Device for advertisement, method for advertisement and program for advertisement

Abstract

This invention provides an advertising device that implements advertisements tailored to the user's needs based on computer-generated dialogue records. The advertising device includes: a dialogue record memory means for storing dialogue record information for each user; a dialogue record analysis means for analyzing the dialogue record information and extracting combinations of topics and products discussed in the dialogue; a model generation means for generating a machine learning model that learns from a set of learning data consisting of combinations of topics and products, and outputs the specific product if the topic is input; a topic extraction means for analyzing a user's dialogue and extracting the topic discussed in the dialogue; a product information generation means for inputting the topic into the learned machine learning model to output the specific product; and an advertising implementation means for performing advertising related to the specific product on the user.

Description

Technical Field

[0001] This invention relates to an advertising technique. Prior Technology

[0002] Advertising refers to the active and widespread dissemination of information related to goods or services to the public in order to promote the sales of goods / services. In the past, despite frequent advertising related to goods or services, it was still difficult to implement effective advertising. In such cases, the development of technologies to improve the effectiveness of advertising is increasing. For example, Patent Document 1 proposes a technology for appropriately determining the delivery of advertisements. [Known Technical Literature] [Patent Literature]

[0003] [Patent Document 1] Japanese Patent No. 6125700 Summary of the Invention

[0004] [Problem to be solved by the present invention] However, the above-mentioned prior art has problems such as the inability to fully grasp customer needs and implement advertising that meets those needs.

[0005] Therefore, in view of the above problems, the object of the present invention is to provide an advertising device that implements advertising that meets the needs of users based on computer conversation records. [Technical means to solve the problem]

[0006] One aspect of the advertising device disclosed in this invention is characterized by comprising: a dialogue record memory means for remembering dialogue record information of each user using a computer; a dialogue record analysis means for analyzing the dialogue record information and extracting the combination of topics and products discussed in the dialogue; a model generation means for generating a machine learning model that learns a set of learning data consisting of the combination of the topic and the product, and outputs the specific product if the topic is input; a topic extraction means for analyzing the dialogue conducted by a user and extracting the topic discussed in the dialogue; a product information generation means for inputting the topic into the machine learning model after it has been learned, so that the specific product is output; and an advertising execution means for performing advertising related to the specific product on the user. [Effects of the Invention]

[0007] The advertising device disclosed in this invention implements advertisements that meet the user's needs based on computer-generated dialogue records. Simple Explanation of the Diagram

[0008] Figure 1 is a schematic diagram showing the advertising device of this embodiment. Figure 2 is a functional block diagram of the advertising device in this embodiment. Figure 3 is a diagram showing an example of the hardware configuration of the advertising device in this embodiment. Figure 4 is a flowchart showing a processing example of the advertising device in this embodiment. Implementation

[0009] Referring to the drawings, the configurations used to implement the present invention are described. (The operating principle of the advertising device in this embodiment)

[0010] Using Figures 1 and 2, the operating principle of the advertising device (hereinafter referred to as "the device") 100 of this embodiment will be explained. Figure 1 is a diagram showing the connection relationship between the device 100 and other devices, and Figure 2 is a functional block diagram of the device 100.

[0011] As shown in Figure 1, this device 100 is connected to the user terminal 310 operated by the user 300 and other devices 320 for advertising delivery via a communication network 330. The communication network 330 can be either wired or wireless. The user terminal 310 can be a portable information terminal such as a smartphone or a personal computer.

[0012] As shown in Figure 2, this device 100 has a learning data memory means 110, a model information memory means 120, a dialogue record memory means 130, a purchase record memory means 140, a dialogue record analysis means 150, a model generation means 160, a topic extraction means 170, a product information generation means 180, and an advertising implementation means 190. Learning materials and memory techniques 110, memory techniques described later, learning materials collection 270. Model information memory method 120, memory of the parameters of the action of the machine learning model 280 as described later.

[0013] Dialogue record storage method 130, 300 per user, storing dialogue record information 230 via computer. Dialogue record storage method 130, 300 per user, storing dialogue record information 230 between the user and the dialogue system. Purchase record memory method 140, remembers 300 purchase record information for each user 290.

[0014] Alternatively, the device 100 may also be configured such that it does not have direct memory means 110, 120, 130, 140, but instead utilizes external memory means 110, 120, 130, 140 connected via a communication network 330.

[0015] The dialogue record analysis method 150 analyzes the dialogue record information 230 stored in the dialogue record memory method 130 and extracts the combination 260 of the topics 240 and products 250 discussed in each dialogue 220.

[0016] Furthermore, the dialogue record analysis method 150 analyzes the dialogue record information 230 stored in the dialogue record memory method 130 and the purchase record information 290 stored in the purchase record memory method 140. The dialogue record analysis method 150 may also take the form of extracting the combination 260 of the topics discussed in each dialogue 220 and the goods 250 purchased by the speakers 300 within a predetermined period from the start of the dialogue 220.

[0017] Model generation means 160 generates a machine learning model 280, which learns from a learning dataset 270 composed of combinations 260 extracted by dialogue record analysis means 150. If a topic 240 is input, a specific product 250 is output. Furthermore, the learning algorithm in model generation means 160 is not particularly limited.

[0018] Topic extraction method 170 analyzes a conversation 220 conducted by a user 300 and extracts a topic 240 discussed in the conversation 220. Topic extraction method 170 analyzes a conversation 220 conducted by a user 300 and extracts a topic 240 discussed in the conversation 220.

[0019] The product information generation method 180 inputs a topic 240 extracted by the topic extraction method 170 into the machine learning model 280 after it has been learned, and outputs a specific product 250.

[0020] The advertising implementation method 190 performs the following processing: It displays an advertisement 210 related to a specific product 250 obtained from the product information generation method 180 to a user 300. The advertisement 210 can be implemented directly by the device 100, or it can be a form in which the advertising implementation method 190 instructs an external device 320 to implement the advertisement 210. Furthermore, the advertisement 210 can be any advertising medium or method; for example, it can also be delivered to a user terminal 310 operated by a user 300 by specifying user terminal identification information (e.g., advertisement ID, advertisement identification code). Based on the above operating principle, this device 100 uses computer dialogue records 230 to display advertisements 210 that meet the needs of user 300. (Hardware configuration of the advertising device in this embodiment)

[0021] Using FIG3, an example of the hardware configuration of the present device 100 will be described. FIG3 is a diagram showing an example of the hardware configuration of the present device 100. As shown in FIG3, the present device 100 includes: CPU (Central Processing Unit) 510, ROM (Read-Only Memory) 520, RAM (Random Access Memory) 530, auxiliary memory device 540, communication interface (Interface (I / F)) 550, input device 560, display device 570, and memory media interface 580.

[0022] CPU 510 is a device that executes programs stored in ROM 520. Following the program's instructions, it processes the data loaded into RAM 530 and controls the entire device 100. ROM 520 stores the programs and data executed by CPU 510. RAM 530, when CPU 510 executes the programs stored in ROM 520, loads the executed programs and data and temporarily stores the processing data during the operation.

[0023] The auxiliary memory device 540 is a device that stores the basic software, i.e., the OS (Operating System), the applications of this embodiment, and related data. The auxiliary memory device 540 may be, for example, an HDD (Hard Disk Drive) or flash memory, and includes a learning data storage method 110, a model information storage method 120, a conversation record storage method 130, and a purchase record storage method 140.

[0024] The communication interface 550 is used to connect to a wired / wireless LAN (Local Area Network), Internet or other communication network 330, and to send and receive data with other devices 310 and 320 that provide communication functions.

[0025] Input device 560 is a device such as a keyboard used for inputting data into this device 100. Display device (output device) 570 is a device such as an LCD (Liquid Crystal Display) that functions as a user interface when the user utilizes the functions of this device 100 or makes various settings. Memory media interface 580 is an interface used for receiving and transmitting data with memory media 590 such as CD-ROM, DVD-ROM, and USB memory.

[0026] The various means provided by this device 100 can be configured to enable the CPU 510 to execute programs corresponding to each means stored in the ROM 520 or auxiliary memory device 540. Furthermore, the various means provided by this device 100 can also be configured to implement the related processing of each means as hardware. Additionally, it can be configured to read the program of this invention from an external server device via the communication interface 550, or read the program of this invention from the memory medium 590 via the memory medium interface 580, thereby enabling the device 100 to execute the program. (Example of processing performed using an advertising device in this embodiment) Using Figure 4, a processing example performed by this apparatus 100 will be described. Figure 4 is a flowchart showing the process of the processing example performed by this apparatus 100.

[0027] In S10, the dialogue record analysis method 150 analyzes the dialogue record information 230 stored in the dialogue record memory method 130 and extracts the combination 260 of the topics 240 and products 250 discussed in each dialogue 220.

[0028] Furthermore, the dialogue record analysis method 150 analyzes the dialogue record information 230 stored in the dialogue record memory method 130 and the purchase record information 290 stored in the purchase record memory method 140. The dialogue record analysis method 150 may also take the form of extracting the combination 260 of the topics discussed in each dialogue 220 and the goods 250 purchased by the speakers 300 within a predetermined period from the start of the dialogue 220.

[0029] In S20, the model generation means 160 generates a machine learning model 280, which learns from the learning data set 270 composed of the combinations 260 extracted in S10. If the input topic 240 is given, the specific product 250 is output. Furthermore, the learning algorithm in the model generation means 160 is not particularly limited.

[0030] In S30, topic extraction method 170 analyzes the dialogue 220 being conducted by a user 300 and extracts a topic 240 discussed in the dialogue 220. Topic extraction method 170 analyzes the currently ongoing dialogue 220 of a user 300 and extracts a topic 240 discussed in the dialogue 220.

[0031] In S40, the product information generation method 180 inputs a topic 240 extracted in S30 into the machine learning model 280 learned in S20, so that a specific product 250 is output.

[0032] In S50, the advertising implementation means 190 performs the following process: an advertisement 210 related to a specific product 250 obtained in S40 is displayed to a user 300. The advertisement 210 can be displayed directly by the device 100, or it can be a form in which the advertising implementation means 190 instructs an external device 320 to display the advertisement 210. Furthermore, the advertisement 210 can be any advertising medium or method; for example, it can also be in the form of specifying user terminal identification information (e.g., advertisement ID, advertisement identification code) and delivering the advertisement 210 to the user terminal 310 operated by the user 300.

[0033] By performing the process described above, this device 100, based on the computer's dialogue record 230, implements an advertisement 210 that meets the needs of the user 300.

[0034] While the embodiments of the present invention have been described in detail above, the present invention is not limited to these specific embodiments. Various modifications and alterations can be made within the scope of the spirit of the present invention as described in the claims.

[0035] 100: Advertising installations 110: Learning Material Memorization Techniques 120: Model Information Memory Methods 130: Dialogue Recording Memory Techniques 140: Purchase recording and memory methods 150: Dialogue Record Analysis Methods 160: Model Generation Methods 170: Topic Extraction Techniques 180: Methods of Generating Product Information 190: Advertising Implementation Methods 210: Advertisement 220: Dialogue 230: Conversation Record Information 240: Topic 250:Commodity 260: The combination of topics and products 270: Collection of Learning Materials 280: Machine Learning Model 290: Purchase Record Information 300: User (Interlocutor) 310: User Terminal 320: Advertising delivery device 330: Communication Network 510: CPU 520:ROM 530: RAM 540: Auxiliary Memory Device 550: Communication Interface 560: Input device 570: Output device 580: Memory Media Interface 590: Memory Media

Claims

1. An advertising device comprising: a dialogue record memory means for memorizing dialogue record information of each user using a computer; a dialogue record analysis means for analyzing the dialogue record information and extracting combinations of topics and products discussed in the dialogue; a model generation means for generating a machine learning model that learns a set of learning data consisting of combinations of topics and products, and outputs a specific product if a topic is input; a topic extraction means for analyzing a dialogue conducted by a user and extracting a topic discussed in the dialogue; a product information generation means for inputting the topic into the learned machine learning model to output a specific product; and an advertising execution means for performing advertising related to the specific product on the user.

2. The advertising device as requested in item 1, wherein, The system includes a purchase record memory mechanism that stores purchase record information for each user; a dialogue record analysis mechanism that analyzes the dialogue record information and the purchase record information to extract the combination of the topic and the goods purchased by the speakers within a predetermined period from the start of the dialogue; and a model generation mechanism that enables the machine learning model to learn a set of learning data consisting of the combination of the topic and the goods purchased by the speakers within a predetermined period from the start of the dialogue.

3. The advertising device as requested in item 1, wherein, It has: a learning data memory mechanism to memorize the learning data set; and a model information memory mechanism to memorize the parameters that specify the actions of the machine learning model.

4. An advertising method implemented in a computer equipped with a dialogue record memory device; the dialogue record memory device remembers dialogue record information of each user using the computer; the advertising method is characterized by comprising the following steps: using a dialogue record analysis device to analyze the dialogue record information and extract the combination of topics and products discussed in the dialogue; using a model generation device to generate a machine learning model, the machine learning model learning a set of learning data consisting of the combination of the topic and the product, and outputting the specific product if the topic is input; using a topic extraction device to analyze a dialogue conducted by a user and extract the topic discussed in the dialogue; using a product information generation device to input the topic into the machine learning model after learning, and output the specific product; and using an advertising implementation device to perform processing to display an advertisement related to the specific product to the user.

5. The advertising method as described in request item 4, wherein, The computer has a purchase record memory mechanism that stores the purchase record information of each user; a dialogue record analysis mechanism that analyzes the dialogue record information and the purchase record information, and extracts the combination of the topic and the goods purchased by the interlocutor within a predetermined period from the start of the dialogue; and a model generation mechanism that enables the machine learning model to learn a set of learning data consisting of the combination of the topic and the goods purchased by the interlocutor within a predetermined period from the start of the dialogue.

6. The advertising method as described in request item 4, wherein, The computer has a learning data memory method for storing the learning data set, and a model information memory method for storing the parameters that define the actions of the machine learning model.

7. An advertising program for causing a computer to perform an advertising method as described in any one of claims 4 to 6.

Images