Keyword output system, keyword output method and program

Abstract

To provide a keyword output system, a keyword output method and a program for efficiently selecting keywords appropriate for advertisers (clients) who place search-linked advertisements. [Solution] In a keyword output system that is an information processing system in which a server device (100), a client device (110), a user terminal device (120), and a server device (130) are communicatively connected via a network (150), the server device has at least one or more control units. The control unit acquires telephone call data. The call data is telephone call data of the client. The control unit also generates text information by transcribing the call data, acquires keywords to be targeted for the client's search-related advertising based on the text information, and outputs the keywords.

Description

[Technical Field]

[0001] The present invention relates to a keyword output system, a keyword output method, and a program. [Background technology]

[0002] One type of internet advertising is search engine advertising, which displays ads related to keywords searched by general users on search engines on the search results screen. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2024-024188 Summary of the Invention [Problem to be solved by the invention]

[0004] It is a difficult problem for advertisers (clients) who place search-related ads to efficiently select appropriate keywords. [Means for solving the problem]

[0005] According to one aspect of the present invention, there is provided a keyword output system. The keyword output system has at least one or more control units. The control unit acquires telephone call data. The call data is telephone call data of a client. The control unit generates text information by transcribing the call data. The control unit acquires keywords to be targeted for search-related advertising for the client based on the text information. The control unit outputs the keywords. [Brief explanation of the drawings]

[0006] [Figure 1] FIG. 1 is a diagram illustrating an example of a system configuration of an information processing system. [Figure 2]FIG. 2 is a diagram illustrating an example of a hardware configuration of the server device. [Figure 3] FIG. 3 is a flowchart illustrating an example of information processing in the information processing system. [Figure 4] FIG. 4 is a flowchart showing an example of information processing in the information processing system of the first modified example. [Figure 5] FIG. 5 is a flowchart showing an example of information processing in the information processing system of the second modification. [Figure 6] FIG. 6 is a flowchart showing an example of information processing in the information processing system of the third modified example. DETAILED DESCRIPTION OF THE INVENTION

[0007] Hereinafter, the embodiments will be described with reference to the drawings. Various features shown in the following embodiments (including the modified examples shown below, the same applies hereinafter) can be combined with each other.

[0008] <Embodiment 1> 1. System Configuration FIG. 1 is a diagram illustrating an example of a system configuration of an information processing system 1000. As illustrated in FIG. 1, the information processing system 1000 includes, as a system configuration, at least a server device 100, a client device 110, a user terminal device 120, and a server device 130. The server device 100, the client device 110, the user terminal device 120, and the server device 130 are communicatively connected via a network 150. The network 150 includes a wide area network (WAN) and / or a local area network (LAN) constructed using wired and / or wireless methods. The network 150 also includes a telephone network. The information processing system 1000 is an example of a keyword output system.

[0009] The server device 100 is a device that executes the main processes described in the specification, and also provides the function of an automated telephone answering service.

[0010] The client device 110 is a device of a user company (hereinafter simply referred to as a company) that uses the functions of the telephone automated response service. The user terminal device 120 is the terminal device of a user who calls the company.

[0011] The server device 130 is a device that provides services for a large-scale language model. The large-scale language model is a computer language model composed of an artificial neural network with a large number of parameters, and is a language model trained by self-supervised learning or semi-supervised learning using a huge amount of unlabeled text.

[0012] Although only one user terminal device 120 is shown in FIG. 1 for simplicity, the information processing system 1000 may include a plurality of user terminal devices 120 and user terminals 130 .

[0013] In addition, in the example of FIG. 1, smartphones are illustrated as examples of the client device 110 and the user terminal device 120, but the present invention is not limited to smartphones.

[0014] Here, the claimed information processing system may be composed of multiple devices or may be composed of a single device. When the claimed information processing system is composed of a single device, an example of that device is server device 100. When the claimed information processing system is composed of a plurality of devices, an example of the plurality of devices is a cloud system that provides the functions of server device 100. The cloud system is composed of, for example, a plurality of server devices.

[0015] 2. Hardware Configuration 2 is a diagram showing an example of the hardware configuration of server device 100. Server device 100 includes, as its hardware configuration, a control unit 210, a storage unit 220, a communication unit 230, and an internal bus 240. Control unit 210, storage unit 220, and communication unit 230 are electrically connected via internal bus 240.

[0016] The control unit 210 is a CPU (Central Processing Unit) or the like, and controls the entire server device 100 .

[0017] The memory unit 220 is one of an HDD (Hard Disk Drive), a ROM (Read Only Memory), a RAM (Random Access Memory), an SSD (Solid State Drive), or any combination thereof, and stores programs and data (e.g., call data, blacklists, etc., as described below) used by the control unit 210 when executing processing based on the programs.

[0018] The control unit 210 reads out a program stored in the storage unit 220 and executes processing based on the read program, thereby realizing the processing of the flowcharts in Figs. 3 to 5 described below and the functions of the server device 100. The storage unit 220 is an example of a storage medium. Note that, in the first embodiment, the data used when the control unit 210 executes processing based on the program is described as being stored in the storage unit 220, but the data may also be stored in a storage unit of another device that can communicate with the server device 100. In other words, the data may be stored in a storage unit of any device as long as the data can be referenced and / or acquired by the control unit 210.

[0019] The communication unit 230 connects the server device 100 to the network 150 and controls communication with other devices. The number of each hardware component constituting the server device 100 is not limited to one, and two or more components may be present as necessary. Also, the server device 100 may have hardware components other than those shown in FIG.

[0020] 3. Information Processing The information processing according to the first embodiment will be described below.

[0021] (1) Automatic telephone answering service In response to a request, the control unit 210 controls the company's device (client device 110) to display a setting screen related to the telephone automated answering service. The control unit 210 can create multiple call flows in response to setting operations via the setting screen. A call flow is a series of steps that determine how the server device 100 responds when a customer calls and which route the process will take. The control unit 210 can select from a voice guidance function, an SMS (Short Message Service) sending function, a call forwarding function, a voice recording function, etc. as a branching response after a push.

[0022] The voice guidance function uses AI (Artificial Intelligence) to automatically read out text created according to the business content of the company. The SMS sending function automatically sends pre-set text via SMS to the receiving phone number. This function can be used for a variety of purposes, such as providing the address of a company (store), directions such as routes, event information, and directing to an inquiry form. The call forwarding function forwards calls to a person in charge according to the requirements, date and time, etc. The voice recording function accepts customer requirements via voice recording (answering machine function). Companies can play back the voice data via the settings screen and respond by calling back depending on the content. With the automatic telephone answering service, all calls between companies and customers are recorded.

[0023] (2) Overview of keyword output processing The control unit 210 acquires call data from the automated telephone answering service. The call data is data on calls between a client (company) and a customer. The control unit 210 generates text information by transcribing the call data. The control unit 210 acquires keywords to be targeted for search-linked advertising for the client based on the text information. The control unit 210 outputs the keywords. By adopting such a configuration, it is possible to efficiently select keywords appropriate for advertisers (clients) who place search-related advertisements.

[0024] (3) Details of keyword output processing FIG. 3 is a flowchart showing an example of a keyword output process in the information processing system 1000. In step S310, the control unit 210 acquires customer identification information and call data stored in association with customer identification information that identifies a customer in the storage unit 220 or the like. The call data is call data for a client's phone call. The customer identification information is, for example, the customer's telephone number.

[0025] In step S320, the control unit 210 generates text information by transcribing the call data based on the acquired call data. For example, the control unit 210 inputs the acquired call data into a trained model. The trained model is trained using the call data as input data and the text information transcribed from the call data as output data. The control unit 210 receives the text information output from the trained model.

[0026] In addition, if the text information of the call data has already been generated in the processing related to the automatic telephone answering service and the text information has been stored in the memory unit 220 or the like in association with the customer identification information, the control unit 210 may be configured to acquire the text information of the call data from the memory unit 220 or the like.

[0027] In step S330, the control unit 210 generates a prompt for the large-scale language model provided by the server device 130. The prompt includes the text information generated in step S320. That is, the control unit 210 generates a prompt including the text information. The prompt further includes an instruction to extract keywords to be targeted by the client's search-related advertising from the text information. More preferably, the prompt includes an instruction to extract keywords and an instruction to generate keywords similar to the extracted keywords (hereinafter also referred to as expanded keywords).

[0028] The control unit 210 transmits the generated prompt to the service providing device 130 and requests acquisition of data corresponding to the prompt. Upon receiving the prompt, the large-scale language model of the service providing device 130 outputs data based on the instructions included in the prompt. The output data includes multiple keywords targeted by the client's search-related advertising and their respective expansion keywords. Hereinafter, the multiple keywords and their respective expansion keywords will also be referred to simply as a keyword group.

[0029] In step S340, control unit 210 determines whether or not it has received data (keyword group) output from service providing device 130. If control unit 210 determines that it has received the data output from service providing device 130, it proceeds to step S350. If control unit 210 determines that it has not received the data output from service providing device 130, it repeats the process of step S340. Note that if it has not been able to receive the data output from service providing device 130 even after repeating the process of step S340 for a predetermined time, control unit 210 ends the process of the flowchart shown in FIG. 3. The processing of step S330, or steps S330 and S340, is an example of processing in which the control unit 210 acquires candidate keywords to be targeted by a client's search-related advertising based on text information. The processing of step S330, or steps S330 and S340, can also be considered an example of processing in which data output from a large-scale language model based on a prompt is acquired as keywords.

[0030] In step S350, the control unit 210 temporarily stores the keyword group received from the service providing device 130 in the storage unit 220 or the like. The keyword group temporarily stored in the storage unit 220 or the like can also be considered keyword candidates. The control unit 210 then deletes from the keyword candidates keywords included in a blacklist stored in a predetermined storage area such as the storage unit 220. Here, the blacklist is a list of keywords that are inappropriate as keywords targeted by a client's search-based advertising. Keywords that are inappropriate as keywords targeted by a client's search-based advertising include, for example, malicious words, obscene words, etc. The blacklist is an example of a list of predetermined keywords to be excluded. The process of step S350 can also be considered an example of a process in which control unit 210 selects a keyword from multiple keyword candidates based on the list.

[0031] In step S360, the control unit 210 deletes keywords from the keyword candidates based on the search volume of each of the multiple keywords included in the keyword candidates stored in the storage unit 220 or the like. Search volume refers to the number of searches performed on a search engine within a predetermined period (e.g., within the past month). The search volume of a keyword searched 10,000 times in one month is "10,000." For example, the control unit 210 obtains the search volume of each of the multiple keywords from a predetermined system. Here, the predetermined system is a system that outputs the search volume for each keyword. An example of a predetermined system is a keyword planner or the like. For example, the control unit 210 can obtain the search volume of each of the multiple keywords via an API (Application Programming Interface) provided by the keyword planner or the like.

[0032] The control unit 210 deletes keywords with a search volume equal to or greater than a first threshold from the keyword group. The first threshold is a threshold (e.g., 100,000, etc.) that indicates that the search volume is too high. Keywords with too high a keyword volume can display advertisements to many users, but competition is also fierce. The control unit 210 deletes keywords with a search volume less than a second threshold from the keyword group. The second threshold is a threshold (e.g., 50, etc.) that indicates that the search volume is too low. This is because few users search for the relevant keywords in the first place, making it difficult to expect advertising effectiveness. The first threshold and second threshold are assumed to be stored in advance in the storage unit 220, etc.

[0033] The process of step S360 is an example of a process for acquiring search volume information for each of multiple keyword candidates, and for selecting a keyword from multiple keyword candidates based on the search volume information.

[0034] In step S370, the control unit 210 deletes keywords that appear frequently and have low relevance, which are included in the text information of the call data, from the keyword candidates temporarily stored in the storage unit 220, etc. For example, the control unit 210 determines the relevance between keywords using a predetermined model. Examples of the predetermined model include BERT (language model), a word2vec trained model, etc.

[0035] The process of step S370 is an example of a process for selecting keywords from multiple keyword candidates by excluding keywords that appear frequently and have low relevance to keywords included in the text information from the keyword candidates. The process of step S370 is also an example of a process for excluding keywords that appear frequently and have low relevance to keywords included in the text information from the keyword candidates using a predetermined model. The predetermined model is a model that outputs the relevance between input keywords.

[0036] In step S380, the control unit 210 outputs the keyword candidates temporarily stored in the storage unit 220 or the like as keywords to be targeted by the client's search-linked advertising. The control unit 210 may output the keywords by transmitting the keyword candidates to the client device 110 with which communication is possible via the network 150, or may output the keywords by describing the keyword candidates in a file or the like and storing them in a predetermined recording area such as the storage unit 220. The control unit 210 may also output the keywords by setting the keyword candidates temporarily stored in the storage unit 220 or the like in a predetermined system as keywords to be targeted by the client's search-linked advertising.

[0037] The process of step S380 is an example of a process for outputting a keyword.

[0038] According to the first embodiment, it is possible to efficiently select keywords that are appropriate for advertisers (clients) who place search-related advertisements.

[0039] (Variation 1) The following describes Modification 1 of Embodiment 1. Modification 1 is included in Embodiment 1, but is not a different embodiment from Embodiment 1. Configurations and processes not described in Modification 1 are the same as those in Embodiment 1.

[0040] FIG. 4 is a flowchart showing an example of information processing in the information processing system of the first modified example. In step S410, the control unit 210 of Modification Example 1 (hereinafter simply referred to as the control unit 210) acquires the company's web data. The web data is data stored on a computer called a web server and transmitted to a client such as a web browser using the HTTP protocol. Examples of web data include text data, image data, video data, audio data, and structured data. The control unit 210 acquires the company's web data using, for example, web scraping technology. The process of step S410 is an example of a process for acquiring data of a client's web page.

[0041] In step S420, control unit 210 generates a prompt for the large-scale language model provided by server device 130. Control unit 210 generates the prompt based on the text information generated in step S320 and the web data acquired in step S410. For example, the prompt includes an instruction to extract keywords from the text information that will be targeted for the client's search-based advertising, taking into account the company's web data.

[0042] The processing in step S420, or steps S420 and S340, is an example of processing for acquiring keywords based on text information and web page data.

[0043] According to the first modification, keywords can be acquired that take into consideration data on a company's web page.

[0044] (Variation 2) The following describes Modification 2 of Embodiment 1. Modification 2 is included in Embodiment 1, and is not a different embodiment from Embodiment 1. Configurations and processes not described in Modification 2 are the same as those in Embodiment 1.

[0045] FIG. 5 is a flowchart showing an example of information processing in the information processing system of the second modification. In step S510, the control unit 210 of Modification 2 (hereinafter simply referred to as the control unit 210) acquires business information about the company. The business information about the company is information that indicates what kind of business the company is engaged in, what products and / or services it offers, and how it earns revenue. The control unit 210 may acquire the business information about the company from the company's web data, or may acquire the business information based on the results of a survey of company personnel, etc. The process of step S510 is an example of a process for acquiring business information indicating the content of the client's business.

[0046] In step S520, the control unit 210 generates a prompt for the large-scale language model provided by the server device 130. The control unit 210 generates the prompt based on the text information generated in step S320 and the client's business details acquired in step S510. For example, the prompt includes an instruction to extract keywords from the text information that will be targeted for the client's search-based advertising, taking into account the business information of the company.

[0047] The processing in step S520, or steps S520 and S340, is an example of processing for acquiring keywords based on character information and business information.

[0048] According to the second modification, keywords can be acquired that take into consideration the business information of the company.

[0049] (Variation 3) The following describes Modification 3 of Embodiment 1. Modification 3 is included in Embodiment 1, and is not a different embodiment from Embodiment 1. Configurations and processes not described in Modification 3 are the same as those in Embodiment 1.

[0050] FIG. 6 is a flowchart showing an example of information processing in the information processing system of the third modified example. The flowchart in FIG. 6 includes step S610 in addition to the flowchart in FIG. In step S610, the control unit 210 generates text for a search-linked advertisement related to the keyword output in step S380. For example, the control unit 210 generates a prompt including a keyword and an instruction to generate text for a search-related advertisement to be displayed in conjunction with the keyword, and transmits the generated prompt to the large-scale language model.The control unit 210 then receives the text for the search-related advertisement to be displayed in conjunction with the keyword, which is output from the large-scale language model.

[0051] In step S620, the control unit 210 outputs the generated text of the search-related advertisement. The control unit 210 may output the text of the search-related advertisement by transmitting the text of the search-related advertisement to the client device 110 that is capable of communicating via the network 150, or may output the text of the search-related advertisement by writing the text of the search-related advertisement in a file or the like and storing it in a predetermined storage area such as the memory unit 220. The control unit 210 may also output keywords by setting the text of the search-related advertisement that is temporarily stored in the memory unit 220 or the like in a predetermined system as a document of the client's search-related advertisement.

[0052] According to the third modification, it is possible to generate text for a search-related advertisement relating to a keyword.

[0053] <Additional Notes> The present embodiment includes the following disclosure.

[0054] (Appendix 1) A keyword output system, having at least one or more control units, The control unit Get phone call data, The call data is call data of a client's telephone; generating text information by transcribing the call data; acquiring keywords to be targeted by the client's search-related advertising based on the character information; outputting the keyword; Keyword output system.

[0055] (Appendix 2) 10. The keyword output system according to claim 1, The control unit Obtaining business information indicating the business content of the client; acquiring the keywords based on the character information and the business information; Keyword output system.

[0056] (Appendix 3) 10. The keyword output system according to claim 1, The control unit Obtaining data of the client's web page; acquiring the keywords based on the character information and the data of the web page; Keyword output system.

[0057] (Appendix 4) 10. The keyword output system according to claim 1, The control unit generating a prompt including the textual information; the prompt further includes an instruction to extract keywords to be targeted for the client's search advertising from the text information; acquiring data output from a large-scale language model based on the prompt as the keyword; Keyword output system.

[0058] (Appendix 5) 5. The keyword output system according to claim 4, The prompt includes an instruction to extract the keyword and an instruction to generate a keyword similar to the extracted keyword. Keyword output system.

[0059] (Appendix 6) 10. The keyword output system according to claim 1, The control unit acquiring a plurality of keyword candidates to be targeted for the client's search-related advertising based on the character information; Obtaining search volume information for each of the plurality of keyword candidates; selecting a keyword from the plurality of keyword candidates based on the search volume information; Keyword output system.

[0060] (Appendix 7) 10. The keyword output system according to claim 1, The control unit acquiring a plurality of keyword candidates to be targeted for the client's search-related advertising based on the character information; selecting a keyword from the plurality of keyword candidates by excluding keywords that appear frequently and keywords that are less relevant to the keywords included in the character information from the keyword candidates; Keyword output system.

[0061] (Appendix 8) 8. The keyword output system according to claim 7, The control unit Using a predetermined model, keywords that appear frequently and keywords that are less relevant to the character information are excluded from the keyword candidates; the predetermined model is a model that outputs the relevance between input keywords, Keyword output system.

[0062] (Appendix 9) 10. The keyword output system according to claim 1, The control unit acquiring a plurality of keyword candidates to be targeted for the client's search-related advertising based on the character information; Obtain a list of predefined keywords to be excluded, selecting a keyword from the plurality of keyword candidates based on the list; Keyword output system.

[0063] (Appendix 10) 10. The keyword output system according to claim 1, The control unit generating a sentence relating to the search advertising; Keyword output system.

[0064] (Appendix 11) A keyword output method executed by a keyword output system, Get phone call data, The call data is call data of a client's telephone; generating text information by transcribing the call data; acquiring keywords to be targeted by the client's search-related advertising based on the character information; outputting the keyword; Keyword output method.

[0065] (Appendix 12) A program, Computer, A program for functioning as the keyword output system described in any one of Appendix 1 to Appendix 10.

[0066] While various embodiments of the present invention have been described, these are presented as examples and are not intended to limit the scope of the invention. Novel embodiments may be embodied in various other forms, and various omissions, substitutions, and modifications may be made without departing from the spirit of the invention. The embodiments and their modifications are intended to fall within the scope and spirit of the invention, as well as within the scope of the inventions and their equivalents as set forth in the accompanying claims.

[0067] The above-described embodiments and modifications may be implemented in any combination. [Explanation of symbols]

[0068] 100: Server device 210: Control unit 220: Storage section 230: Communications Department 240: Internal bus

Claims

1. A keyword output system, having at least one or more control units, The control unit Acquire call data from the automated answering service. the call data is call data of a call between a client who uses the function of the telephone automatic answering service and a customer of the client; generating text information by transcribing the call data; Obtaining business information indicating the business content of the client; generating a prompt including the textual information; the prompt further includes an instruction to extract, from the text information, a plurality of keyword candidates to be targeted for search advertising for the client, taking into account the business information; obtaining data output from a large-scale language model based on the prompt as a plurality of keyword candidates; Obtaining a search volume for each of the plurality of keyword candidates; removing keyword candidates having a search volume equal to or greater than a first threshold and keyword candidates having a search volume less than a second threshold from the plurality of keyword candidates; outputting the deleted keyword candidates as keywords to be targeted for search-related advertising by the client; Keyword output system.

2. A keyword output system, having at least one or more control units, The control unit Acquire call data from the automated answering service. the call data is call data of a call between a client who uses the function of the telephone automatic answering service and a customer of the client; generating text information by transcribing the call data; Acquire the data of the client's web page; generating a prompt including the textual information; The prompt further includes an instruction to extract, from the text information, a plurality of keyword candidates to be targeted for search advertising by the client, taking into account the data of the web page; obtaining data output from a large-scale language model based on the prompt as a plurality of keyword candidates; Obtaining a search volume for each of the plurality of keyword candidates; removing keyword candidates having a search volume equal to or greater than a first threshold and keyword candidates having a search volume less than a second threshold from the plurality of keyword candidates; outputting the deleted keyword candidates as keywords to be targeted for search-related advertising by the client; Keyword output system.

3. A keyword output method executed by a keyword output system, Acquire call data from the automated answering service. the call data is call data of a call between a client who uses the function of the telephone automatic answering service and a customer of the client; generating text information by transcribing the call data; Obtaining business information indicating the business content of the client; generating a prompt including the textual information; the prompt further includes an instruction to extract, from the text information, a plurality of keyword candidates to be targeted for search advertising for the client, taking into account the business information; obtaining data output from a large-scale language model based on the prompt as a plurality of keyword candidates; Obtaining a search volume for each of the plurality of keyword candidates; removing keyword candidates having a search volume equal to or greater than a first threshold and keyword candidates having a search volume less than a second threshold from the plurality of keyword candidates; outputting the deleted keyword candidates as keywords to be targeted for search-related advertising by the client; Keyword output method.

4. A keyword output method executed by a keyword output system, Acquire call data from the automated answering service. the call data is call data of a call between a client who uses the function of the telephone automatic answering service and a customer of the client; generating text information by transcribing the call data; Acquire the data of the client's web page; generating a prompt including the textual information; The prompt further includes an instruction to extract, from the text information, a plurality of keyword candidates to be targeted for search advertising by the client, taking into account the data of the web page; obtaining data output from a large-scale language model based on the prompt as a plurality of keyword candidates; Obtaining a search volume for each of the plurality of keyword candidates; removing keyword candidates having a search volume equal to or greater than a first threshold and keyword candidates having a search volume less than a second threshold from the plurality of keyword candidates; outputting the deleted keyword candidates as keywords to be targeted for search-related advertising by the client; Keyword output method.

5. A program, Computer, A program for causing the keyword output system according to claim 1 or 2 to function.

Images