Apparatus and method for generating product detail page on basis of artificial intelligence

Abstract

The present invention relates to an apparatus and a method for generating a product detail page on the basis of artificial intelligence. The apparatus for generating a product detail page on the basis of artificial intelligence, according to one aspect of the present invention, comprises: a question database for storing question content of each of preset question items, and a question identifier for distinguishing each question; an answer acquisition unit for displaying the question content of each of the question items stored in the question database, acquiring a voice answer to each question and converting same into text, and matching the converted text to the question identifier of each question item and storing same; a product description prompt generation unit, which uses the text stored in the answer acquisition unit so as to generate a product description prompt including, as content, answers to the preset question items; an integrated prompt generation unit for generating an integrated prompt by combining a pre-stored additional description prompt with the product description prompt in order to guide the generation direction of a product detail page; and a product detail page generation unit, which inputs the integrated prompt into a preset generative model so as to generate and output the product detail page.

Description

Device and method for generating product detail pages based on artificial intelligence

[0001] The present invention relates to a technology for automatically generating a product detail page for describing a product.

[0002] As e-commerce technology advances and becomes more widespread, individuals and small businesses can now register on e-commerce platforms and sell products directly to consumers. Encouraging online purchases requires uploading an appropriate product detail page that facilitates understanding of the product. Sellers unfamiliar with product detail page creation often struggle with the process itself, or even after completing the page, they often struggle to properly include product information, resulting in poor sales.

[0003] In this regard, a technology has been disclosed to facilitate the creation of product detail pages. This technology generates product detail pages by receiving product information and reference images from a terminal. This conventional technology automatically generates product detail pages based on the design layout of the reference images, requiring users to manually find the reference images. Furthermore, if an inappropriate reference image is registered, the product detail page may not be properly created.

[0004] The purpose of the present invention is to solve the above problem, and to provide a technology for automatically generating a prompt for a query to a generative model using a user's answer to a question and creating a product detail page using the generative model.

[0005] The purpose of the present invention is not limited to the purposes mentioned above, and other purposes not mentioned can be clearly understood from the description below.

[0006] In order to achieve the above-described object, an artificial intelligence-based product detail page generation device according to one aspect of the present invention includes a question database storing question contents of each of preset question items including at least one of a product name, a place of production, a producer, a story, a product type, a storage method, and a sales location, and a question identifier for distinguishing them from other questions; an answer acquisition unit displaying the question contents of each of the question items stored in the question database, acquiring a voice answer, converting the answer into text, and matching the converted text with the question identifier of each question item and storing it; a product description prompt generation unit generating a product description prompt including answers to preset question items as content using the text stored in the answer acquisition unit; an integrated prompt generation unit generating an integrated prompt by combining a pre-stored additional description prompt with the product description prompt to guide the direction of product detail page generation by specifying a producer as a speaker describing a product, and including content such as the amount of text, text structure, speech style, and whether the writing is in spoken or written language depending on the writing location; and a product detail page generation unit inputting the integrated prompt into a preset generative model to generate and output a product detail page.

[0007] According to another aspect of the present invention, a method for generating a product detail page based on artificial intelligence is provided, which method is performed by an electronic device, and includes the steps of: displaying pre-stored question contents for each of preset question items including at least one of a product name, a place of production, a producer, a story, a product type, a storage method, and a sales location; obtaining answers by obtaining voice answers and converting them into text; and matching the converted text with a pre-stored question identifier for the preset question items and storing it; generating a product description prompt including answers to the preset question items as content using the text stored in the step of obtaining answers; specifying a producer as a speaker describing a product, and combining a pre-stored additional description prompt with the product description prompt to guide the direction of generating a product detail page including content such as the amount of text, text structure, speech style, and whether the text is written in spoken or written language depending on the location of writing, to generate an integrated prompt; and inputting the integrated prompt into a preset generative model to generate and output a product detail page.

[0008] According to the present invention, a prompt for a query is automatically generated in a generative model using a user's answer to a question, thereby enabling the creation of a product detail page easily without the difficulty of creating a prompt.

[0009] Additionally, by reflecting the speech style selected by the user in the prompt, it has the effect of providing the user with a product detail page written in the speech style of the online platform they want.

[0010] The effects of the present invention are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the description of the claims.

[0011] FIG. 1 is a block diagram of an artificial intelligence-based product detail page generation device according to one embodiment of the present invention.

[0012] Figure 2 is a flowchart of a method for generating an artificial intelligence-based product detail page according to another embodiment of the present invention.

[0013] FIG. 3 is a drawing showing a first input screen provided according to embodiments of the present invention.

[0014] FIG. 4 is a drawing showing a second input screen provided according to embodiments of the present invention.

[0015] FIG. 5 is a drawing showing a third input screen provided according to embodiments of the present invention.

[0016] FIG. 6 is a drawing showing a display screen provided according to embodiments of the present invention.

[0017] In order to achieve the above-described object, an artificial intelligence-based product detail page generation device according to one aspect of the present invention includes a question database storing question contents of each of preset question items including at least one of a product name, a place of production, a producer, a story, a product type, a storage method, and a sales location, and a question identifier for distinguishing them from other questions; an answer acquisition unit displaying the question contents of each of the question items stored in the question database, acquiring a voice answer, converting the answer into text, and matching the converted text with the question identifier of each question item and storing it; a product description prompt generation unit generating a product description prompt including answers to preset question items as content using the text stored in the answer acquisition unit; an integrated prompt generation unit generating an integrated prompt by combining a pre-stored additional description prompt with the product description prompt to guide the direction of product detail page generation by specifying a producer as a speaker describing a product, and including content such as the amount of text, text structure, speech style, and whether the writing is in spoken or written language depending on the writing location; and a product detail page generation unit inputting the integrated prompt into a preset generative model to generate and output a product detail page.

[0018] The above integrated prompt generation unit displays the product description prompt and can select any one of the preset speech styles according to the user's input, outputs a display screen for receiving a completion request according to the user's input, receives a completion request according to the user's input on the display screen, and generates an integrated prompt by combining the additional description prompt with the product description prompt if the speech style is not selected according to whether the speech style is selected on the display screen.

[0019] The above integrated prompt generation unit, when a speech style is selected, loads a pre-saved sentence template including a data field corresponding to the speech style, inputs pre-saved text for the selected speech style, generates a speech guide sentence, generates a speech reflection prompt that reflects the speech guide sentence in the additional description prompt, and combines the speech reflection prompt with the product description prompt to generate an integrated prompt.

[0020] The above product description prompt generation unit loads a pre-saved prompt template including a data field matching each question identifier for the above question items to include answers to preset question items as content, and inputs the text stored in the answer acquisition unit into each data field of the prompt template according to the matched question identifier to generate a product description prompt.

[0021] According to another aspect of the present invention, a method for generating an artificial intelligence-based product detail page is provided, which is performed by an electronic device, and includes the steps of: displaying pre-stored question content for each of preset question items, including at least one of a product name, a place of production, a producer, a story, a product type, a storage method, and a sales location; obtaining an answer by obtaining a voice answer, converting the voice answer into text, and matching the converted text with a pre-stored question identifier for the preset question items and storing it; generating a product description prompt including answers to the preset question items as content using the text stored in the step of obtaining the answer; specifying a producer as a speaker describing the product, and combining a pre-stored additional description prompt with the product description prompt to guide the direction of generating the product detail page, including content such as the amount of text, text structure, speech style, and whether the text is written in spoken or written language depending on the location of writing, to generate an integrated prompt; and inputting the integrated prompt into a preset generative model to generate and output a product detail page.

[0022] The step of generating the above integrated prompt includes the step of displaying the product description prompt and selecting any one of preset speech styles according to a user's input, outputting a display screen for receiving a completion request according to the user's input, the step of receiving a completion request according to the user's input on the display screen, the step of determining whether a speech style is selected on the display screen, and the step of generating the integrated prompt by combining the product description prompt with the additional description prompt if the speech style is not selected.

[0023] The step of generating the above integrated prompt further includes the step of determining whether a speech style is selected on the display screen, and if the speech style is selected, the step of loading a pre-stored sentence template including a data field corresponding to the speech style and inputting pre-stored text for the selected speech style to generate a speech guide sentence, the step of generating a speech reflection prompt that reflects the speech guide sentence in the additional description prompt, and the step of combining the speech reflection prompt with the product description prompt to generate an integrated prompt.

[0024] The step of generating the above product description prompt includes loading a pre-saved prompt template including a data field matching each question identifier for the above question items to include answers to the preset question items as content, and inputting the text stored in the answer acquisition section into each data field of the prompt template according to the matched question identifier to generate the product description prompt.

[0025] The advantages and features of the present invention, and the methods for achieving them, will become clearer with reference to the embodiments described in detail below together with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in various different forms. These embodiments are provided only to ensure that the disclosure of the present invention is complete and to fully inform those skilled in the art of the scope of the invention, and the present invention is defined only by the description of the claims. Meanwhile, the terminology used in this specification is for the purpose of describing the embodiments and is not intended to limit the present invention. In this specification, the singular includes the plural unless specifically stated otherwise.

[0026] The present invention relates to a technology for automatically generating a product detail page.

[0027] In particular, the present invention is characterized by providing sufficient information about a product based on an answer to a question and providing a product detail page that applies the speech style of an online platform desired by the user according to the speech style selected by the user.

[0028] These technical features can be achieved by a configuration that automatically generates prompts for querying the generative model based on the user's answers to questions and the speaking style chosen by the user.

[0029] Hereinafter, with reference to the attached drawings, the artificial intelligence-based product detail page generation technology according to the present invention will be described in detail.

[0030] Referring to FIG. 1, an artificial intelligence-based product detail page generation device (10) according to one embodiment of the present invention may include a question database (110), an answer acquisition unit (120), a product description prompt generation unit (130), an integrated prompt generation unit (140), a product detail page generation unit (150), and a communication unit (160).

[0031] The method for generating an artificial intelligence-based product detail page according to another embodiment of the present invention can be performed by an artificial intelligence-based product detail page generating device (10) according to one embodiment of the present invention.

[0032] For convenience of explanation, the drawing symbols are matched for functionally identical contents and duplicate explanations are avoided.

[0033] The question database (110) can store the product name, the question content of each preset question item, and a question identifier to distinguish it from other questions.

[0034] Pre-defined questions may include product name, country of origin, producer, story, differentiation, customer value, product type, packaging method, customer response, storage method, and sales location.

[0035] And, for each question item, the following questions can be saved: ‘Please tell me the name of the product.’ for the product name, ‘Please tell me where it was produced.’ for the production site, ‘Please tell me the name of the company, farm, or representative that produced it.’ for the producer, ‘Please tell me if the producer has a special story.’ for the story, ‘Please tell me what makes this product different.’ for differentiation, ‘What value will customers who purchase this product receive?’ for customer value, ‘What value will customers who purchase this product receive?’ for the product type, ‘Is this product a low-cost product or a high-quality premium product?’ for the packaging method, ‘How is this product packaged?’ for the customer response, ‘What were the responses of customers who purchased this product?’ for the storage method, ‘Please tell me the precautions for storage.’ for the sales location, ‘Is there a special sales location that sells this product?’

[0036] The answer acquisition unit (120) can acquire answers by displaying the question content of each question item stored in the question database (110), acquiring a voice answer for each question item, converting it into text, and storing the converted text by matching it with the question identifier of each question item (S110).

[0037] The answer acquisition unit (120) outputs a first input screen for receiving either voice or text as an answer method according to the user's input, so that either voice or text can be selected as an answer method according to the user's input (S111).

[0038] Referring to FIG. 3, the first input screen (1100) may include a voice start button (1110) for receiving voice input in response to a user's input and a writing start button (1120) for receiving text input in response to a user's input.

[0039] The answer acquisition unit (120) determines whether the selected answer method is voice (S112), and if the selected answer method is voice, it displays the contents of each question item in a preset order for the question items stored in the question database (110), outputs a second input screen for obtaining a voice answer for each question item according to the user's input (S113), and can obtain a voice answer for each question item according to the user's input on the second input screen (S115).

[0040] Here, the second input screen exists in multiple numbers corresponding to preset question items, is output in a predetermined order, and can be switched to the next order in response to the voice answer obtained for each question item according to the user's input.

[0041] Referring to FIG. 4, the second input screen (1200) may be composed of a question display screen (1210) and a voice recognition screen (1220).

[0042] The question display screen (1210) may include a microphone icon (1211) for performing voice recognition according to the user's input and may display the question content of each question item.

[0043] The voice recognition screen (1220) can be switched in response to a user input to the microphone icon (1211) on the question display screen (1210).

[0044] The voice recognition screen (1220) may include a plurality of buttons (1221) for starting, pausing, and ending voice recording according to the user's input, and a completion button (1223) for obtaining the recorded voice as a voice response according to the user's input.

[0045] The answer acquisition unit (120) can convert the voice answers acquired for each question item into text (S117) and store them by matching them with question identifiers for each question item (S118).

[0046] At this time, converting a voice response into text can be performed using a speech to text conversion model that has been developed and is currently in use, and as it is not a core matter of the present invention, a detailed description thereof will be omitted.

[0047] The answer acquisition unit (120) determines whether the selected answer method is voice in the step (S112) where it is determined that the selected answer method is not voice, displays the contents of each question item in a preset order for the question items stored in the question database (110), outputs a third input screen for obtaining a text answer for each question item according to the user's input (S114), obtains a text answer for each question item according to the user's input on the third input screen (S116), and stores the text obtained for each question item by matching it with a question identifier (S118).

[0048] Referring to FIG. 5, the third input screen (1300) may include text displaying the question content for each question item, an input area for receiving text for each question from the user by activating the keyboard area according to the user's input, and a next button for obtaining the text entered in the input area according to the user's input as a text answer for each question item.

[0049] After the step (S118) of matching text for each question item with a question identifier and storing it, the answer acquisition unit (120) counts the number of characters in the text matched to each question identifier and compares whether the counted value exceeds a preset minimum value (e.g., 2). If the counted value does not exceed the preset minimum value, it can be determined that no answer has been given.

[0050] If there is a question item that is judged to have not been answered, the answer acquisition unit (120) displays the question content for the question item and outputs an input screen for obtaining a voice answer (a text answer if the answer method is selected as text) so that the answer can be obtained again.

[0051] The product description prompt generation unit (130) can generate a product description prompt by loading a pre-saved prompt template and entering text stored in the answer acquisition unit into each data field of the prompt template according to the matched question identifier (S120).

[0052] A prompt template may include a plurality of data fields, each of which has question identifiers corresponding to preset question items, to include content for answers to preset question items, and preset text to form sentences before and after each data field.

[0053] A prompt template can include multiple data fields (${}) matching the question identifier and predefined text to construct sentences, such as "Create a product detail page that can sell products online. Reference the top 100 best-selling products from ${Seller}. The name of this product is ${Product Name}, it was produced in {$Country}, and {$Manufacturer} made it. … "

[0054] The integrated prompt generation unit (140) can generate an integrated prompt by combining a pre-saved additional description prompt with the product description prompt generated by the product description prompt generation unit to guide the direction of creation of a product detail page by specifying a producer as a speaker describing a product and including content such as the amount of text, text structure, speech style, and whether the text is written in spoken or written language depending on the location of writing (S130).

[0055] Here, concatenation means combining at least two strings to create one string, and conventional character concatenation functions (concatenatem, strcat, +, &) can be used.

[0056] That is, the integrated prompt generation unit (140) can generate an integrated prompt consisting of a single combined string by combining the string of the product description prompt and the string of the additional description prompt in the order of the string of the product description prompt and the string of the additional description prompt (S130).

[0057] The integrated prompt generation unit (140) displays the product description prompt generated by the product description prompt generation unit (130), and can select any preset speech style according to the user's input, and outputs a display screen for receiving a completion request according to the user's input, thereby displaying the product description prompt (S131), and can receive a completion request according to the user's input on the display screen (S132).

[0058] Referring to FIG. 6, the display screen (2000) may include a first area (2100), a second area (2200), and a completion button (2300).

[0059] The first area (2100) can display a product description prompt.

[0060] The second area (2200) may include a plurality of style buttons (2210, 2220) for selecting a speech style according to user input in response to preset speech styles.

[0061] Here, multiple speech styles can be preset to correspond to existing online platforms (Instagram, Naver Blog, etc.) where product-related posts can be posted.

[0062] The integrated prompt generation unit (140) can select the first speech style corresponding to the first style button when a user input occurs on any of the first style buttons in the second area of ​​the display screen.

[0063] The integrated prompt generation unit (140) can deselect the first speech style when a user input occurs again on the first style button while the first speech style is selected, and can deselect the first speech style and select the second speech style when a user input occurs on the second style button corresponding to a second speech style different from the first speech style while the first speech style is selected.

[0064] The completion button (2300) is for receiving a completion request based on user input, and the integrated prompt generation unit (140) can receive a completion request when a user input occurs in the completion button (2300).

[0065] When a completion request is input through a display screen (S131), the integrated prompt generation unit (140) determines whether a speech style has been selected through the display screen (S133), and if the selected speech style does not exist, the integrated prompt can be generated by combining a pre-saved additional description prompt with the product description prompt generated by the product description prompt generation unit (130) to guide the direction of creation of a product detail page (S135).

[0066] Here, additional description prompts can be pre-saved, including a first sentence of text that identifies the producer as the speaker describing the product, a second sentence of text that defines the length and structure of the text, a third sentence of text that defines whether the writing is spoken or written based on the location, and a fourth sentence of text that defines the speaking style.

[0067] Additional explanation prompts may include the first sentence, "The product description script should be written as if the producer wrote it himself. The producer should be referred to as 'we' and used as the subject.", the second sentence, "The script should be approximately 2000 bytes, and each paragraph should consist of three sentences.", the third sentence, "For each paragraph, write a subheading that summarizes the content, and use a polite tone to create a sense of trust. After writing the first three paragraphs in a formal tone, write the next paragraph using quotation marks as if the producer were talking to a customer. The phrases in quotation marks should be polite but colloquial.", and the fourth sentence, "You can speak in the Naver blog style, but you should express it in quotation marks."

[0068] In the step (S133) of determining whether a speech style is selected through a display screen, if it is determined that the selected speech style exists, the integrated prompt generation unit (140) loads a pre-stored sentence template including a data field corresponding to the speech style, inputs pre-stored text for the selected speech style, and generates a speech guide sentence (S134), and can generate a speech reflection prompt that reflects the speech guide sentence in a pre-stored additional explanation prompt (S135).

[0069] Here, the sentence template can be something like 'You should speak in ${speech style} and express it by enclosing it in quotation marks.'

[0070] The integrated prompt generation unit (140) can generate a speech style reflection prompt by replacing the fourth sentence, which is composed of text that limits the speech style among the sentences that constitute the additional description prompt, with a speech style guide sentence (S135), and can generate an integrated prompt by combining the speech style reflection prompt with the product description prompt generated by the product description prompt generation unit (130) (S138).

[0071] In summary, when a completion request is input from a user, the integrated prompt generation unit (140) can generate an integrated prompt by combining a pre-saved additional description prompt to guide the creation of a detailed page with a product description prompt if the speech style is not selected, and can generate an integrated prompt by combining the product description prompt with the speech style selected in the additional description prompt.

[0072] Therefore, the difficulty of having users directly write prompts to input into a generative model to create product detail pages is solved, while allowing users to selectively reflect the speech style of the online platform they want in the prompts.

[0073] The product detail page generation unit (150) can input the integrated prompt generated by the integrated prompt generation unit (140) into a preset generation model to generate and output a product detail page (S140).

[0074] Here, a generative model can be applied, which is an interactive language model that receives data corresponding to content and a prompt as input and generates data that is transformed from the data corresponding to the content according to the prompt.

[0075] The generative model can be trained in advance to receive text and images from product detail pages posted for each product on predetermined online platforms and generate product detail pages by referencing previously published product detail pages.

[0076] The communication unit (160) is connected to the user terminal (20) so as to be communicable through a network, and outputs screens provided by the answer acquisition unit (120), the integrated prompt generation unit (140), and the product detail page generation unit (150) to the user terminal (20) and receives user input for the screen from the user terminal (20).

[0077] The user terminal (20) can be implemented as a smartphone, tablet, desktop, laptop, etc.

[0078] According to the present invention, a product description prompt is generated based on a user's answer to a question, and an integrated prompt is generated for querying a generative model by combining the product description prompt with a pre-stored additional description prompt to guide the creation of a detail page, thereby enabling the creation of a product detail page using the generative model.

[0079] That is, when a user answers a question, a prompt is automatically generated, which solves the difficulty of the user writing the prompt directly.

[0080] Additionally, by reflecting the speech style selected by the user in the prompt, it has the advantage of providing a product detail page that reflects the speech style desired by the user without the hassle of separately finding reference materials for creating a product detail page.

[0081] Meanwhile, the blocks in the attached block diagram and the steps in the flowchart may be implemented as computer instructions that are loaded into the processor or memory of an electronic device capable of data processing (e.g., a general-purpose computer, a special-purpose computer, a portable laptop computer, a network computer) and perform designated functions. Since these computer program instructions can be stored in a computer-readable memory, the functions described in the blocks in the block diagram or the steps in the flowchart may also be produced as a product that includes instruction means for performing them.

[0082] Those skilled in the art will appreciate that the present invention can be implemented in other specific forms without altering its technical spirit or essential characteristics. Therefore, the embodiments described above should be understood as illustrative in all respects and not restrictive. The scope of the present invention is defined by the claims below rather than the detailed description above, and all changes or modifications derived from the claims and their equivalents should be construed as being included within the scope of the present invention.

[0083] The present invention relates to a technology for automatically generating a product detail page for describing a product.

Claims

1. A question database that stores the question content of each preset question item, including at least one of product name, production site, producer, story, product type, storage method, and sales location, and a question identifier to distinguish it from other questions; An answer acquisition unit that displays the question content of each question item stored in the above question database, obtains a voice answer, converts it into text, and stores the converted text by matching it with the question identifier of each question item; A product description prompt generation unit that generates a product description prompt including answers to preset question items as content using the text stored in the above answer acquisition unit; An integrated prompt generation unit that combines pre-saved additional description prompts with the product description prompt to guide the direction of product detail page creation, including the content of the producer as the speaker describing the product, the length of the text, the structure of the text, the style of speech, and whether the text is spoken or written depending on the location of writing, to create an integrated prompt; and An artificial intelligence-based product detail page generation device including a product detail page generation unit that inputs the above integrated prompt into a preset generative model to generate and output a product detail page.

2. In paragraph 1, The above integrated prompt generation unit Display the above product description prompt and select one of the preset speech styles based on the user's input, and output a display screen for receiving a completion request based on the user's input. A completion request is received based on the user's input on the above display screen. If a speech style is not selected in the above display screen, the additional description prompt is combined with the product description prompt to create an integrated prompt. A device for generating product detail pages based on artificial intelligence.

3. In paragraph 2, The above integrated prompt generation unit When a speech style is selected, a pre-saved sentence template is loaded, including data fields corresponding to the speech style, and a speech guide sentence is generated by entering the pre-saved text for the selected speech style. Create a speech reflection prompt that reflects the speech guide sentences in the additional explanation prompt above, Create an integrated prompt by combining the above speech reflection prompt with the above product description prompt. A device for generating product detail pages based on artificial intelligence.

4. In paragraph 1, The above product description prompt generation section In order to include answers to preset question items as content, a pre-saved prompt template including a data field matching each question identifier for the above question items is loaded, and the text stored in the answer acquisition section is entered into each data field of the prompt template according to the matched question identifier to create a product description prompt. A device for generating product detail pages based on artificial intelligence.

5. Regarding a method performed by an electronic device, A step of displaying pre-stored question content for each of the preset question items, including at least one of product name, production site, producer, story, product type, storage method, and sales location, obtaining a voice answer, converting it into text, and storing the converted text by matching it with a pre-stored question identifier for the preset question items to obtain an answer; A step of generating a product description prompt including answers to preset question items as content using the text saved in the step of obtaining the above answers; A step of creating an integrated prompt by combining a pre-saved additional description prompt with the product description prompt to guide the direction of creating a product detail page, including the content such as the producer as the speaker describing the product, the length of the text, the structure of the text, the style of speech, and whether the text is spoken or written depending on the location of writing; and An artificial intelligence-based product detail page generation method, comprising: a step of inputting the above integrated prompt into a preset generative model to generate and output a product detail page; 6. In paragraph 5, The steps to create the above integrated prompt are A step of displaying the above product description prompt and selecting one of the preset speech styles based on the user's input, and outputting a display screen for receiving a completion request based on the user's input. A step of receiving a completion request based on the user's input on the above display screen; A step for determining whether a speech style is selected on the above display screen; If no speech style is selected, the step of combining the above additional description prompt with the above product description prompt to create a unified prompt is included. How to create a product detail page based on artificial intelligence.

7. In paragraph 6, The steps to create the above integrated prompt are After the step of determining whether the speech style is selected in the above display screen, When a speech style is selected, a step is taken to load a pre-saved sentence template including data fields corresponding to the speech style and create a speech guide sentence by entering the pre-saved text for the selected speech style. A step for creating a speech reflection prompt that reflects the speech guide sentence in the above additional explanation prompt; and Further comprising the step of combining the above speech reflection prompt with the above product description prompt to create an integrated prompt. How to create a product detail page based on artificial intelligence.

8. In paragraph 5, The steps to create the above product description prompt are: In order to include answers to preset question items as content, a pre-saved prompt template including a data field matching each question identifier for the above question items is loaded, and the text stored in the answer acquisition section is entered into each data field of the prompt template according to the matched question identifier to create a product description prompt. How to create a product detail page based on artificial intelligence.

Images