WO2021075337A1 - Dispositif, procédé et programme de traitement d'informations - Google Patents

Dispositif, procédé et programme de traitement d'informations Download PDF

Info

Publication number
WO2021075337A1
WO2021075337A1 PCT/JP2020/038078 JP2020038078W WO2021075337A1 WO 2021075337 A1 WO2021075337 A1 WO 2021075337A1 JP 2020038078 W JP2020038078 W JP 2020038078W WO 2021075337 A1 WO2021075337 A1 WO 2021075337A1
Authority
WO
WIPO (PCT)
Prior art keywords
advertisement
user
information processing
product
information
Prior art date
Application number
PCT/JP2020/038078
Other languages
English (en)
Japanese (ja)
Inventor
範亘 高橋
Original Assignee
ソニー株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by ソニー株式会社 filed Critical ソニー株式会社
Publication of WO2021075337A1 publication Critical patent/WO2021075337A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F13/00—Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
    • G06F3/048—Interaction techniques based on graphical user interfaces [GUI]
    • G06F3/0484—Interaction techniques based on graphical user interfaces [GUI] for the control of specific functions or operations, e.g. selecting or manipulating an object, an image or a displayed text element, setting a parameter value or selecting a range
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G—PHYSICS
    • G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
    • G09F—DISPLAYING; ADVERTISING; SIGNS; LABELS OR NAME-PLATES; SEALS
    • G09F19/00—Advertising or display means not otherwise provided for
    • G—PHYSICS
    • G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
    • G09F—DISPLAYING; ADVERTISING; SIGNS; LABELS OR NAME-PLATES; SEALS
    • G09F27/00—Combined visual and audible advertising or displaying, e.g. for public address

Definitions

  • This technology relates to information processing devices, information processing methods, and information processing programs.
  • the usage status of the user's service, etc. is collected, the preference is analyzed according to the collected usage status, and when the user is determined to be a potential customer of a specific product, the product is transferred.
  • a technique has been proposed in which a user is made to output a guidance message to elicit the interest of the user to draw out the potential request of the user and present more effective advertising information (Patent Document 1).
  • advertisements may be unnecessary information for users.
  • the advertisement may become an obstacle or take the user's time, which may give a high load to the user and cause discomfort or disadvantage.
  • This technology was made in view of these points, and an object of the present technology is to provide an information processing device, an information processing method, and an information processing program capable of presenting an advertisement to a user with a low load. ..
  • the first technique is a first advertisement for a product to be presented to the user and a second advertisement for the product to be presented to the user when the user is interested in the first advertisement. It is an information processing device including an advertisement determination unit for determining an advertisement of the above and a determination unit for determining whether or not the user is interested in the first advertisement.
  • the second technique also determines a first advertisement for the product to be presented to the user and a second advertisement for the product to be presented to the user when the user is interested in the first advertisement, and the user. Is an information processing method for determining whether or not is interested in the first advertisement.
  • the third technique determines a first advertisement for the product to be presented to the user and a second advertisement for the product to be presented to the user when the user is interested in the first advertisement, and the user.
  • Embodiment> [1-1. Configuration of information processing system 10] [1-2. Configuration of agent terminal 100] [1-3. Configuration of information processing device 200] [1-4. Processing by information processing device 200] ⁇ 2. Modification example>
  • the information processing system 10 includes an agent terminal 100 and an information processing device 200.
  • the agent terminal 100 and the information processing device 200 are connected to each other via a network such as the Internet.
  • the agent terminal 100 is a device used by a user who is the target of an advertisement realized by processing by the information processing device 200, and the agent operates.
  • An agent is a user interface technology in which a computer performs information processing and situational judgment based on learning results and / or autonomously, communicates with a user, and executes instructions from the user. Agents are often configured as virtual life forms such as people and animals.
  • the agent is displayed as a CG (Computer Graphics) character or the like on the display unit of the agent terminal 100, and can communicate with the user, present information to the user, present an advertisement to the user, and the like by its operation and voice output.
  • CG Computer Graphics
  • the information processing device 200 operates on a server, a cloud, or the like, and determines an advertisement to be executed on the agent terminal 100 and performs various processes for executing the advertisement.
  • the agent terminal 100 includes a control unit 101, a storage unit 102, an interface 103, an input unit 104, a display unit 105, a speaker 106, a camera unit 107, a microphone 108, and a sensor unit 109.
  • the control unit 101 is composed of a CPU (Central Processing Unit), a RAM (Random Access Memory), a ROM (Read Only Memory), and the like.
  • the CPU controls the entire agent terminal 100 and each part by issuing commands by executing various processes according to the program stored in the ROM.
  • the storage unit 102 is a large-capacity storage medium such as a hard disk or a flash memory. Various applications and data used by the agent terminal 100 are stored in the storage unit 102.
  • the interface 103 is an interface between the agent terminal 100 and the information processing device 200.
  • the interface 103 may include a wired or wireless communication interface.
  • the Internet may intervene between the agent terminal 100 and the information processing device 200.
  • the wired or wireless communication interface is cellular communication such as 3G / LTE, Wi-Fi, Bluetooth (registered trademark), NFC (Near Field Communication), Ethernet (registered trademark), HDMI (registered trademark). (High-Definition Multimedia Interface), USB (Universal Serial Bus), etc. can be included.
  • the interface 103 may include a bus in the device, a data reference in the program module, and the like (hereinafter, these). Is also called the interface inside the device). Also, if the agent terminals 100 are distributed across a plurality of devices, the interface 103 may include different types of interfaces for each device. For example, the interface 103 may include both a communication interface and an interface within the device.
  • the input unit 104 is for the user to input various instructions to the agent terminal 100.
  • a control signal corresponding to the input is generated and supplied to the control unit 101.
  • the control unit 101 performs various processes corresponding to the control signal.
  • the input unit 104 includes a touch panel, voice input by voice recognition, gesture input by human body recognition, and the like.
  • the display unit 105 is a display device such as a display that displays an agent, an advertisement, an image / video, a GUI (Graphical User Interface), and the like.
  • the speaker 106 is a voice output device that outputs voice of an agent, voice accompanying an advertisement, voice message, voice of video, and the like.
  • the camera unit 107 is composed of a lens, an image sensor, a video signal processing circuit, and the like, and is for photographing the surrounding environment of the user and the agent terminal 100.
  • the image / video captured by the camera unit 107 may or may not be used for image recognition processing or the like. Further, an image / video taken by a camera separate from the agent terminal 100 may be used for image recognition processing in the agent terminal 100 or the information processing device 200.
  • the microphone 108 is used by the user to input voice to the agent terminal 100.
  • the user can perform various operations of the agent terminal 100 by voice input using the voice recognition technology. Further, the microphone 108 is used not only for the user's intentional voice input but also for the agent terminal 100 to collect the user's voice and the voice of the environment around the agent terminal 100.
  • the sensor unit 109 is a sensor that detects various types of information by sensing. Sensors include various biosensors that detect biometric information such as heartbeat data, blood flow data, fingerprint data, voiceprint data, face data, and vein data, and various environmental sensors that detect environmental information such as temperature data, humidity data, and illuminance data. Etc. may be included.
  • the sensor unit 109 may acquire an image or sound in the vicinity of the user or the agent terminal 100 as data by using a camera, a microphone, various sensors described above, or the like. Further, the sensor unit 109 may include a position detecting means for detecting an indoor or outdoor position. Specifically, the position detection means includes a GNSS (Global Navigation Satellite System) receiver, for example, a GPS (Global Positioning System) receiver, a GLONASS (Global Navigation Satellite System) receiver, a BDS (BeiDou Navigation Satellite System) receiver, and the like. / Or may include communication devices and the like.
  • GNSS Global Navigation Satellite System
  • GPS Global Positioning System
  • GLONASS Global Navigation Satellite System
  • BDS BeiDou Navigation Satellite System
  • Communication devices include, for example, Wi-fi, MIMO (Multi-Input Multi-Output), cellular communication (for example, position detection using a mobile base station, femtocell), or short-range wireless communication (for example, BLE (Bluetooth Low Energy)). Position is detected using technology such as Bluetooth®.
  • MIMO Multi-Input Multi-Output
  • cellular communication for example, position detection using a mobile base station, femtocell
  • short-range wireless communication for example, BLE (Bluetooth Low Energy)
  • BLE Bluetooth Low Energy
  • the device including the sensor When the above-mentioned sensor detects the position and situation (including biological information) of the user, the device including the sensor is carried or worn by the user, for example. Alternatively, even when a device including a sensor is installed in the user's living environment, it may be possible to detect the user's position and situation (including biological information). For example, the pulse of the user can be detected by analyzing an image including the user's face acquired by a camera fixedly installed indoors or the like.
  • the sensor unit 109 is a processor or process that converts a signal or data acquired by the sensor into a predetermined format (for example, converting an analog signal into a digital signal or encoding image or audio data).
  • the circuit may be included.
  • the sensor unit 109 may output the acquired signal or data to the interface 103 without converting it into a predetermined format. In that case, the signal or data acquired by the sensor is converted into an operation command by the processing unit 200.
  • the software that acquires information from an external service uses, for example, the API (Application Program Interface) of the external service to acquire various types of information provided by the external service.
  • the software may, for example, acquire information from the server of the external service, or may acquire the information from the application software of the service executed on the client device.
  • the software may obtain information such as texts and images posted by the user or other users to external services such as social media, for example.
  • the information acquired does not necessarily have to be intentionally posted by the user or another user, and may be, for example, a log of operations performed by the user or another user.
  • the acquired information is not limited to the personal information of the user or other users, and is not limited to an unspecified number of people such as news, weather forecasts, traffic information, POI (Point Of Interest), or advertisements.
  • the information may be distributed to the user.
  • the information acquired from the external service includes information acquired by the various sensors described above, such as acceleration, angular velocity, orientation, altitude, illuminance, temperature, atmospheric pressure, pulse, sweating, brain wave, touch, smell, taste, and the like.
  • Biological information, emotions, location information, and the like may be detected by a sensor included in another system linked with an external service, and may include information generated by posting to the external service.
  • the agent terminal 100 is configured as described above.
  • the agent terminal 100 includes, for example, an agent-dedicated display device as shown in FIG. In the example of FIG. 3, a humanoid agent is displayed on the display unit 105, and the agent and the user can communicate with each other such as a dialogue.
  • the information processing device 200 includes a context user recognition unit 201, a general context acquisition unit 202, a user information database 203, an advertisement determination unit 204, a matching table 205, an advertisement-related information database 206, an operation generation unit 207, a determination unit 208, and learning.
  • -It is configured to include an evaluation unit 209.
  • the context user recognition unit 201 is a user to be advertised based on an image / video taken by the camera unit 107 of the agent terminal 100, a sound collected by the microphone 108, and sensor information detected by the sensor unit 109. And get the user context.
  • the user context includes analysis of images / videos taken by the camera unit 107 of the agent terminal 100, sensor information acquired by the sensor unit 109, recognition of voice collected by the microphone 108, and history of the Internet accessed via the agent terminal 100. It can be obtained from such as.
  • the user context may be acquired from a device other than the agent terminal 100, for example, a smartphone, a personal computer, a tablet terminal, or other IoT (Internet of Things) device used by the user.
  • a device other than the agent terminal 100 for example, a smartphone, a personal computer, a tablet terminal, or other IoT (Internet of Things) device used by the user.
  • IoT Internet of Things
  • User contexts include user position, user posture (standing, sitting, sleeping, etc.), user behavior (watching TV, eating, drinking, talking). Etc.), what you purchased, what order you canceled, Internet browsing history, mail order site purchase history, TV viewing history, posting and operation history on social media, etc.
  • the word in the conversation may be recognized by voice recognition and the conversation content may be set as the user context.
  • the general context acquisition unit 202 acquires the current general context.
  • General contexts are, for example, date and time, weather, season, temperature, humidity, current affairs (pollen season, elections, Olympics, year-end and New Year, etc.).
  • the general context can be obtained from an information providing service as an external service on the Internet, various service APIs (Application Programming Interfaces), various sensors, various devices, an agent terminal 100, and the like.
  • the user information database 203 stores a user profile input in advance, various information about the user recognized by the context user recognition unit 201, an ID for identifying the user, and the like.
  • the user profile can also be obtained from an external information bank 301, an external information service database 302, or the like.
  • the user information database 203 may be provided as an external service (external service database).
  • User profiles include user name, nationality, language used, age, birthday, gender, address, occupation, hobbies, relationship, purchase history, location information history, friendship, etc.
  • the advertisement determination unit 204 constructs a matching table 205 based on the user profile, the user context, and the general context, and determines the advertisement of the product to be performed by the agent to the user.
  • Products are not limited to anything, but are subject to commercial transactions such as tangible objects such as products and foods, intangible objects such as services, information and programs, contents such as movies and music, and animals and plants such as pets and flowers. Anything can be used.
  • the product may be provided for a fee or may be provided free of charge.
  • the advertisement-related information database 206 stores advertisement operation information, advertisement frequency information, and advertisement prohibition conditions as information related to product advertisement.
  • the advertisement operation information defines the operation of the agent when the agent presents the advertisement as the first advertisement or the second advertisement, which is preset for each product. For example, when the product is drinking water, the action is to drink, when the product is food, the action is to eat, and when the product is a book, the action is to read.
  • the agent service provider needs to set in advance an operation that can sufficiently convey the attractiveness of the advertisement product to the user. It should be noted that a plurality of advertisement operation information may be set for one product, selected according to the user profile, context, and the like, and executed by the agent.
  • the advertisement can also be performed in a mode of displaying an image or a video about the product on the display unit 105 on which the agent is displayed, and the operation of the agent is included here including such a mode.
  • the advertisement frequency information is the setting for each product how often the agent presents the advertisement. For example, drinking water 6 times a day (4 hours before the next ad), food once a day, books once every 3 days, cars once a week, etc. .. In addition, for example, except for the time zone when the user is sleeping (for example, from 0:00 to 8:00) as a predetermined time zone, the advertisement is presented four times a day every four hours for the remaining time. Is also possible. It should be noted that the advertisement by the agent can be set to be once a day regardless of what the product is.
  • the advertisement frequency information may include a time zone in which the advertisement is presented.
  • the advertisement prohibition condition defines the condition that the agent does not present the advertisement in advance based on the user's preference, the user's lifestyle, the social situation, the general social convention, and the like.
  • the first example of the advertisement prohibition condition defines the timing at which the advertisement is not presented by setting a threshold value for the frequency of the prohibited advertisement. For example, the advertisement is not presented for 5 minutes from the previous advertisement.
  • the second example of the advertisement prohibition condition is the time zone when the advertisement is prohibited. For example, advertisements are not presented during the time period from 1:00 am to 7:00 am.
  • the third example of the advertisement prohibition condition is to prohibit advertisement by the condition of social and environmental conditions. For example, advertisements will not be presented in situations where there is a state of emergency declared by the national or local government, a disaster emergency proclamation, or an emergency proclamation based on the police Law. Further, the fourth example of the advertisement prohibition condition prohibits the advertisement based on the specific situation of the user. For example, if the user is crying, making a loud voice above a certain amount, or uttering a specific word (for example, a word related to illness or obituary), the advertisement is not presented. And so on.
  • Advertisement operation information, advertisement frequency information, and advertisement prohibition conditions need to be set in advance by the agent service provider by the information processing device 200.
  • the advertisement prohibition condition may be arbitrarily set by the user who uses the agent terminal 100. Further, the user's preference for advertisement may be learned and updated by machine learning. Further, the advertisement prohibition condition may be set individually for each user, or the same condition may be set for all users when the information processing apparatus 200 targets the information processing device 200. When setting the advertisement prohibition condition for each user, it is necessary to associate the user ID stored in the user information database 203 with the advertisement prohibition condition. Further, the target user may be set for each condition. When set for each user, the advertising behavior for another user is not restricted by the advertising prohibition condition targeting only the user A.
  • the agent service provider needs to set advertisement operation information and advertisement frequency information for the increased product candidates at any time. ..
  • the advertisement frequency information is not essential information. If there is no advertisement frequency information, the advertisement will be performed at a predetermined timing unless the advertisement prohibition condition is met.
  • the motion generation unit 207 generates motion control information for causing the agent to execute the motion of presenting an advertisement in the agent terminal 100.
  • the operation control information is transmitted to the agent terminal 100 via the network.
  • the agent terminal 100 controls the operation of the agent based on the operation control information, so that the advertisement is presented to the user.
  • the determination unit 208 may determine the user's interest by using an existing emotion estimation algorithm using biological information that can be detected by the sensor unit 109.
  • the learning / evaluation unit 209 relearns the learning model based on the user profile, user context, general context, advertisement content, judgment result, etc. acquired in a series of processes for presenting the advertisement to the user, and the matching table 205. It is for rebuilding. This makes it possible to improve the accuracy of advertisements.
  • the information bank 301 manages data by utilizing a system such as PDS (Personal Data Store) based on contracts related to data utilization with individuals, companies, organizations, organizations, etc., and also gives instructions from the above-mentioned individuals, etc. It is a business that provides data to a third party after judging the validity on behalf of an individual, etc. based on the conditions specified in advance.
  • the information bank 301 stores the data transmitted from the data provider (user in this embodiment), and provides the data in response to a request from the data user (agent service provider in this embodiment). To do.
  • the information bank 301 gives the data provider an incentive obtained from the data user in response to the provision of the data to the data user, and also obtains a part of the incentive.
  • the user profile and the like can be obtained from the information bank 301.
  • the information processing device 200 is realized by executing a program, and the program may be installed in a server or the like in advance, or may be distributed by download, storage medium, or the like and installed by an agent service provider. Further, the information processing device 200 may be realized not only by a program but also by combining a dedicated device, a circuit, or the like by hardware having the function.
  • the information processing device 200 communicates with the agent terminal 100 by using a communication function (omitted in FIG. 4) provided in a server or the like on which the information processing device 200 operates.
  • step S101 it is confirmed whether or not the user is within the communication range of the agent. If the user is not within the communication range of the agent, the process ends (No in step S101).
  • the communication range is a range in which the user can recognize the operation of the agent, the voice of the agent, and the like, and is determined by the characteristics of the agent, the performance of the agent terminal 100, and the like. Alternatively, it is set in the agent terminal 100.
  • Whether or not the user is within the communication range of the agent can be determined by analyzing the image around the agent terminal 100 taken by the camera unit 107 of the agent terminal 100 and analyzing the user's voice collected by the microphone 108. It can be done by grasping the position.
  • the agent terminal 100 may determine whether or not the user is within the communication range of the agent, and the information processing device 200 may receive the determination result, or the information processing device 200 may receive the user's position information from the agent terminal 100. May be received and determined.
  • step S102 If the user is within the communication range of the agent (Yes in step S101), then the user ID is confirmed in step S102.
  • the user ID is confirmed by identifying who the user is by face recognition processing on the image taken by the camera unit 107 of the agent terminal 100, voiceprint recognition on the voice collected by the microphone 108, and the like, and the user information database 203 is confirmed. This can be done by checking the stored user ID.
  • step S103 the advertisement determination unit 204 refers to the advertisement prohibition condition of the advertisement-related information database 206 set in advance, and confirms whether or not the current advertisement prohibition condition is not satisfied. If the current advertisement prohibition condition is met, the process ends (No in step S103).
  • step S104 the advertisement determination unit 204 acquires a user profile from the user information database 203 based on the user ID.
  • the user profile may be acquired from the agent terminal 100 or other devices.
  • step S105 the general context acquisition unit 202 acquires the current general context and supplies it to the advertisement determination unit 204.
  • step S106 the context user recognition unit 201 acquires the current user context and supplies it to the advertisement determination unit 204.
  • step S107 the advertisement determination unit 204 determines the advertisement of the product to be presented to the user on the agent terminal 100 by the matching process.
  • the advertisement determination unit 204 constructs a matching table 205 as shown in FIG. 6 based on the user profile, the user context, the general context, and the detected user behavior.
  • the matching table 205 is composed of the user's status information, the product corresponding to the status information, and the user's reaction to the product.
  • User status information is composed of a user profile, a user context, and a general context.
  • the product describes the superordinate concept in FIG. 6, it may be a specific name such as a specific product name, a work name, or a service name.
  • the user status information includes at least the user name, age, time, and behavior. Any information that can be obtained from the user profile, the user context, and the general context may be included in the state information. It should be noted that each item of action shown in FIG. 6 is merely an example, and the present technology is not limited thereto.
  • the user's reaction includes a positive reaction and a negative reaction.
  • the positive reaction and the negative reaction classify the user's reaction to a specific product into positive and negative according to the type of the reaction. Goods and positive and negative reactions are taken from the user context. Therefore, it can be said that the matching table 205 centrally manages the user profile, the user context, the general context, the product, the user's positive reaction, and the negative reaction.
  • Positive reactions are criticisms such as purchasing products on online shopping sites, "Like” posts about products on social media, accessing advertising links about products, making restaurant reservations, etc. It is a reaction that has no positive or negative elements.
  • a negative reaction is a reaction that includes critical and negative elements such as canceling an ordered product on an online shopping site, blocking or muting posts related to the product on social media, and turning off the display of advertisements related to the product. is there.
  • Scores are associated with each of the positive and negative reactions according to the user's behavior with respect to the target. Positive responses are scored positive, such as +2 for purchases, +1 for likes on social media, and +1.5 for access to ad links in the example in Figure 6.
  • negative reactions are given a negative score
  • order cancellation is -2
  • social media mute is -1.5
  • ad removal is -1 and so on.
  • the score associated with each of the positive reaction and the negative reaction needs to be set in advance by the agent service provider. The value of this score is merely an example, and the present technology is not limited thereto.
  • user A is 20 to 25 years old, and is in a state of drinking, eating, and talking between 10 and 11 o'clock. It shows that they took a positive reaction of "purchasing" to the target "beer".
  • the user B is 35 to 40 years old and is talking to the target “coffee” between 13:00 and 14:00. It shows that he made a negative reaction of "cancelling the order".
  • the matching table 205 is constructed by accumulating the state information and the positive and negative reactions of the user in this way, when new state information is input, the matching table 205 is used and the state information is shown. It is possible to calculate a product that is likely to have a positive reaction with a high score in the state of the user. The more information in the matching table 205, the more advertisements for products that can be presented to the user in response to various state information of the user. Therefore, it is preferable that the information processing apparatus 200 continuously updates the matching table 205 while presenting the advertisement to the user.
  • the calculation can be performed using, for example, Factorization Machines (Equation 1), which is a model of supervised learning that handles combination features.
  • Factorization Machines is realistic because it requires only a linear amount of calculation for the number of non-zero elements when learning sparse data and when inferring from sparse feature vectors.
  • Equation 1 y is the calculation result, x is the state, and each parameter is adjusted in the learning process. This makes it possible to find out what products have high scores and are likely to take positive actions from the acquired profile, general context, and user context. Therefore, by having the agent present an advertisement for a product that is most likely to take positive actions, the effectiveness of the advertisement can be enhanced and the advertisement can be efficiently linked to profits.
  • step S108 the motion generation unit 207 generates motion control information for causing the agent to execute the first advertisement for the product at the agent terminal 100.
  • the operation control information is transmitted to the agent terminal 100 via the network.
  • the control unit 101 of the agent terminal 100 that has received the operation control information performs display control on the display unit 105 and / or voice output control from the speaker 106, and causes the agent to execute the presentation of the first advertisement.
  • the advertisement there are a first advertisement and a second advertisement in the advertisement.
  • the first advertisement has a lower cognitive load on the user than the second advertisement. Therefore, it can be said that the second advertisement has a higher cognitive load on the user than the first advertisement.
  • the cognitive load is a load applied to a person when recognizing information, and in the present technology, it is a load applied to a user when recognizing an advertisement presented by an agent. Since an advertisement with a high cognitive load may cause discomfort or disadvantage to the user, in the present technology, the first advertisement with a low cognitive load is first presented to the user, and the second is only when the user is interested. Present an advertisement for.
  • the first advertisement having a low cognitive load examples include an advertisement without audio output, an advertisement with audio output but a volume of less than a predetermined value, an advertisement with an audio output time of less than a predetermined time, and an image / video display.
  • advertisements whose display size is smaller than a predetermined size are advertisements in which the number of characters is less than a predetermined amount with the display of characters, advertisements in which the operation time is less than a predetermined time with the movement of the agent, and advertisements accompanied by gentle music.
  • the second advertisement having a high cognitive load examples include an advertisement with audio output, an advertisement with an audio output volume of a predetermined value or more, an advertisement with an audio output time of a predetermined time or more, and an image / video display.
  • the first advertisement when the product is drinking water, for example, as shown in FIG. 7, the first advertisement is performed by the operation of "the agent only holds the drinking water” or "the agent only drinks the drinking water”. I do. In this way, it is preferable that the first advertisement gives the user the impression of being "subtle”. At this time, the agent does not output voice as an example of a method for reducing the cognitive load on the user.
  • the timing at which the agent presents the first advertisement is determined by the advertisement prohibition condition and the advertisement frequency information.
  • the operation control information of the agent terminal 100 is transmitted according to the advertisement frequency information and at a timing that does not correspond to the advertisement prohibition condition, and the agent is made to execute the presentation of the first advertisement. It should be noted that the agent may be made to execute the first advertisement when it is detected from the sensor information that the user has performed a specific action.
  • the score associated with the user's reaction may be associated with the advertisement frequency information so that the frequency of the first advertisement presentation for the product having a high score may be increased.
  • the reaction to the product "drinking water” is Have a "purchase” and have a high score for that reaction.
  • a specific reaction to a specific product is weighted so that the score becomes high.
  • step S109 the determination unit 208 determines whether or not the user is interested in the first advertisement based on the user's behavior. The determination is made by detecting the user's behavior based on the user context such as the image / video taken by the camera unit 107 of the agent terminal 100, the sensor information detected by the sensor unit 109, and the voice collected by the microphone 108. Can be done.
  • the user's face orientation and line of sight turned to the agent for a certain period of time, the user laughed at the agent, and the user spoke about the product of the first advertisement.
  • an action such as ,, is detected, it is determined that the user is interested in the first advertisement.
  • the action used to determine whether or not the user has the above-mentioned interest is detected for a certain period of time. If not, it can be determined that the user is not interested in the first advertisement.
  • step S110 the advertisement determination unit 204 performs a rematching process to determine a new product for which the agent presents the advertisement.
  • the product "car” was filtered out from the matching table 205, and the product was input from Equation 1 using Factorization Machines described above. Calculate a product that has a high score and is likely to take a positive reaction in the state of the user indicated by the state information.
  • products related to the products excluded by filtering may be excluded at the same time.
  • products such as motorcycles, vehicle insurance, and tourist spots for driving that are related / similar to the car are also excluded from the matching table 205.
  • step S108 the agent is made to present the first advertisement for the product determined by the rematching process. Steps S108 to S110 are repeated until it is determined in step S109 that the user is interested.
  • the rematching process is a process of determining the product to which the first advertisement is presented next. Therefore, depending on the user's behavior, the first advertisement may be presented for a product different from the product determined by rematching.
  • step S109 If the user is interested in the first advertisement in step S109, the process proceeds to step S111 (Yes in step S109). Then, in step S111, the motion generation unit 207 transmits motion control information to the agent terminal 100 via the network in order to present a second advertisement for the product that the user is interested in.
  • the control unit 101 of the agent terminal 100 that has received the operation control information performs display control on the display unit 105 and / or voice output control from the speaker 106, and causes the agent to execute the presentation of the second advertisement.
  • the product when the product is drinking water, a method that is more direct and has a larger amount of information than the first advertisement, such as displaying detailed drinking water information superimposed on the agent as shown in FIG.
  • the second advertisement when the product is drinking water, a method that is more direct and has a larger amount of information than the first advertisement, such as displaying detailed drinking water information superimposed on the agent as shown in FIG.
  • the degree of awareness of the product by the user can be increased, and the user can be given a stronger impression of the product than the above-mentioned first advertisement that "the agent only has drinking water".
  • the agent suggests to the user "Would you like to investigate in detail?" About the product that presented the first advertisement, and if the user instructs to investigate, the information about the product is presented as the second advertisement. You may do it.
  • step S112 the learning / evaluation unit 209 re-learns the learning model based on the user profile, user context, general context, product and advertisement content, determination result, etc. acquired in a series of processes for presenting the advertisement to the user. Learning, reconstruction of matching table 205, etc. are performed. This makes it possible to improve the accuracy of subsequent advertisements. As the reconstruction of the matching table 205, the learning / evaluation unit 209 also adds the user's reaction to the executed first advertisement and the user's reaction to the executed second advertisement as new items (id) in the matching table 205. It is good.
  • the processing in the information processing device 200 is performed as described above.
  • the first advertisement having a low cognitive load is first presented to the user, and when the user is interested in the first advertisement, the second advertisement is presented to the user. This prevents the advertisement from interrupting the actions and tasks performed by the user, the advertisement from getting in the way, and the advertisement taking the user's time, causing discomfort or disadvantage to the user. be able to.
  • the second advertisement is presented only when the user is interested in the first advertisement, the effectiveness of the advertisement can be enhanced. The purpose of the advertisement can be achieved while reducing the disadvantages of the advertisement to the user.
  • the advertisement prohibition condition since the first and second advertisements are not presented when the advertisement prohibition condition is met, the advertisement is not presented at an inappropriate timing, so that the user does not feel uncomfortable.
  • information databases of information banks, agent service providers themselves, and other information service providers can be used.
  • information banks etc.
  • personal information is usually stored in an information bank or the like using a user ID as a key.
  • the data user obtains the personal information from an information bank or the like using the user ID as a key. Therefore, it is possible to obtain permission to use personal information in advance, obtain a user ID at that time, obtain personal information using the user ID as a key, and use it as a user profile.
  • the stored personal information is updated sequentially, so it is necessary to perform acquisition work on a regular basis.
  • a business operator that provides an agent service using this technology can feed back personal information acquired and expanded by the agent service to an information bank, its own service, and other information services to update it sequentially.
  • the information processing device 200 may operate on a server, cloud, or the like as described in the embodiment, or may operate on the agent terminal 100.
  • the agent terminal 100 is not limited to the agent-dedicated display device, and may be a smartphone 120 as shown in FIG. 10A, a television 130 as shown in FIG. 10B, or the like. Further, the agent terminal 100 includes a tablet terminal, a personal computer, a laptop computer, a wearable device, a portable game machine, a smart speaker, a humanoid robot, a pet-type robot, AR (Augmented Reality), VR (Virtual Reality), MR (Mixed Reality), and the like. A head-mounted display used for the above may be used.
  • the agent terminal 100 may be a device that does not have a display unit and only outputs audio.
  • Such devices include, for example, smart speakers.
  • the present technology can be implemented by setting the volume of the voice of the first advertisement having a low cognitive load to a predetermined amount or less and making the volume of the second advertisement louder than that of the first advertisement.
  • the agent when the user is interested in the first advertisement, the agent directly asks the user for details and background of the user's behavior such as "what is being read” and “why is laughing", and the user's response to it is given. It may be used to determine the advertisement to be presented. As a result, it is possible to present the advertisement more accurately and according to the situation.
  • the response from the user itself can be used for other services of the information bank and the agent service provider as valuable information that cannot be known from the existing log.
  • the information processing apparatus 200 includes the motion generation unit 207 and the context user recognition unit 201 in the embodiment, the agent terminal 100 may include them.
  • the state of the product owned by the user may be confirmed by the image / video taken by the camera unit 107, and the first advertisement may be presented according to the state.
  • the product is toothpaste
  • the remaining amount of toothpaste is grasped from the image taken by the camera unit 107
  • the first advertisement is presented when the remaining amount becomes less than a predetermined amount.
  • the behavior, dialogue, and wording when the agent presents the advertisement may be changed by using the score value and the user context. For that purpose, it is necessary to store in advance data such as gestures, lines, and phrases of many patterns in the advertisement-related information.
  • the present technology can also have the following configurations.
  • An advertisement decision unit that determines a first advertisement for a product to be presented to a user, a second advertisement for the product to be presented to the user when the user is interested in the first advertisement, and an advertisement determination unit.
  • An information processing device including a determination unit for determining whether or not the user is interested in the first advertisement.
  • the advertisement determination unit determines the product as a product different from the product advertised in the first advertisement (1). Or the information processing apparatus according to (2).
  • the information processing device determines the product based on the type of reaction of the user to the product candidate associated with the state information.
  • the state information is any or a combination of the user's profile, the user's context, and a general context.
  • the information processing device (4), wherein the type of reaction of the user is a positive reaction and a negative reaction.
  • the advertisement determination unit presets the product candidate, the positive reaction to the product candidate, and the negative reaction based on the context of the user.
  • the information processing device wherein the contents of the first advertisement and the second advertisement are preset for each product candidate.
  • the information processing device wherein the frequencies of the first advertisement and the second advertisement are preset for each product candidate.
  • the information processing device according to any one of (1) to (9), wherein the first advertisement is not accompanied by audio output, and the second advertisement is accompanied by audio output.
  • the first advertisement and the second advertisement are accompanied by the display of information about the product, at least the first advertisement has a smaller display size or a shorter display time than the second advertisement (1).
  • the information processing apparatus according to any one of 10).
  • the first advertisement and the second advertisement are accompanied by audio output of information about the product, at least the first advertisement is lower in volume or has a shorter audio output time than the second advertisement (1).
  • the information processing apparatus according to any one of (11).
  • the advertisement determination unit determines a timing for presenting the first advertisement to the user based on a preset advertisement prohibition condition.
  • the first advertisement and the second advertisement are presented to the user by an agent operating in an agent terminal.
  • the agent is displayed on the agent terminal including at least a display unit.
  • the information processing device according to any one of (1) to (15), wherein the user profile is acquired from an information bank.
  • a first advertisement for the product to be presented to the user and a second advertisement for the product to be presented to the user when the user is interested in the first advertisement are determined.
  • An information processing method for determining whether or not the user is interested in the first advertisement is determined.
  • a first advertisement for the product to be presented to the user and a second advertisement for the product to be presented to the user when the user is interested in the first advertisement are determined.
  • An information processing program that causes a computer to execute an information processing method for determining whether or not the user is interested in the first advertisement.
  • Agent terminal 200 Agent terminal 200 .
  • Information processing device 204 Advertisement decision unit 208 .

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Strategic Management (AREA)
  • General Engineering & Computer Science (AREA)
  • Economics (AREA)
  • Accounting & Taxation (AREA)
  • Marketing (AREA)
  • Development Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Game Theory and Decision Science (AREA)
  • Finance (AREA)
  • General Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Human Computer Interaction (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Information Transfer Between Computers (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

L'invention concerne un dispositif de traitement d'informations équipé des éléments suivants : une unité de détermination de publicité qui détermine une première publicité relative à un produit à présenter à un utilisateur, et une seconde publicité relative au produit à présenter à l'utilisateur lorsque l'utilisateur est intéressé par la première publicité ; et une unité d'évaluation qui évalue si l'utilisateur est intéressé par la première publicité.　FIG. 4 :
PCT/JP2020/038078 2019-10-17 2020-10-08 Dispositif, procédé et programme de traitement d'informations WO2021075337A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2019-189966 2019-10-17
JP2019189966 2019-10-17

Publications (1)

Publication Number Publication Date
WO2021075337A1 true WO2021075337A1 (fr) 2021-04-22

Family

ID=75537463

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2020/038078 WO2021075337A1 (fr) 2019-10-17 2020-10-08 Dispositif, procédé et programme de traitement d'informations

Country Status (1)

Country Link
WO (1) WO2021075337A1 (fr)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114331570A (zh) * 2022-03-09 2022-04-12 广东联讯科技发展股份有限公司 基于大数据的移动终端广告智能推送系统及方法
WO2022269862A1 (fr) * 2021-06-24 2022-12-29 日本電信電話株式会社 Système d'extraction de client, système de gestion d'informations, dispositif de demande d'analyse, procédé d'extraction de client et programme

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2002297626A (ja) * 2001-03-30 2002-10-11 Just Syst Corp 情報提示装置、情報提示プログラム、及び情報提示方法
JP2012508419A (ja) * 2008-11-03 2012-04-05 グーグル・インコーポレーテッド 二次コンテンツ配信システム
WO2019088661A1 (fr) * 2017-10-30 2019-05-09 코오롱인더스트리 주식회사 Dispositif, système et procédé de fourniture de services associés à une publicité et à un achat de produit au moyen d'une technologie d'intelligence artificielle

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2002297626A (ja) * 2001-03-30 2002-10-11 Just Syst Corp 情報提示装置、情報提示プログラム、及び情報提示方法
JP2012508419A (ja) * 2008-11-03 2012-04-05 グーグル・インコーポレーテッド 二次コンテンツ配信システム
WO2019088661A1 (fr) * 2017-10-30 2019-05-09 코오롱인더스트리 주식회사 Dispositif, système et procédé de fourniture de services associés à une publicité et à un achat de produit au moyen d'une technologie d'intelligence artificielle

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2022269862A1 (fr) * 2021-06-24 2022-12-29 日本電信電話株式会社 Système d'extraction de client, système de gestion d'informations, dispositif de demande d'analyse, procédé d'extraction de client et programme
CN114331570A (zh) * 2022-03-09 2022-04-12 广东联讯科技发展股份有限公司 基于大数据的移动终端广告智能推送系统及方法
CN114331570B (zh) * 2022-03-09 2022-05-24 广东联讯科技发展股份有限公司 基于大数据的移动终端广告智能推送系统及方法

Similar Documents

Publication Publication Date Title
JP7111376B2 (ja) プログラムおよび情報処理方法
EP3479588B1 (fr) Dispositif à réalité augmentée, et son fonctionnement
JP5671133B2 (ja) 動的広告コンテンツ選択
US20190318385A1 (en) Presenting an advertisement in a vehicle
US10726438B2 (en) Personalized contextual coupon engine
US20200125322A1 (en) Systems and methods for customization of augmented reality user interface
US20190236361A1 (en) Systems and methods for using persistent, passive, electronic information capturing devices
WO2016195474A1 (fr) Procédé d'analyse d'état complet d'un sujet
US20150242895A1 (en) Real-time coupling of a request to a personal message broadcast system
CN104699746A (zh) 情境感知主动数字助理
EP2917890A2 (fr) Fourniture de schémas d'achat augmentés
WO2015061297A1 (fr) Système de transaction en ligne à mise en correspondance personnelle
CN111512617B (zh) 推荐联系人信息的装置和方法
WO2021075337A1 (fr) Dispositif, procédé et programme de traitement d'informations
US20170316451A1 (en) Information processing apparatus and information processing method
US20190378171A1 (en) Targeted advertisement system
TW201322034A (zh) 結合搜尋引擎服務的廣告播放系統及方法
KR20150142610A (ko) 관계 유형을 고려한 코칭 방법 및 시스템
JP2020091907A (ja) インタラクティブなエクスペリエンスに適合させた自然言語文法
Langheinrich et al. Privacy in mobile and pervasive computing
US10515423B2 (en) Shareability score
JP2016177443A (ja) 情報処理装置及び方法
Du et al. Cross-Cultural Investigation of the Effects of Explanations on Drivers’ Trust, Preference, and Anxiety in Highly Automated Vehicles
CN108657090A (zh) 车载娱乐系统及其方法
JP2016177442A (ja) 情報処理装置及び方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20877707

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20877707

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: JP