be received from the control device of the artificial object P configured separately from the artificial object P.

  Further, the accompanying with the artifact P that is a premise of the intimacy determination is not limited to real but may be virtual. That is, a user or a user who is out of the activity base of the artifact P is considered. At this time, as a software agent realized by an information processing apparatus (for example, a smartphone), the determination unit also determines the intimacy for a virtual companion who enters the smartphone and virtually interacts with the user. Based on.

  The activity base here is a geographical area (for example, home, site, office space, building, etc.) where the user or artifact is located for a relatively long time, and is determined by statistical processing of location history. Alternatively, the determination is made based on the result of such statistical processing and the section of the area represented by the map information or the plan view information.

  Further, the closeness may be determined based on the handling characteristics of the information processing apparatus used for the virtual accompanying as described above. For example, for a smartphone containing the software agent, whether the physical handling is polite or violent is measured with an accelerometer built into the smartphone, and it is determined that the intimacy is politely higher.

  In addition, as another example of handling characteristics, the temperature is detected by a temperature sensor etc. for a smartphone when the user carries the smartphone in a trouser pocket, an inner pocket of clothes, a smartphone holder, etc. Then, in many time zones (a ratio or frequency that is greater than or equal to a predetermined value in a predetermined time zone), there is a high possibility of staying in the 30 degree Celsius range or a predetermined range above and below it. In this case, it can be determined that the familiarity is high.

  The determination unit can also determine that the closeness is high when detecting an interaction while performing a predetermined action while going out. Here, the interaction is not limited to voice, but may be a touch operation with a finger. In addition, the predetermined action here is a social activity that usually involves negative elements such as difficulties, disadvantages, disabilities, and resistance in the operation of information equipment.

  For example, if the user interacts with a walking smartphone, interacts during work, during a lecture, or works part-time (part-time job), or has a high interaction frequency, it can be determined that the intimacy is high.

  Further, the determination unit can determine that the degree of intimacy is high when an action related to an object that appears in the interaction of the user or the artifact P in the activity base is detected while going out of the activity base. . For example, this is a case where the user actually visits the ramen shop within a predetermined period (for example, within one week, within one month, etc.) from an interaction such as “recommended ramen of XX” with the name of the shop.

  The interaction in this case is not limited to advertisement output. In addition, actions such as actually going to a ramen shop are not limited to when accompanied by an artifact P, visit judgment based on location history acquired by the car navigation application or map application used by the user, Oral reports from the user to the artifacts (for example, “I ’m talking about OO ramen, I ate today”).

[3. (Ad selection according to intimacy)
The intimacy determined as described above can be utilized for advertisement allocation (allocation, selection). In this case, in the advertisement server A, the information of the advertisement put out by the advertiser is stored in advance in the advertisement storage unit A1 together with the evaluation value of the advertisement (FIG. 3), and the selection unit A2 of the advertisement server A (the change unit) Is selected from the advertisement storage unit A1 and assigned as an output target for the artificial object P stored in the user information storage unit C4 and having a higher familiarity (FIG. 2) (step S4). ).

  The criteria for selecting an advertisement include not only the matching of the closeness and the height of the evaluation value, but also whether attributes such as age and gender and interest match with the conditions specified by the advertiser.

  Then, the transmission unit A4 transmits the advertisement selected by the selection unit A2 to the artifact P (Step S5). Here, in order to simplify the explanation, information on the content to be output from the artifact P as an advertisement (for example, text to be uttered as an advertisement, product name of the object, etc.) itself is output with an output deadline. Is transmitted from the transmission unit A4 to the artifact P (FIG. 4).

  However, for the transmission of the advertisement from the transmission unit A4 to the artifact P, only the identification information such as an ID for specifying the advertisement and the download URL is transmitted, and this identification information is used on the artifact P side. You may make it download from this advertisement delivery server (illustration omitted).

  For example, when the advertisement is output from the artifact P, the evaluation value of the advertisement used for selecting the advertisement is a rate at which a predetermined result (for example, a user applied for purchase or requested a document) was obtained. (For example, FIG. 3 and FIG. 7. Here, it is temporarily called “response rate”), or an advertisement unit price that is a bid amount bid for and associated with an advertisement may be used. Further, the evaluation value may be a value obtained by multiplying the advertisement unit price associated with the advertisement and the response rate.

  Then, the artifact P stores the advertisement transmitted by the transmission unit A4 in the advertisement storage unit 45 that is a storage unit together with the output deadline (step S5). Thereafter, when the timing determination unit 70, which is a determination unit, determines that the predetermined output timing is suitable for the output of the advertisement based on the acquisition of information about the user and the interaction with the user (step S7: "YES"), the advertisement The output unit 50 outputs the advertisement stored in the advertisement storage unit 45 (step S8).

  In addition, the information regarding the user used for timing determination is not limited to the information during the interaction, but is general information regarding the user obtained from the sensor group of the artifact P. For example, the position and posture of the user in the home (standing) , Sitting on the couch, etc.), behavior (during a meal, on the phone, etc.), utterance content, etc.

  The predetermined output timing suitable for the output of the advertisement is, for example, when the user talks to the artifact P or when the user takes a breath. As an example, if the user sits after acquiring a drink and is not talking to another person, the user is likely to take a breath.

  Moreover, you may utilize a user's life stage understood by the continuous relationship with the artifact P for the selection of advertisement. In this case, the stage determination unit C6 of the cloud server C determines the life stage of the user based on information obtained regarding continuous interaction with the artifact P. Life stage includes, for example, the person's age, sex, occupation, spouse or child presence, number of children, child age or age, child gender, type of school the child attends, type of housing, mortgage And the existence or amount of savings, etc.

  Then, the selection unit A3 gives priority to an advertisement in which the life stage of the user determined by the stage determination unit C6 matches the life stage associated with the advertisement as an advertisement to be selected for the artifact P, The advertisement is assigned to the artifact P. For example, when the life stage associated with the advertisement is that there is a child up to elementary school, the user has made an interaction P including an utterance such as “Son is also a 4th son from this spring” by the user. Is preferred in the selection of the advertisement.

  It is possible to change in what logical operation relationship the conditions such as evaluation value, intimacy, attribute, life stage, etc., and in what weight are prioritized according to the purpose of the advertisement and other conditions. Further, the life stage is not limited to the one determined from the interaction with the artifact P, and may be used together with, for example, one explicitly registered by user registration or the like.

[4. (Advertisement output in a mode according to intimacy)
The mode of outputting the advertisement from the artifact P to the user can also be switched according to the familiarity. The target of such switching is not limited to advertisements selected based on intimacy, but may be advertisements that are output indiscriminately or according to only user attributes and life stages other than intimacy, for example. In this case, the advertisement output unit 50 of the artifact P outputs an advertisement from the artifact P to the user in an output manner corresponding to the familiarity stored in the user information storage unit C4 of the advertisement server A.

  For example, when the artifact P recognizes that the intimacy is equal to or greater than a predetermined value and the user is within a predetermined range, the advertisement output unit 50 uses the self-speaking about the advertisement (presence, content, target) as the advertisement output. Speaks to the artifact P.

  For example, the self-speaking about the advertisement may be the content of the advertisement (for example, “New model of XX company, fuel consumption is △ △ km or so”), or simply ask the brand name or company name to be advertised. Also, the presence of the advertisement to be output may be attracted to the user by intentionally uttering the presence of the advertisement, for example, “That story of the new model car.

  As another example, when the intimacy is greater than or equal to a predetermined value, the advertisement output unit 50 determines that the artificial object P recognizes that the user is within a predetermined range and an object related to the advertisement. An advertisement may be output to the object P. Further, the advertisement output unit 50 may cause the artifact P to output an advertisement by appropriately indicating the existence of the object recognized by the artifact P and the relationship between the object and the advertisement.

  For example, when a user who returns home retrieves a car information magazine from his bag, he says, “You bought a car magazine. Do you know a new car from XX? .

  Further, as another example, the advertisement output unit 50 has a familiarity with a certain degree or more of the advertisement content related to the display contents of the website, television, and other screens browsed by the user, such as in daily conversation. As an appropriate indication, the artifact P may be uttered.

  For example, in the case where an advertisement related to the release of a new model car of XX Company is output to a user who is given a user attribute representing interest in a car, the artificial object P is “ Speaking of which, it seems that a new model car will come out from XX company ", and the advertisement is output in the form of displaying the product image on the display screen of the artifact P itself accordingly.

  The output mode according to the familiarity as described above is realized as follows, for example. For example, advertisement information including a plurality of element information respectively corresponding to a plurality of output modes is stored in the advertisement storage unit 45 which is a storage unit (FIG. 3). Then, the advertisement output unit 50 uses the element information corresponding to the output mode corresponding to the familiarity among the plurality of element information stored in the advertisement storage unit 45 (step S6), and sends the advertisement from the artifact P. Output to the user (step S8).

  For example, in an advertisement for outputting an advertisement of B company “b1 toothbrush” to a user who bought a toothbrush a month ago, the above element information is, for example, a low price as a feature that should be appealed for the advertising object. A case is considered in which information to be emphasized (for example, it is 10% cheaper than the conventional product) and information to emphasize the high quality (for example, the result of user review is excellent) is included.

  In this case, when the intimacy is a predetermined level or more, the advertisement output unit 50 displays an advertisement from the artifact P in an output mode (in the above example, the user review result is excellent) that emphasizes high quality over low cost. Output to the user, that is, make a product recommendation. On the other hand, when the intimacy is less than a predetermined level, the advertisement output unit 50 displays the advertisement from the artifact P in an output mode (in the above example, it is 10% cheaper than the conventional product) that emphasizes the low price rather than the high quality. Output, that is, product recommendation.

  Note that the advertisement output timing may be determined based on activity patterns such as lifestyle habits. In this case, the extraction unit (pattern extraction unit 60 in FIG. 1) extracts an activity pattern such as a user's lifestyle based on acquisition of information about the user and interaction with the user, and stores the activity pattern in the pattern storage unit 65. Then, the determination unit (timing determination unit 70 in FIG. 1) determines a predetermined output timing suitable for the output of the advertisement from the activity patterns extracted by the pattern extraction unit 60 and stored in the pattern storage unit 65 (step S7). ).

  The advertisement output unit 50 outputs the advertisement from the artifact P to the user in an output manner according to the familiarity when the timing determination unit 70 determines that the output timing is reached (step S7: “YES”) (step S8). ). For example, if the user's bathing time period is almost stable every day, an advertisement of canned beer may be output in a manner similar to recommendation or one after the bathing, that is, after the bath.

[5. (Interaction advertising effectiveness evaluation)
As described above, the advertising effect of the advertisement output from the artifact P can be performed based on the subsequent interaction. For example, the advertisement output from the artifact P to the user and the advertisement object are associated with each other and stored in the advertisement storage unit 45 (step S9). In order to store the output advertisement, for example, a value may be set in the output date / time field of the advertisement storage unit 45 that was blank before output.

  Then, the evaluation unit A4 is based on the appearance in the interaction with the user of the advertisement (a coined word or catchphrase that characterizes the advertisement) or an object (for example, a brand name or a product name) stored in the advertisement storage unit 45. The advertisement effect of the advertisement is evaluated (step S10).

  Evaluation can be performed in two stages. In the first stage, the local evaluation unit 80 of each artifact P obtains an evaluation value (individual evaluation value) based on the corresponding user interaction and stores it in the advertisement storage unit 45 (FIG. 4). In the second stage, the evaluation unit A4 of the advertisement server A collects the individual evaluation values from the respective artifacts P, adds them to the average evaluation value, stores them in the advertisement storage unit A1, and updates them thereafter (FIG. 3).

  Many variations are possible for examples where advertisements and objects appear in the interaction. For example, the local evaluation unit 80 requests a re-output of an advertisement from the user to the artifact P in the interaction (for example, “What's the △△ you said yesterday?” Evaluate advertising effectiveness based on

  That is, the local evaluation unit 80 adds the positive evaluation addition value with a predetermined weight to the advertisement effect value (for example, the individual evaluation value in FIG. 4) based on the request for such re-output. To do. Another example of the criteria is that the user responded positively to the interaction or conversation continuation following the output of the advertisement. For example, in response to an utterance of an advertisement from an artifact P, "Do you know the new model of XX company?", The user responded to the conversation like "It seems popular. .

  Another example of the criterion is that the user requested the artificial object P to perform predetermined information processing for an advertisement or an advertisement target (hereinafter referred to as “advertisement etc.”) in an interaction after the advertisement is output. Here, as the predetermined information processing, for example, posting to SNS, that is, social networking service, web search, and the like can be considered. For example, in response to the introduction of a new car that is an advertisement from the artifact P "Do you know the new car of XX company?", As a request for a web search, for example, "What is the price range of the new car?" “Can the image search result of the new model be put on your screen?”

  Still other examples of the criteria are: (a) the user discussed the advertisement again after a predetermined time from the output of the advertisement; (b) the user discussed the advertisement etc. more than a predetermined number of times after the advertisement output. (C) After the advertisement is output, the user has talked about the advertisement etc. at a predetermined frequency or more. (D) After the advertisement is output, the user talks about the advertisement with other users or other artifacts. It is that.

  Yet another example of the criterion is that the user other than the target user who has output the advertisement is also affected. That is, the evaluation unit A4 is a user other than the user (hereinafter referred to as “person”) who is the target of the advertisement output among the users identified by the artifact P after the advertisement is output, The advertisement effect is evaluated based on the fact that a user having the relationship of the above has performed a predetermined action related to an advertisement or the like.

  A user who has a predetermined relationship with the person is, for example, a user who appears in the field of view at the same time or more than the person, as viewed from the artifact P, a user who speaks with the other person's name unilaterally or physically, In addition to the relationship that can be extracted from sensor information and interaction history, such as a user who touches, a living family in which face patterns, voiceprints, and other information are registered in advance in the artifact P or cloud server C may be used.

  In addition to the requests for re-output, the requests for predetermined information processing, and the topics that have been mentioned above as the predetermined actions related to advertisements, etc., there are purchase applications, document requests, advertisement objects, etc. Others are to obtain the product itself and present it in the field of view of the camera of the artifact P.

  By using one or a combination of the above-mentioned criteria, for each advertisement, an individual evaluation value when output from the artifact P to the user is obtained (FIG. 4), and an average evaluation value obtained by tabulating the individual evaluation values ( Fig. 3) Used to select advertisements.

[6. effect〕
As described above, in this embodiment,
(1-1) By determining the intimacy between the artificial object and the user who continuously interacts, and controlling the behavior of the artificial object such as changing information to be output according to the intimacy, It is possible to realize an interaction that matches the relationship.

(1-2) Further, in this embodiment, maintenance management with a burden such as time and labor increases as it becomes more intimate. Therefore, the maintenance management is used as a criterion for determining the familiarity, thereby determining the familiarity more appropriately. it can.

(1-3) Moreover, in this embodiment, since the intimacy increases as the attachment to the object is stronger, the intimacy can be determined more appropriately by using the behavior representing attachment as a criterion for intimacy determination.

(1-4) Further, in the present embodiment, since an opponent to be accompanied by an action related to an important interest of a human tends to be an intimate partner, an important predetermined action is used as a reference for intimacy determination. By doing so, it is possible to more appropriately determine the intimacy between the user and the artifact.

(1-5) Further, in this embodiment, humans are allowed to accompany a partner with a higher intimacy for a long time, far away, even at a high cost. Thus, the intimacy can be determined more appropriately.

(1-6) Also, in this embodiment, a server such as a so-called cloud serving as a back end of AI (artificial intelligence) is compared with the case of determining the intimacy for individual artifacts or users by an absolute standard alone. The intimacy is determined by relative evaluation based on a plurality of artifacts and users. As a result, even if there is a change in the environment such as a social trend or fashion concerning how to contact the artifact, a certain ratio or number can be extracted as having a high degree of familiarity.

  Accordingly, it is possible to realize a sharpness that makes the content of the interaction different among users under any circumstances. For example, when a predetermined advertisement is output as the content of interaction that varies between users, a desired amount of advertisement inventory, that is, an opportunity for advertisement output can be secured. In addition, as a content of interaction that varies among users, when a privilege grant or other promotion measure is applied to a highly intimate user, a desired amount of promotion effect can be ensured.

(1-7) Further, in the present embodiment, a case where a software agent enters a smartphone and interacts with a user while virtually going out with the user is added to the determination of intimacy, thereby making various artificial Intimateness can be determined appropriately according to the object and interaction mode.

(1-8) In this embodiment, the degree of intimacy can be determined in a multifaceted manner by using a software agent as a criterion for intimacy determination, which is handled as an information processing apparatus virtually including an artifact. it can.

(1-9) Further, in this embodiment, the degree of familiarity is particularly high by determining that the degree of familiarity is high when interacting even when the obstacle or resistance is high, such as when going out and active. It becomes easier to clarify the disparity in intimacy between artifacts and users.

(1-10) Also, in this embodiment, since humans are easily affected by interaction with an intimate partner, it is possible to understand that interaction with an artifact in a base also affects user behavior while going out. By using the density determination criterion, the familiarity can be determined appropriately.

(2-1) Moreover, in this embodiment, an advertisement output opportunity can be used effectively by assigning an advertisement having a high evaluation value to a user who has a high degree of closeness with the user. In addition, advertisers can implement a crisp advertising strategy by assigning a large advertising budget to advertising output that can be expected to have a great influence by using a high bid price as the evaluation value for the advertisement that they want to prioritize. .

(2-2) Further, in this embodiment, the higher the bid unit price is, the higher the advertising effect that affects the user is expected by allocating to an artificial object having a higher closeness with the user who is the interaction partner. An advertisement output opportunity that can be used is assigned a high-value advertisement, which is the unit price of advertisement, and the advertisement output opportunity with a high value can be appropriately utilized.

(2-3) Further, in the present embodiment, when the advertisement is output or when a predetermined result is obtained by the output, the advertising unit price that becomes the revenue from the advertiser to the advertising company and the advertisement brings the predetermined result. By using the value multiplied by the rate as the evaluation value, the most effective advertisement allocation can be realized in terms of both the magnitude of the value generated by the advertisement and its ratio.

(2-4) Further, in the present embodiment, the advertisement effect can be expected by storing the selected advertisement, determining a suitable timing, and outputting the advertisement.

(2-5) Further, in the present embodiment, an excellent advertising effect can be obtained by presenting an advertisement (including a recommendation) suitable for the life stage of the user that can be understood by continuous interaction.

(3-1) Further, in the present embodiment, for an artifact that performs interactive interaction with the user, the maximum advertisement can be obtained regardless of the degree of intimacy by switching the output mode of the advertisement according to the intimacy with the user. The effect can be expected.

(3-2) Also, in this embodiment, since the self-speaking of the partner with high intimacy attracts attention, the user and the artificial object with high intimacy say self-speaking about the advertisement, so that the content of the advertisement, the target or the advertisement Can attract the user's interest in itself, and can achieve excellent advertising effects such as strong memorization and high conversion rate.

(3-3) Further, in the present embodiment, it is possible to effectively attract interest to an advertisement by outputting an advertisement when a user with a high degree of closeness is nearby and there is a thing related to the advertisement.

(3-4) Further, in this embodiment, since the user does not always recognize the thing recognized by the artifact and the relationship between the thing and the advertisement, the advertisement is accompanied with the appropriate indication. By outputting, it is possible to assist the user's understanding of why the story has come out, and to receive the advertisement content without any sudden feeling.

(3-5) Further, in the present embodiment, the content of the advertisement related to the display content of the screen viewed by the user is appropriately indicated for the fact that the topic is newly presented or the story is expanded in a daily conversation with a close partner. Talk to it like this. Thereby, without feeling uncomfortable for the user who is browsing the content, it is possible to lead attention to the content of the advertisement, and it is possible to guide to subsequent actions such as recording of the advertisement target, request for material, and purchase application.

(3-6) Moreover, in this embodiment, the element information for every output mode according to the familiarity is included in the advertisement information in advance, and by using the element information according to the actually used output mode, the artifact This eliminates the burden of adapting advertisements according to the output mode.

(3-7) Also, in this embodiment, when the quality is appealed from an intimate partner, it tends to be purchased even at a slightly higher price. By outputting the advertisement in a manner, an excellent advertising effect such as conversion can be obtained.

(3-8) Also, in this embodiment, an excellent advertisement can be obtained by extracting the user's activity pattern from the joint life with the user, judging the timing for effectively transmitting the advertisement (recommendation), and outputting the advertisement. The effect can be expected.

(4-1) Further, in the present embodiment, based on the appearance in the subsequent interactive interaction between the user and the artifact, the advertisement object output to the user from the artifact performing the interactive interaction. By evaluating it, it is possible to appropriately evaluate the advertising effectiveness including memorization according to the characteristic of interactive interaction called continuous communication. In addition, it is possible to appropriately determine the advertisement value and allocate the advertisement by appropriately evaluating the advertisement effect.

(4-2) Also, in this embodiment, when an advertisement is output and then requested to be output again, it is understood that the advertisement or target object has attracted attention. Can be evaluated.

(4-3) Further, in this embodiment, when the user responds positively to the interaction following the advertisement output, it is understood that the user is interested in the advertisement and the object. Can be evaluated appropriately.

(4-4) Also, in this embodiment, after the advertisement is output, it is understood that the user has been interested in the advertisement or the target object when the user requests the artificial object for predetermined information about the advertisement or the target object. Based on the request, the advertising effect can be appropriately evaluated.

(4-5) Further, in this embodiment, when the user has talked about the advertisement and its target object again after a time has passed since the advertisement output, the advertisement and the target object are attracted and recorded. Since it is understood, the advertising effectiveness can be appropriately evaluated based on the request.

(4-6) Further, in the present embodiment, after the advertisement is output, if the user has repeatedly talked about the advertisement or the target object, it is understood that the advertisement or the target object is strongly attracted and recorded. Therefore, the advertising effectiveness can be appropriately evaluated based on the request.

(4-7) Further, in this embodiment, after the advertisement is output, if the user frequently talks about the advertisement or its target, it is understood that the advertisement or target is strongly interested and recorded. Based on the request, the advertising effect can be appropriately evaluated.

(4-8) In the present embodiment, after the advertisement is output, when the user talks about the advertisement or the target object with another person or another artifact, the information is transmitted to the other person or the information is collected from the other person. Since it is understood that the advertisement and the target object are attracted and recorded so that the advertisement effect is recorded, the advertising effect can be appropriately evaluated based on the topic.

(4-9) Also, in this embodiment, after an advertisement is output, if another user related to the user who is the target of the advertisement output takes a predetermined action on the advertisement or the target object, he / she is interested in the other user. Since it is understood that interest and other influences have spread, it is possible to appropriately evaluate the advertising effectiveness based on the facts. In particular, since the purchase of goods and services becomes larger, the intentions of family members and other friends are also affected, so that the spread to other users is significant.

[7. Other embodiments]
In addition, the content of the said embodiment and figure is only an illustration, and the presence or absence and arrangement | positioning of each element, the order and content of a process, etc. can be changed suitably. For this reason, this invention also includes the modification illustrated below and other embodiments. For example, each element shown in FIG. 1 (FIG. 1), the advertisement server A or the cloud server C, or the artifact P can be freely determined according to the implementation.

  Further, the advertisement server A and the cloud server C may be integrated. In addition, as in the above-described embodiment, the present invention is not necessarily realized as a system including a server connected via the communication network N, and includes all functions including determination of intimacy and selection of advertisements. The processing apparatus may be realized as a single artifact P, or may be implemented by being arranged across a plurality of artifacts P operating in cooperation with each other. That is, one or both function groups of the cloud server C and the advertisement server A may be included in the artifact P itself.

  The storage means is not limited to local storage in the apparatus, but may be remote storage such as network computing (cloud). The storage means may include not only a data storage area but also functions such as data input / output and management. The unit of the storage means shown in this application is for convenience of explanation, and the configuration can be divided or integrated as appropriate. Further, in addition to the storage means for specifying, storage means for processing targets, work areas, results, and the like are used as appropriate.

  Note that the arrows in the figure (for example, FIG. 1) supplementarily indicate the main direction of the flow of data, control, etc., and do not mean the denial of other flows or the limitation of the direction. Further, each means other than the storage means is a processing means for realizing and executing the information processing functions and operations described in the above embodiment (for example, FIG. 1), but these are functional units arranged for explanation. It doesn't matter if it matches the actual hardware elements or software modules.

  Moreover, each aspect of the present invention can be understood as other categories (methods, programs, systems including terminals, etc.) that are not specified. In the category of method or program, “means” shown in the category of apparatus is appropriately read as “process” or “step”. In addition, since “XX” (unit, section, module, etc.) shown as an element of the invention or embodiment is not limited to a hardware element or a software element, all or part of “XX means” Can be read as

  Also, the processes and steps shown in the embodiment can be changed by changing the order, executing some of them collectively, or executing them part by part. In addition, CPUs, cores, threads, and the like that implement and execute individual means, processes, and steps may be common, or may differ for each means, process, step, or timing.

  The individual means shown in the present application may be realized by calling a function provided by an external server by an API (Application Program Interface) or network computing (so-called cloud or the like). Furthermore, each element such as means relating to the present invention may be realized by other information processing mechanisms such as a physical electronic circuit as well as a computer control unit.

3 arithmetic control unit 4 storage device 5 communication device 10 input / recognition unit 20 inference response unit 25 history storage unit 30 output control unit 40 local determination unit 45 advertisement storage unit 50 advertisement output unit 60 pattern extraction unit 65 pattern storage unit 70 timing determination Unit 80 local evaluation unit A advertisement server A1 advertisement storage unit A2 selection unit A3 selection unit A4 transmission unit A4 evaluation unit C cloud server C1 inference determination unit C2 action storage unit C3 relative determination unit C4 user information storage unit C5 change unit C6 stage determination Part N Communication network P Artifact SM Sensory / effector group

Claims (10)

An object storage unit that stores an advertisement output to the user from an artifact that performs interactive interaction with the user and the object of the advertisement in association with each other;
In the interaction with the user, an evaluation unit that evaluates the advertisement effect of the advertisement based on the advertisement stored in the object storage unit or the appearance of the object;
An information processing apparatus comprising:   The information processing apparatus according to claim 1, wherein the evaluation unit evaluates the advertisement effect based on a request for the re-output of the advertisement from the user to the artifact in the interaction. .   The information processing apparatus according to claim 1, wherein the evaluation unit evaluates the advertisement effect based on a user's positive response to an interaction following the output of the advertisement.   The evaluation unit evaluates the advertisement effect based on the fact that the user has requested the artificial object to perform information processing predetermined for the advertisement or the target object in an interaction after the output of the advertisement. The information processing apparatus according to any one of claims 1 to 3, wherein the information processing apparatus is characterized.   The evaluation unit evaluates the advertising effect based on the user talking about the advertisement or the target object again after elapse of a predetermined time from the output of the advertisement. The information processing apparatus according to any one of 1 to 4.   The evaluation unit evaluates the advertisement effect based on the user talking about the advertisement or the target object a predetermined number of times or more after the advertisement is output. The information processing apparatus according to any one of claims 5 to 6.   The evaluation unit evaluates the advertising effect based on the user talking about the advertisement or the object at a frequency more than a predetermined frequency after the advertisement is output. The information processing apparatus according to any one of claims 1 to 6.   The said evaluation part evaluates the said advertisement effect based on the said user having talked about the said advertisement or the said object with the other user or another artifact after the said advertisement was output. The information processing apparatus according to any one of 1 to 7.   The evaluation unit is a user other than the user who is the target of the advertisement output among the users identified by the artifact after the advertisement is output, and has a predetermined relationship with the target user. The information processing apparatus according to claim 1, wherein the user has an evaluation of the advertisement effect based on a predetermined action related to the advertisement or the target object. A process of associating and storing an advertisement output to the user from an artifact that performs interactive interaction with the user and the advertisement object;
In the interaction with the user, an evaluation process for evaluating the advertising effectiveness of the advertisement based on the appearance of the stored advertisement or the object;
An information processing method characterized in that a computer executes.

Images