Information processing system, information processing method, and program

Abstract

Provided is an information processing system, an information processing method, and a program that further reduce the burden of creating different documents every day for sales or advertising purposes. [Solution] In an information processing system in which an information processing device and a user terminal can communicate over a network, a control unit of the information processing device executes step S001 in which a reception unit receives advertising target information from a user, and step S002 in which a generation unit generates one or more sentences related to advertising using a learning model based on the advertising target information. The advertising target information includes at least information on the advertising target and date information related to the date. The learning model is a model trained so that advertising target information is input and sentences can be output, and the sentences include related information related to the date information.

Description

[Technical field]

[0001] The present invention relates to an information processing system, an information processing method, and a program. [Background technology]

[0002] For example, Patent Document 1 discloses a "system for eliminating user hassle and promoting sales and customer attraction through SNS posts." The system includes a processor capable of executing a program that executes the steps of acquiring an image related to a store, generating a comment according to the content of the image by analyzing the acquired image, generating a posted article in which the generated comment is added to the image, and automatically posting the generated posted article on a predetermined SNS account for the store. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2023-145312 A Summary of the Invention [Problem to be solved by the invention]

[0004] However, there is still room for improvement in the functions of the above-mentioned known techniques. In particular, there is a need for a technique that further reduces the burden involved in creating different posts or emails every day for sales or advertising purposes.

[0005] In view of the above circumstances, the present invention provides an information processing system and the like that can further reduce the burden involved in creating different documents every day for sales or advertising purposes. [Means for solving the problem]

[0006] According to one aspect of the present invention, there is provided an information processing system including a processor capable of executing a program to perform the following steps: in a reception step, input of advertising target information is received from a user, and the advertising target information includes at least information regarding the subject of the advertising and date information regarding the date; in a first generation step, one or more sentences regarding advertising are generated by a learning model based on the advertising target information, the learning model being a model that has been trained so as to be able to take the advertising target information as input and output sentences, and the sentences include relevant information related to the date information.

[0007] According to this embodiment, the burden of creating different documents every day for sales or advertising purposes can be further reduced. [Brief description of the drawings]

[0008] [Figure 1] 1 is a configuration diagram illustrating an information processing system 1. [Diagram 2] 2 is a block diagram showing a hardware configuration of an information processing device 2. FIG. [Diagram 3] FIG. 2 is a block diagram showing a hardware configuration of a user terminal 3. [Figure 4] 2 is a functional block diagram showing functions of an information processing device 2. FIG. [Diagram 5] 2 is a flowchart showing an overview of processing executed by the information processing system 1. [Figure 6] 2 is an activity diagram showing a specific example of processing executed by the information processing system 1. FIG. [Figure 7] FIG. 2 is a schematic diagram of a related information table T1. [Figure 8] 1 shows an example of an input screen 5 displayed on the display unit 34 of the user terminal 3 of the user X. [Figure 9] 1 shows an example of a generated result screen 6 displayed on the display unit 34 of the user terminal 3 of the user X. [Figure 10] 2 shows an example of a search result screen 7 displayed on the display unit 34 of the user terminal 3. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described with reference to the drawings. Various characteristic features shown in the following embodiments can be combined with each other.

[0010] Incidentally, a program for realizing the software appearing in one embodiment may be provided as a non-transitory computer-readable recording medium, or may be provided so as to be downloadable from an external server, or may be provided so that the program is launched on an external computer and its functions are realized on a client terminal (so-called cloud computing).

[0011] In addition, in various information processing according to an embodiment, an input and an output according to the input can be realized. Here, as long as an output is obtained as a result of the input, the form of information referenced in such information processing (hereinafter referred to as reference information) is not limited. The reference information may be, for example, rule-based information such as a database, a lookup table, or a predetermined function (including a judgment formula such as a regression formula constructed by a statistical method), or may be a trained model that has previously trained the correlation between the input and the output, or may be a generation AI such as a large-scale language model or a visual language model that can output a desired result by inputting a prompt.

[0012] In one embodiment, the term "unit" may include, for example, a combination of hardware resources implemented by a circuit in the broad sense and software information processing that can be specifically realized by these hardware resources. In one embodiment, various information is handled, and this information is represented by, for example, physical values ​​of signal values ​​representing voltage and current, high and low signal values ​​as a binary bit collection consisting of 0 or 1, or quantum superposition (so-called quantum bits), and communication and calculation can be performed on the circuit in the broad sense.

[0013] Furthermore, a circuit in the broad sense is a circuit realized by at least appropriately combining a circuit, circuitry, a processor, and a memory. The processor may be a general-purpose processor or a dedicated circuit. In other words, it includes an application specific integrated circuit (ASIC), a programmable logic device (e.g., a simple programmable logic device (SPLD), a complex programmable logic device (CPLD), and a field programmable gate array (FPGA)), etc.

[0014] 1. Hardware Configuration In this section, the hardware configuration of one embodiment will be described.

[0015] 1.1 Information Processing System 1 FIG. 1 is a configuration diagram showing an information processing system 1. The information processing system 1 includes an information processing device 2 and a user terminal 3. The information processing device 2 and the user terminal 3 are configured to be able to communicate with each other via a telecommunications line. In one embodiment, the information processing system 1 is made up of one or more devices or components. For example, if the information processing system 1 is made up of only the information processing device 2, the information processing system 1 can be the information processing device 2. These components will be described below.

[0016] 1.2 Information processing device 2 2 is a block diagram showing a hardware configuration of the information processing device 2. The information processing device 2 includes a communication bus 20, a communication unit 21, a storage unit 22, and a control unit (processor) 23. The communication unit 21, the storage unit 22, and the processor 23 are electrically connected via the communication bus 20 inside the information processing device 2.

[0017] <Communication unit 21> Although the wired communication means such as USB, IEEE1394, Thunderbolt (registered trademark), wired LAN network communication, etc. are preferable for the communication unit 21, wireless LAN network communication, mobile communication such as 3G / LTE / 5G, BLUETOOTH (registered trademark) communication, etc. may be included as necessary. That is, it is more preferable to implement as a set of these plurality of communication means. That is, the information processing apparatus 2 may communicate various information from the outside via the communication unit 21 and the network.

[0018] <Storage unit 22> The storage unit 22 stores various information defined by the foregoing description. This may be implemented as a storage device such as a solid state drive (SSD) that stores various programs and the like related to the information processing apparatus 2 executed by the processor 23, or as a memory such as a random access memory (RAM) that stores temporarily necessary information (arguments, arrays, etc.) related to the calculation of the program. The storage unit 22 stores various programs, variables, etc. related to the information processing apparatus 2 executed by the processor 23.

[0019] <Processor 23> The processor 23 performs processing and control of the overall operation related to the information processing apparatus 2. The processor 23 is, for example, a central processing unit (CPU) not shown. The processor 23 realizes various functions related to the information processing apparatus 2 by reading a predetermined program stored in the storage unit 22. That is, the information processing by software stored in the storage unit 22 is specifically realized by the processor 23 which is an example of hardware, and can be executed as each functional unit included in the processor 23. These will be described in more detail in the next section. Note that the processor 23 is not limited to being single, and may be implemented to have a plurality of processors 23 for each function. Or a combination thereof may be used.

[0020] 1.3 User terminal 3 The user terminal 3 is a terminal owned by a user. The user terminal 3 is also assumed to be a terminal 3 operated by a user. The user terminal may be in any form, such as a smartphone, a tablet terminal, a computer, or any other device that can access the information processing device 2 via a telecommunication line.

[0021] 3 is a block diagram showing a hardware configuration of the user terminal 3. The following describes the user terminal 3 as an example. The user terminal 3 includes a communication bus 30, a communication unit 31, a storage unit 32, a processor 33, a display unit 34, and an input unit 35. The communication unit 31, the storage unit 32, the processor 33, the display unit 34, and the input unit 35 are electrically connected via the communication bus 30 inside the user terminal 3. Descriptions of the communication unit 31, the storage unit 32, and the processor 33 will be omitted because they are similar to the descriptions of the respective units in the information processing device 2.

[0022] <Display section 34> The display unit 34 displays a screen of a graphical user interface (GUI) that can be operated by the user. The display unit 34 may be included in the housing of the user terminal 3, or may be attached externally. Specifically, the display unit 34 may be implemented as a display device such as a CRT display, a liquid crystal display, an organic EL display, or a plasma display. It is preferable that these display devices are implemented by selectively using them according to the type of the user terminal 3.

[0023] <Input section 35> The input unit 35 accepts an operation input made by a user. The operation input is transferred as a command signal to the processor 33 via the communication bus 30. The processor 33 may execute a predetermined control or calculation based on the transferred command signal as necessary. The input unit 35 may be included in the housing of the user terminal 3, or may be externally attached. For example, the input unit 35 may be implemented as a touch panel integrated with the display unit 34. When the input unit 35 is implemented as a touch panel, the user can input a tap operation, a swipe operation, or the like to the input unit 35. As the input unit 35, a switch button, a mouse, a QWERTY keyboard, or the like can be adopted instead of a touch panel.

[0024] 2. Functional configuration In this section, the functional configuration will be described. Fig. 4 is a functional block diagram showing the functions of the information processing device 2. As described above, information processing by software stored in the storage unit 22 can be specifically realized by hardware (specifically, the processor 23) and executed as each functional unit included in the processor 23. In other words, the information processing system 1 includes each functional unit in a program.

[0025] Specifically, the processor 23 includes, as its functional units, a reception unit 231, an acquisition unit 232, an identification unit 233, a generation unit 234, an artificial intelligence unit 235, and a display control unit 236.

[0026] The reception unit 231 is configured to receive various information as a reception step. Specifically, the reception unit 231 is configured to receive information via the communication unit 21 or the storage unit 22 and to be able to read the information into a working memory. Preferably, the reception unit 231 receives advertisement target information IF1 relating to an advertisement target via the input unit 35 of the user terminal 3 or the like. The advertisement target is, for example, a business in which the user is involved.

[0027] The acquiring unit 232 is configured to acquire various information as an acquiring step. Specifically, the acquiring unit 232 acquires the advertisement target information IF1 accepted by the accepting unit 231 from the user.

[0028] The identification unit 233 is configured to identify various pieces of information as an identification step. Specifically, the identification unit 233 identifies related information IF2 related to date information IF12 based on date information IF12 included in the advertisement target information IF1 and a related information table T1 which is predetermined reference information. The related information table T1 is configured to be able to refer to the date included in the date information IF12 and the related information IF2. In addition, the related information IF2 is information including information derived from the date.

[0029] The generating unit 234 is configured to generate various information as a generating step. Specifically, the generating unit 234 may be configured to generate a prompt that instructs outputting a text related to an advertisement, using the related information IF2, etc., identified by the identifying unit 233 as an input. Note that "generate" may be read as "create".

[0030] Furthermore, the generation unit 234 is configured to be able to instruct the artificial intelligence unit 235 to generate text related to the advertisement of the advertised target based on the advertising target information IF1 and related information IF2. The artificial intelligence unit 235 is an example of artificial intelligence, and has a learning model LM that receives the advertising target information IF1 and related information IF2 as input and outputs text related to the advertisement of the advertised target. The learning model LM is a learning model that creates / generates and outputs text related to the advertisement of the advertised target based on the advertising target information IF1 and related information IF2.

[0031] This learning model LM may be a model generated by performing so-called supervised machine learning, in which advertising target information IF1, which includes various information about the business in which the user is involved, and related information IF2 about dates are used as input data for learning (explanatory variables), texts actually posted by the user on SNS or the like are used as output data for learning (objective variables), and the input data and output data are used as teacher data in which the input data and output data are associated in pairs.

[0032] The learning model may be generated by learning a corpus (a database in which natural language sentences are structured and accumulated on a large scale), and may include the advertising target information IF1 and related information IF2 related to dates, etc., as one of the large corpora. The artificial intelligence unit 235 generates advertising sentences derived from the input advertising target information IF1, etc., using this learning model, and outputs the generated sentences. The display control unit 236 also functions as an example of a display unit that displays the advertising sentences output by the artificial intelligence unit 235 based on instructions from the generation unit 234.

[0033] The artificial intelligence unit 235 is configured to receive input from each functional unit and return an instructed output. The artificial intelligence used by each functional unit of the information processing device 2 may be a common one, or may be prepared individually for each functional unit.

[0034] The artificial intelligence unit 235 is an AI (Artificial Intelligence) equipped with language models such as Transformers including GPT (Generative Pretrained Transformer, including GPT-1, GPT-2, GPT-3, and GPT-4), BERT (Bidirectional Encoder Representations from Transformers), and BART (Bidirectional and Auto-regressive Transformer), and Recurrent Neural Networks (RNN), and includes a generative AI.

[0035] The language model is an example of a learning model based on a machine learning algorithm. Specific examples of machine learning algorithms include nearest neighbor methods, naive Bayes methods, decision trees, support vector machines, and deep learning using neural networks. The artificial intelligence unit 235 can apply the above algorithms as appropriate.

[0036] The artificial intelligence unit 235 has a trained model constructed by a learning method such as supervised learning, unsupervised learning, or self-supervised learning. In supervised learning, machine learning is performed using training data (training data). The training data is composed of a pair of input data for learning and output data (correct answer data). In addition, the language model may not only be one trained for a specific task, but also a general-purpose model that can be used for a wide range of tasks.

[0037] The artificial intelligence unit 235 includes, as the artificial intelligence, a general-purpose natural language processing learning model such as a large-scale language model (LLM) that has learned a huge amount of data. Such a general-purpose learning model includes a language model that can handle various tasks without fine tuning by one-shot learning, few-shot learning, etc. Furthermore, the general-purpose learning model can also handle various tasks by zero-shot learning. The artificial intelligence used in each functional unit of the control unit 23 may be a separate learning model, or may be a common general-purpose learning model.

[0038] The artificial intelligence unit 235 may receive the advertisement target information IF1 and the related information IF2 as input, and based on the prompt generated by the generation unit 234, output a sentence related to advertising the business or the like in which the user is involved.

[0039] As a display control step, the display control unit 236 displays various information stored in the storage unit 22 or a screen including the information in a manner viewable on a terminal such as the user terminal 3. This allows various information to be presented to a user who operates the user terminal 3. When the phrase "display" is used, it does not matter whether the display medium to be displayed is in a local environment or whether processing for displaying is performed via the network 11. The display control unit 236 controls visual information such as a screen, an image, an icon, a message, etc. to be displayed on the display of the terminal. The display control unit 236 may generate only rendering information for displaying the visual information on the terminal.

[0040] 3. Information Processing The information processing of the first embodiment will be described below.

[0041] 3.1 Processing Overview The information processing system 1 includes at least one processor 23, and the processor 23 includes a program for executing each of the following units. In other words, the information processing method includes each of the steps of the information processing system 1 described below. From another perspective, the program causes a computer to execute each of the steps of the information processing system 1. FIG. 5 is a flowchart showing an outline of the processing executed by the information processing system 1. Each step shown in FIG. 5 will be described below.

[0042] First, the receiving unit 231 receives an input of advertising target information IF1 from a user as a receiving step (step S001). Here, the advertising target information IF1 includes at least information related to the advertising target and date information IF12 related to the date. Next, the generating unit 234 generates one or more sentences related to advertising using a learning model LM based on the advertising target information IF1 as a first generating step (step S002). The learning model LM is a model that has been trained so that it can take the advertising target information IF1 as input and output a sentence. In addition, the sentence includes related information IF2 related to the date information IF12.

[0043] According to this embodiment, when posting to SNS or sending business e-mails every day, text containing information related to the date of the day can be easily created. In other words, it becomes easy to post or send e-mails with different contents every day. Therefore, the burden on the user who creates text can be reduced.

[0044] 3.2 Specific examples of information processing 6 is an activity diagram showing a specific example of processing executed by the information processing system 1. The specific example may be included within the scope defined in the above-mentioned overview. In the specific example, an example will be described in which the information processing system 1 generates text for a user X to post on an SNS or the like regarding his or her own business. Here, the user X is a manager of a restaurant, and the example will be described in which the user X generates text for posting an advertisement for the restaurant on an SNS.

[0045] First, the reception unit 231 receives an input request for advertising target information IF1 from the user X as a reception step (activity A101). The advertising target information IF1 includes information on the target of advertising. In this embodiment, the advertising target information IF1 will be described as including information on the restaurant managed by the user X.

[0046] Here, the advertised target information IF1 includes at least date information IF12 related to a date. The date included in the date information IF12 may be, for example, the date on which the text to be generated is posted. Also, for example, the date included in the date information IF12 may be the date on which an event is scheduled to be held at the store to be advertised or in the area where the store is located. The date included in the date information IF12 is not limited to this.

[0047] Next, the display control unit 236, as a display control step, displays an input screen for inputting the advertised information IF1 on the display unit 34 of the user terminal 3 of the user X (activity A102). A specific example of the input screen will be described with reference to FIG.

[0048] Next, as a receiving step, the receiving unit 231 receives input of the advertisement target information IF1 from the user X via the input screen (activity A103). Furthermore, as a receiving step, the receiving unit 231 receives a request to generate text from the user X (activity A104).

[0049] Next, the identifying unit 233 identifies related information IF2 based on the date information IF12 and predetermined reference information (activity A105). More specifically, the identifying unit 233 identifies related information IF2 related to the date information IF12 by matching the date information IF12 included in the advertised information IF1 with the related information table T1 (an example of reference information). In this embodiment, the reference information will be described as being the related information table T1. The related information table T1 is configured to be able to refer to at least the date information IF12 and the related information IF2. Next, the acquiring unit 232 acquires the related information IF2 identified in the activity A105.

[0050] The method of acquiring the related information IF2 from the related information table T1 may be, for example, RAG (Retrieval Augmented Generation). More specifically, in activity A105 of FIG. 6, the control unit 23 may read out date information IF12 from the storage unit 22. Then, the control unit 23 may access the related information table T1 via the communication unit 21. The control unit 23 may acquire the related information IF2 that matches the read out date information IF12 from the related information table T1. The control unit 23 may store the acquired related information IF2 in the storage unit 22.

[0051] Next, the generation unit 234 receives the advertising target information IF1 acquired by the acquisition unit 232 and the related information IF2 identified by the identification unit 233 as input, and generates a prompt PR that instructs outputting a text related to the advertisement (activity A106).

[0052] Next, as a first generation step, the generation unit 234 inputs the advertising target information IF1 and an instruction to generate a sentence based on the advertising target information IF1 to the learning model LM, which is a large-scale language model, to generate a sentence (activity A107). For example, the generation unit 234 instructs the artificial intelligence unit 235 to generate a sentence by the prompt PR generated in activity A106. The artificial intelligence unit 235 generates a sentence based on the prompt PR.

[0053] In this embodiment, the learning model LM is described as being the large-scale language model LM1, but this is not limited to the present invention.

[0054] Furthermore, the generating unit 234 may generate keywords related to the text using the large-scale language model LM1 (learning model LM) as a second generating step. Then, the artificial intelligence unit 235 may output the hash tag HT.

[0055] Next, the display control unit 236 displays the text content TC in which the hashtag HT (an example of a keyword) and a sentence are associated with each other so that the user X can view it (activity A108). Specifically, the display control unit 236 displays the text content TC on the display unit 34 of the user terminal 3. A specific example of the display screen will be described in FIG. 9. Note that the text content TC may include a configuration that allows the user X to post it to a predetermined site.

[0056] The hashtag HT includes a structure in which a hash symbol "#" is added before text including a keyword. The keyword may include a keyword related to the advertised information IF1 input from the user X in the activity A103 or a sentence generated in the activity A106.

[0057] According to this embodiment, a hashtag corresponding to the posted text or information on a store or the like is generated together with the posted text, which further reduces the burden on User X in posting to the SNS. It is also expected that the posting will be more likely to be viewed by viewers who are interested in the keyword indicated by the hashtag.

[0058] It should be noted that the text content TC does not necessarily have to include the hashtag HT. The configuration of the text content TC is not limited to this.

[0059] Furthermore, the method of generating the text and hashtag HT in activity A107 may be, for example, RAG. More specifically, the control unit 23 may read out the advertised information IF1, the related information IF2, and the prompt PR generated in activity A106. Next, the control unit 23 may read out the large-scale language model LM1. Then, the control unit 23 may input the advertised information IF1, the related information IF2, and the prompt PR generated in activity A106 to the large-scale language model LM1 to generate the text and hashtag HT. The control unit 23 may store the generated text and hashtag HT in the storage unit 22.

[0060] 4. Related technical matters

[0061] (Related information table T1) Fig. 7 is a schematic diagram of related information table T1. Related information table T1 is an example of a database referenced in activity A105. As shown in Fig. 7, related information table T1 may include information such as date (T11) and related information IF2 (T12 to T16). Date (T11) indicates a date that can be identified from date information IF12.

[0062] The related information IF2 managed in association with the date (T11) includes information indicating an object or event related to a date identifiable from the date information IF12. As shown in FIG. 7, for example, the related information IF2 corresponding to December 13 may be "Twins Day," "Vitamin Day," "Wright Brothers Day," and "Beauty Salon Day." Specifically, "Twins Day" corresponding to December 13 originates from the fact that December 13, 1874, was the day when the first law regarding twins was enacted in Japan. In addition, for example, "Beauty Salon Day" corresponding to December 13 originates from the fact that December is the busy season for beauty salons, and that the combination of the numbers 1 and 3 forms a shape that symbolizes the letter B, which is the first letter of "Beauty."

[0063] The related information table T1 may be managed in the storage unit 22 of the information processing device 2, may be managed in an external storage medium, or may be managed in a distributed manner through block chain technology. In addition, some of the various information included in the related information table T1 may be changeable or updatable.

[0064] According to this embodiment, since a sentence based on information derived from a date such as "What day is it today" is generated, it is possible to generate sentences with different contents every day. Therefore, User X can post different contents every day, which increases the possibility of the posts being ranked higher on the search engine search result page and viewed by more viewers.

[0065] (Input screen 5) 8 shows an example of an input screen 5 displayed on the display unit 34 of the user terminal 3 of user X. The input screen 5 is an example of a screen displayed in activity A103 of FIG.

[0066] Areas 51 to 56 are areas where advertisement target information IF1 is input by user X. Area 51 is an area that accepts input of, for example, the name of a store. Area 52 is an area that accepts input of date information IF12. Area 53 is an area that accepts input of information related to the genre of the business. More specifically, area 53 may be an area that accepts input of the contents of the service provided or the service category. Area 54 is an area that accepts input of information related to the target customer group of the business, that is, the target viewers of the post. Area 55 is an area that accepts input of information related to the appealing points of the business. Area 56 is an area that accepts input of the name of the product or service provided.

[0067] The advertising target information IF1 input into the areas 51-56 may be used to generate a prompt PR in the activity A106 of Fig. 6. That is, the prompt PR, which is an instruction, may include content for including in a sentence the name of a store or the name of a product or service that can be specified based on the advertising target information IF1.

[0068] For example, the generation unit 234 may generate a prompt PR that reads, "Generate text advertising the store "Takomaru Akasaka," which offers the product "Melted Cheese Takoyaki," to be posted on "Twins Day" on December 13." In this case, "December 13," "Takomaru Akasaka," and "Melted Cheese Takoyaki" are information that can be specified based on the advertising target information IF1, and are words that are used in the text to be generated. Also, "Twins Day" is information that can be specified based on the date information IF12 included in the advertising target information IF1 and the related information table T1, and is a word that is used in the text to be generated.

[0069] According to this embodiment, the generated text can include the name of the store or the name of the menu of the user X, and the text can be more original. Therefore, the user X can post a text that can be expected to have a greater advertising effect.

[0070] Also, for example, the prompt PR, which is an instruction, may include content for including in the text a style that matches the attributes of the target reader of the text, which can be specified based on the advertising target information IF1. In this embodiment, the "attributes of the target reader" may be "college student", which is information about the target who has received input in the area 54. According to this aspect, it is possible to generate text in a style that matches the type of customer or potential customer who is the target of the advertising. Therefore, the user X can make a post that will leave a lasting impression on the customers who are readers.

[0071] For example, the generation unit 234 may generate a prompt PR such as "Generate text advertising the store 'Takomaru Akasaka' for 'college students' to be posted on 'Twins Day' on December 13th." In this case, "college students" is information that can be specified based on the advertising target information IF1 as an attribute of the target audience of the text. Therefore, the large-scale language model LM1 creates posted text that includes a style of writing that is natural to the reader, such as that commonly used by college students. Specifically, the posted text may have a structure that includes a symbol such as "☆" or "♪" at the end of the text.

[0072] Preferably, the prompt PR may be generated by managing predetermined templates and applying the advertisement target information IF1 received via input to the areas 51 to 56 to each of the templates. Also preferably, the prompt PR may be generated by a large-scale language model LM1 based on the advertisement target information IF1. The method of generating the prompt PR is not limited to this.

[0073] The object 57 is an object for receiving from the user X a request for generating a sentence based on the information input in the areas 51 to 56. For example, by receiving an input to the object 57 from the user X, the generation unit 234 starts generating a sentence. That is, the processor 23 executes the processes of the activities A105 to A107 in FIG. 6.

[0074] (Generated Result Screen 6) FIG. 9 shows an example of the generated result screen 6 displayed on the display unit 34 of the user terminal 3 of user X. The generated result screen 6 is an example of the screen displayed in the activity A108 of FIG. 6. The generated result screen 6 includes regions 61 to 69. Note that since regions 61 to 66 have the same configuration as regions 51 to 56 in FIG. 8, the description thereof is omitted.

[0075] Regions 67 to 69 are regions each including text contents TC1 to TC3 composed of a text generated by a large language model and a hash tag HT corresponding to the text. Specifically, the texts output based on the generated prompt PR with the publicity target information IF1 input to regions 61 to 66 and the related information IF2 specified using the date information IF12 as a key are displayed in a manner included in regions 67 to 69.

[0076] Here, as shown in FIG. 7, the related information table T1, which is reference information, may be configured to be able to refer to a plurality of related information IF2 with respect to the date information IF12. That is, as a first step, the generation unit 234 can create a plurality of texts. Note that the plurality of texts each include different related information IF2.

[0077] As shown in FIG. 9, a plurality of types of text contents TC1 to TC3 related to the publicity of the store are generated by the large language model LM1, and the text contents TC1 to TC3 are displayed in a manner included in regions 67 to 69 respectively. The related information IF2 included in the generated text contents TC1 to TC3 are different from each other.

[0078] According to such a manner, user X can select a posting text that better matches the image of his or her own business or store from among the plurality of texts. In addition, user X can select a posting text so that the posting content is not similar to that of other users who are competitors.

[0079] Here, at least a part of the related information IF2 among the multiple related information IF2 that can be referenced by the date information IF12 may be specified and used to generate the text. For example, information that is highly related to the business content of the store may be specified from the related information IF2 and used to generate the text. Specifically, the related information IF2 included in the related information table T1 may be tagged with a tag corresponding to the information included in the advertising target information IF1.

[0080] In this embodiment, among the multiple pieces of related information IF2 that can be referred to based on the input date information IF12, "Twins Day", "Vitamin Day", and "Beauty Salon Day" may be identified as related information IF2 that is particularly highly related to the business content (genre) of the store. The specific information of related information IF2 managed in the related information table T1 is not limited to this.

[0081] Furthermore, a sentence generated by the large-scale language model LM1 will be described by taking the text content TC1 displayed in the area 67 as an example. The text content TC1 includes related information IF2 (Twins Day) identified based on the date information IF12. The text content TC1 also includes the store name (Takomaru Akasaka) and menu names (Melted Cheese Takoyaki, Japanese-style Takoyaki) input by the user X as the advertising target information IF1. The hashtag HT1 also includes keywords (Twins Day, Takoyaki) related to the advertising target information IF1 input by the user X to the area 63 or the generated sentence.

[0082] In this way, the text generated by the artificial intelligence unit 235 includes the advertising target information IF1 related to the store that is the target of the advertising input by the user X. This allows the user X to post a highly effective advertising message.

[0083] (Search results screen 7) 10 shows an example of the search result screen 7 displayed on the display unit 34 of the user terminal 3. The search result screen 7 is a screen that shows the results when, for example, a user other than user X (hereinafter, referred to as a viewer) uses a search engine via the user terminal 3. The search result screen 7 includes areas 70 to 72, an area 71A, and an object 72A.

[0084] The search result screen 7 is a screen showing the results of information search based on a predetermined keyword (Akasaka takoyaki) input by the viewer. The area 70 is an area for accepting input of a keyword for search from the viewer. The area 71 is a predetermined area including profile information of the store hit by the search using the keyword input to the area 70 as a key. The profile information may include, for example, the name of the store, the location of the store, or the average budget. The area 72 is an area where map information for showing the location of the store included in the area 71 is displayed. As shown in FIG. 10, the location of the store included in the area 71A is shown by an object 72A in the area 72. A rating given to the store may be drawn in the object 72A.

[0085] The store information whose profile information is displayed at the top of the search result screen 7 is determined based on, for example, the frequency with which information is posted to a specific site or the content of the posts in correspondence with the profile information. In other words, the higher the frequency with which text is posted and the more diverse the content of the posts, the higher its profile information is displayed on the screen. In other words, the higher the profile information of such a store is ranked on the search engine search result screen. For example, "Takoyaki Cafe Takomaru Akasaka 1-chome Store" included in area 71A is a store that posts to SNS more frequently than other stores or posts more diverse content.

[0086] According to this embodiment, since texts with different contents can be generated according to the date, User X can post more frequently and can diversify the contents of his / her daily posts. Therefore, the store's profile information is displayed at the top of the search results screen via the search engine, and the profile information or posts are more likely to be viewed by more viewers.

[0087] 5.Other The information processing system 1 according to the above embodiment may be configured as follows.

[0088] In the above embodiment, an example was described in which the related information IF2 is information derived from a date, such as "What day is it today?", but this is not limited to this. For example, the related information IF2 may include information about an event in a region including the location of the store. Specifically, in activity A103 of FIG. 6, the reception unit 231 receives advertisement target information IF1 that further includes location information that can identify the location of the store. Next, in activity A105, the identification unit 233 may match the location information and date information IF12 with the related information table T1, and identify information about an event scheduled to be held in the region where the store is located on that date as the related information IF2. According to this embodiment, it is possible to make a post related to the region where the store is located. Therefore, a higher advertising effect can be expected.

[0089] For example, the related information IF2 may further include information related to the Rokuyo. Furthermore, the related information IF2 may include information related to the phases of the moon, such as a new moon or a full moon, or may include information related to a solar eclipse. The related information IF2 may be a combination of these. According to such an embodiment, a variety of posted texts can be created, and an even higher advertising effect can be expected.

[0090] In this embodiment, an example of generating a sentence and a hashtag HT by one prompt PR and a large language model LM1 has been described, but this is not the only case. For example, first, a sentence may be generated by a first large language model, and then a hashtag HT corresponding to the generated sentence may be further generated by a second large language model.

[0091] In one embodiment, the reception unit 231, the acquisition unit 232, the specification unit 233, the generation unit 234, the artificial intelligence unit 235, and the display control unit 236 have been described as functional units realized by the processor 23 of the information processing apparatus 2. However, at least a part of these may be implemented as a functional unit realized by an external server (not shown) or as a functional unit realized by the processor 33 of the user terminal 3.

[0092] The information processing apparatus 2 may be in an on-premises form or a cloud form. As the information processing apparatus 2 in the cloud form, for example, the above functions and processes may be provided in the form of SaaS (Software as a Service), cloud computing, or the like.

[0093] In the above embodiment, the information processing apparatus 2 performs various storage and control. However, instead of the information processing apparatus 2, a plurality of external devices may be used. That is, various information and programs may be distributed and stored in a plurality of external devices using blockchain technology or the like.

[0094] Furthermore, it may be provided in each aspect described below.

[0095] (1) An information processing system comprising a processor capable of executing a program to perform each of the following steps: in a receiving step, input of advertising target information is received from a user, the advertising target information including at least information regarding a subject of the advertising and date information relating to a date; in a first generating step, one or more sentences regarding advertising are generated by a learning model based on the advertising target information, wherein the learning model is a model trained to be capable of receiving the advertising target information as input and outputting the sentences, and the sentences include relevant information related to the date information.

[0096] According to this embodiment, when posting to SNS or sending business e-mails every day, text containing information related to the day can be easily created, and it is possible to post or send e-mails with different contents every day. Therefore, the burden on the user who writes the text can be reduced.

[0097] (2) In the information processing system described in (1) above, further, in the identification step, the related information is identified based on the date information and predetermined reference information, and the reference information is configured to be able to refer to at least the date information and the related information.

[0098] (3) In the information processing system described in (2) above, the reference information is configured to be able to refer to multiple pieces of related information for the date information, and in the first generation step, multiple pieces of sentences can be created, and each of the multiple sentences includes different pieces of related information.

[0099] According to this embodiment, it is possible to select a post from among a plurality of posts, and it is also possible to avoid posting content similar to that of other users.

[0100] (4) In the information processing system described in (1) above, the related information includes information indicating an object or event related to a date that can be identified from the date information.

[0101] According to this embodiment, since a sentence based on information derived from a date, such as "What day is it today?", is generated, it is possible to generate sentences with different contents every day. Therefore, a user can post different contents every day, which increases the possibility of being ranked high on the search result page of a search engine and being viewed by more viewers.

[0102] (5) In the information processing system described in (1) above, in the first generation step, the advertising target information and an instruction to generate the sentence based on the advertising target information are input into the learning model, which is a large-scale language model, to generate the sentence.

[0103] (6) In the information processing system described in (5) above, the instructions include content for including in the text the name of a store or the name of a product or service that can be identified based on the advertising target information.

[0104] According to this embodiment, the generated text can include the user's store name or menu name, etc., making it possible to generate more original text.

[0105] (7) In the information processing system described in (5) above, the instructions include content for including in the text a writing style that matches the attributes of the target viewers of the text, which can be identified based on the advertising target information.

[0106] According to this embodiment, it is possible to generate text that matches the target customers of the advertisement, and therefore the user can post text that leaves a lasting impression on the customers.

[0107] (8) In the information processing system described in (1) above, further, in a second generation step, keywords related to the sentence are generated using the learning model, and in a display control step, text content in which the keywords and the sentence are associated is displayed so as to be visible to the user, and the text content includes a configuration that allows the user to post it to a specified site.

[0108] According to this embodiment, by generating a hashtag corresponding to the generated text, the burden on the user in posting to the SNS can be reduced. Also, the possibility that the post will be viewed by viewers who are interested in the keyword indicated by the hashtag increases.

[0109] (9) An information processing method comprising the steps of the information processing system according to any one of (1) to (8) above.

[0110] (10) A program causing at least one computer to execute each step of the information processing system according to any one of (1) to (8) above. Of course, this is not the case.

[0111] Finally, although 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. The novel embodiments can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. The embodiments and their modifications are included within the scope and spirit of the invention, and are included in the scope of the invention and its equivalents described in the claims. [Explanation of symbols]

[0112] 1: Information processing system 11: Network 2: Information processing equipment 20: Communication bus 21: Communications Department 22: Storage section 23: Processor (control unit) 231: Reception 232: Acquisition Department 233: Specific part 234 :Generation part 235: Artificial Intelligence Department 236: Display control unit 30: Communication bus 31: Communications Department 32: Storage section 33: Processor 34: Display section 35: Input section 5: Input screen 51 :Area 52 :Area 53: Area 54 :Area 55 :Area 56 :Area 57: Object 6:Generation result screen 61 :Area 62 :Area 63 :Area 64 :Area 65: area 66: area 67 :Area 68 :Area 69 :Area 7: Search results screen 70: area 71 :Area 71A :Area 72 :Area 72A : Object 73 :Area HT: Hashtag IF1: Advertising target information IF11: Name IF12 :Date information IF2 : Related information LM: Learning model LM1: Large-scale language model PR: Prompt T1: Related information table TC: Text content TC1: Text content TC2: Text content TC3: Text content X :User

Claims

1. An information processing system, comprising: A processor capable of executing a program to perform the following steps: In the reception step, the input of advertisement target information is received from the user, The advertising target information includes at least information regarding the advertising target and date information indicating a date, In a first generation step, the advertising target information and an instruction to generate one or more sentences related to advertising based on the advertising target information are input to a learning model that is a large-scale language model to generate the sentences, The instruction includes content for including in the text a style of writing that matches attributes of a target person who will read the text, the style being identifiable based on the advertising target information; The text includes relevant information corresponding to the date information; A system in which the related information is information derived from the date and includes different information for each date.

2. 2. The information processing system according to claim 1, Furthermore, in the identification step, the related information is identified based on the date information and predetermined reference information; A system, wherein the reference information is configured to be able to refer to at least the date information and the related information.

3. 3. The information processing system according to claim 2, the reference information is configured to be able to refer to a plurality of pieces of related information with respect to the date information; In the first generation step, a plurality of the sentences can be generated, A system in which each of the multiple sentences contains different related information.

4. 2. The information processing system according to claim 1, A system in which the instructions include content for including in the text the name of a store or the name of a product or service that can be identified based on the advertising target information.

5. 2. The information processing system according to claim 1, Furthermore, in a second generation step, keywords related to the sentence are generated by the learning model; In the display control step, a text content in which the keyword and the sentence are associated with each other is displayed so as to be visually recognizable by the user; The system, wherein the text content includes a composition that can be posted by the user to a given site.

6. 1. An information processing method, comprising: A method comprising the steps of the information processing system according to any one of claims 1 to 5.

7. A program, A program for causing at least one computer to execute each step of the information processing system according to any one of claims 1 to 5.

Images