KR102613621B1 - Method, device and system for generating articles for company promotion using generative artificial intelligence model - Google Patents

Abstract

According to one embodiment, the device acquires company information of the company from the company's terminal, obtains product information of the product from the company's terminal, and collects web pages related to the company online based on the company information and product information, By analyzing the text of the collected web pages using natural language processing techniques, the text of the collected web pages is divided into useful and unused phrases, non-text content from the collected web pages is saved as online content, and corporate information and Product information is entered into a generative artificial intelligence model to obtain a draft article, and additional information including usage phrases and online content is used to correct the draft article to create a promotional article for the company.

Description

Method, device, and system for generating corporate promotional articles using generative artificial intelligence models {METHOD, DEVICE AND SYSTEM FOR GENERATING ARTICLES FOR COMPANY PROMOTION USING GENERATIVE ARTIFICIAL INTELLIGENCE MODEL}

The examples below relate to technology for generating corporate promotional articles using a generative artificial intelligence model.

For companies, public relations and marketing are very important factors in the life and death of a company, and in the public relations, marketing and advertising market, the online market is as important to companies as the offline market.

Additionally, with the development of mobile data networks, many changes have occurred in advertising formats, and many companies are promoting their companies and products through press releases, or articles.

Meanwhile, companies and advertising agencies spend a lot of time and money on writing articles, checking for typos, and editing to produce articles for corporate promotion.

However, as the amount of advertising increases rapidly, it is difficult for one-size-fits-all advertising to attract people's continued attention, and due to the incomprehensibility of information, it does not satisfy consumers' needs, does not provide sufficient interest, is overly stimulating, or is too irritating. Due to various factors, such as not meeting consumers' needs or benefiting them, not all advertisements are able to fulfill their intended role for consumers.

Therefore, research on methods to generate high-quality advertisements with high efficiency is required.

Korean Patent No. 10-2078505 (announced on February 17, 2020) Korean Patent No. 10-1638423 (announced on July 12, 2016) Korean Patent No. 10-2410260 (announced on June 22, 2022) Korean Patent No. 10-0853022 (announced on August 20, 2008)

Embodiments seek to obtain article drafts using a generative artificial intelligence model.

Embodiments seek to create corporate promotional articles by correcting article drafts obtained through a generative artificial intelligence model.

Embodiments seek to generate corporate promotional articles by inserting online content.

According to one embodiment, a method of generating a corporate promotional article using a generative artificial intelligence model includes the name of the company, the industry of the company, the history of the company, the address of the company, and the contact information of the company from the company's terminal. Obtaining corporate information of the company; Obtaining product information of the product including the name of the product provided by the company, the category of the product, the characteristics of the product, and the image of the product from the terminal of the company; collecting web pages related to the company online based on the company information and the product information; Analyzing the collected text of the web page using natural language processing techniques to classify the collected text of the web page into useful phrases and unused phrases; storing content other than the text in the collected web pages as online content; Obtaining a draft article by inputting the company information and the product information into a generative artificial intelligence model; and generating a corporate promotional article by correcting the draft article using additional information including the used phrase and the online content.

In the step of obtaining a draft article by inputting the company information and the product information into the generative artificial intelligence model, the information input into the generative artificial intelligence model is searched on a website using search terms generated based on the product information. This is information about products that are included more than the preset standard number of published materials published within a preset threshold period.

The step of generating a corporate promotional article by correcting the draft article using the use phrase and additional information including the online content includes extracting phrases related to a specific product mentioned in the article draft among the use phrases as related phrases. A step of searching for the specific product online and setting keywords with a higher frequency of appearance than the preset standard frequency among keywords excluding the specific product as trend keywords in the results output, using the related phrases and the trend keywords. Obtaining a revised article by inputting the draft article, the related phrases, and the trend keywords into a generative artificial intelligence model so that the draft article is modified; extracting content related to the specific product from the online content; extracting content related to the specific product from the online content; Giving priority to the extracted content according to the number of views of the original web page containing the content and the rating of the site on which the original web page is posted, the number of times the specific product is mentioned in each paragraph of the article amendment, and A step of assigning priority to each paragraph according to the degree to which it contains a phrase affirming the specific product, and arranging the extracted content according to the priority of each paragraph and the priority of the extracted content. do.

In the step of arranging the extracted content, the step further includes setting the size of the extracted content, and the step of setting the size of the extracted content includes checking the release date of the specific product, and determining the release date of the specific product. generating a confirmation period that is a period between the release date and the current time; determining whether the confirmation period is longer than a preset first reference period; if it is determined that the confirmation period is longer than the first reference period, the extracted Setting the size of the content to a first size, if it is confirmed that the confirmation period is shorter than or equal to the first reference period, determining whether the confirmation period is longer than a preset second reference period, the confirmation period If confirmed to be longer than the second reference period, setting the size of the extracted content to a second size larger than the first size, and if confirmed to be shorter than or equal to the second reference period. , generating a weight in inverse proportion to the confirmation period, setting a third size by applying the generated weight to the second size, and setting the size of the extracted content to the third size.

A method of generating promotional articles for a company using a generative artificial intelligence model includes the steps of acquiring existing promotional articles for the company; Confirming the writing style of the existing promotional article of the company based on the existing promotional article of the company; Correcting the writing style of the company's promotional article based on the writing style of the company's existing promotional article; Selecting an article written within a preset target period among the company's existing promotional articles as a reference article; Based on the reference article, checking phrases used in a preset first part and selecting the most used phrase among the identified phrases as the first phrase; Based on the reference article, checking the phrases used in the preset second part and selecting the most used phrase among the confirmed phrases as the second phrase; And inserting the first phrase into the first part of the corporate promotional article and inserting the second phrase into the second part of the corporate promotional article, using a generative artificial intelligence model. The method of generating a corporate promotional article further includes the step of adding a link related to the specific product to the corporate promotional article, and the step of adding a link related to the specific product to the corporate promotional article includes providing the specific product through SNS. Confirming the influencer who promoted the product, obtaining the influencer's SNS account, and promoting the influencer's product by linking a link containing a promotional post for the specific product based on the influencer's SNS account It includes obtaining a link, confirming a purchase link for purchasing the specific product, and adding the influencer's product promotion link and the purchase link to the company promotional article.

A method of generating a corporate promotional article using a generative artificial intelligence model includes the steps of: when it is confirmed that the corporate promotional article has been sent at a first time, storing information about the company in a first company database; If it is confirmed that the company has confirmed the company promotional article, deleting the company information stored in the first company database and storing the company information in a second company database; According to the enterprise level of the enterprise, setting the storage period of the enterprise information as a first storage period; If it is confirmed that no additional corporate promotional articles have been sent to the company within the first storage period from the first point in time, deleting the information about the company stored in the second company database; If it is confirmed that an additional corporate promotional article has been provided to the company at a second time, setting a second storage period by applying a second weight set to a value less than 1 to the first storage period; And if it is confirmed that no additional corporate promotional articles have been sent to the company within the second storage period from the second point in time, it further includes deleting the information about the company stored in the second company database.

Embodiments may obtain draft articles using a generative artificial intelligence model based on brief information obtained from a company's terminal.

Embodiments may generate corporate promotional articles by correcting article drafts obtained through a generative artificial intelligence model.

Embodiments may generate corporate promotional articles by inserting online content including photos and videos as well as text.

Meanwhile, the effects according to the embodiments are not limited to those mentioned above, and other effects not mentioned can be clearly understood by those skilled in the art from the description below.

1 is a diagram schematically showing the configuration of a system according to an embodiment.
Figure 2 is a flow chart to explain the process of generating a corporate promotional article using a generative artificial intelligence model according to an embodiment.
Figure 3 is a flow chart to explain the process of processing product information by inputting it into a generative artificial intelligence model according to an embodiment.
Figure 4 is a flow chart to explain the process of correcting a draft article and creating a corporate promotional article using additional information according to an embodiment.
Figure 5 is a flowchart for explaining the process of setting the size of extracted content according to an embodiment.
Figure 6 is a flow chart to explain the process of supplementing a corporate promotional article according to an embodiment.
Figure 7 is a flowchart for explaining the process of adding a link related to a specific product according to an embodiment.
Figure 8 is a flow chart to explain the process of managing company information according to an embodiment.
Figure 9 is an exemplary diagram of the configuration of a device according to an embodiment.

Hereinafter, embodiments will be described in detail with reference to the attached drawings. However, various changes can be made to the embodiments, so the scope of the patent application is not limited or limited by these embodiments. It should be understood that all changes, equivalents, or substitutes for the embodiments are included in the scope of rights.

Specific structural or functional descriptions of the embodiments are disclosed for illustrative purposes only and may be modified and implemented in various forms. Accordingly, the embodiments are not limited to the specific disclosed form, and the scope of the present specification includes changes, equivalents, or substitutes included in the technical spirit.

Terms such as first or second may be used to describe various components, but these terms should be interpreted only for the purpose of distinguishing one component from another component. For example, a first component may be named a second component, and similarly, the second component may also be named a first component.

When a component is referred to as being “connected” to another component, it should be understood that it may be directly connected or connected to the other component, but that other components may exist in between.

The terms used in the examples are for descriptive purposes only and should not be construed as limiting. Singular expressions include plural expressions unless the context clearly dictates otherwise. In this specification, terms such as “comprise” or “have” are intended to designate the presence of features, numbers, steps, operations, components, parts, or combinations thereof described in the specification, but are not intended to indicate the presence of one or more other features. It should be understood that this does not exclude in advance the possibility of the existence or addition of elements, numbers, steps, operations, components, parts, or combinations thereof.

Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as generally understood by a person of ordinary skill in the technical field to which the embodiments belong. Terms defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the related technology, and unless explicitly defined in the present application, should not be interpreted in an ideal or excessively formal sense. No.

In addition, when describing with reference to the accompanying drawings, identical components will be assigned the same reference numerals regardless of the reference numerals, and overlapping descriptions thereof will be omitted. In describing the embodiments, if it is determined that detailed descriptions of related known technologies may unnecessarily obscure the gist of the embodiments, the detailed descriptions are omitted.

Embodiments may be implemented in various types of products such as personal computers, laptop computers, tablet computers, smart phones, televisions, smart home appliances, intelligent vehicles, kiosks, and wearable devices.

1 is a diagram schematically showing the configuration of a system according to an embodiment.

Referring to FIG. 1, a system according to an embodiment may include a corporate terminal 100 and a device 200 that can communicate with each other through a communication network.

First, a communication network can be configured regardless of the communication mode, such as wired or wireless, and can be implemented in various forms to enable communication between servers and between servers and terminals.

The corporate terminal 100 is a terminal used by companies that wish to obtain corporate promotional articles that can promote the company through brief information about the company. It can be used through mobile phones, desktop PCs, laptop PCs, tablet PCs, smartphones, etc. It may be implemented, but is not limited to this, and may be implemented as various types of communication devices that can be connected to an external server. For example, as shown in FIG. 1, the corporate terminal 100 may be a smartphone, and may be employed differently depending on the embodiment.

The corporate terminal 100 may be configured to perform all or part of the calculation function, storage/reference function, input/output function, and control function of a typical computer. The company's terminal 100 may be configured to communicate with the device 200 in a wired or wireless manner.

The company's terminal 100 is connected to a website operated by a person or organization that provides services using the device 200, or an application developed and distributed by a person or organization that provides services using the device 200. This can be installed. The company's terminal 100 may be linked to the device 200 through a website or application.

In FIG. 1 and the description below, for convenience of explanation, only the enterprise terminal 100 is shown and described, but the number of terminals may vary depending on the embodiment. As long as the processing capacity of the device 200 allows, the number of terminals is not particularly limited.

The device 200 may be its own server owned by a person or organization that provides services using the device 200, a cloud server, or a p2p (peer-to-peer) set of distributed nodes. It may be possible. The device 200 may be configured to perform all or part of the calculation function, storage/reference function, input/output function, and control function of a typical computer. The device 200 may be configured to communicate wired or wirelessly with the enterprise terminal 100.

Additionally, the device 200 can communicate wired or wirelessly with a website including a web page. At this time, the web page may include a media company's web page, each company's web page, SNS, etc., and SNS may include a blog, This may include cafes, Instagram, Facebook, Twitter, YouTube, etc. The device 200 can access a website and obtain information.

The device 200 obtains the company information of the company from the company terminal 100, obtains product information of the product from the company terminal 100, and displays a web page related to the company online based on the company information and product information. Collects and analyzes the text of the collected web pages using natural language processing techniques to classify the text of the collected web pages into useful and unused phrases, and stores content other than text from the collected web pages as online content. , corporate information and product information can be entered into a generative artificial intelligence model to obtain a draft article, and additional information including usage phrases and online content can be used to correct the draft article to create a corporate promotional article.

Additionally, the device 200 may provide the generated corporate promotional article to the company's terminal 100.

Additionally, the device 200 may transmit a corporate promotional article to a media company based on the media company's web page.

In the present invention, artificial intelligence (AI) refers to technology that imitates human learning ability, reasoning ability, perception ability, etc. and implements this with a computer, and includes concepts such as machine learning and symbolic logic. may include. Machine Learning (ML) is an algorithmic technology that classifies or learns the characteristics of input data on its own. Artificial intelligence technology is a machine learning algorithm that analyzes input data, learns the results of the analysis, and makes judgments or predictions based on the results of the learning. Additionally, technologies that mimic the functions of the human brain, such as cognition and judgment, using machine learning algorithms can also be understood as the category of artificial intelligence. For example, technical fields such as verbal understanding, visual understanding, reasoning/prediction, knowledge representation, and motion control may be included.

Machine learning can refer to the process of training a neural network model using experience processing data. Machine learning can mean that computer software improves its own data processing capabilities. A neural network model is built by modeling the correlation between data, and the correlation can be expressed by a plurality of parameters. A neural network model extracts and analyzes features from given data to derive correlations between data. Repeating this process to optimize the parameters of the neural network model can be called machine learning. For example, a neural network model can learn the mapping (correlation) between input and output for data given as input-output pairs. Alternatively, a neural network model may learn the relationships by deriving regularities between given data even when only input data is given.

An artificial intelligence learning model or neural network model may be designed to implement the human brain structure on a computer, and may include a plurality of network nodes with weights that simulate neurons of a human neural network. A plurality of network nodes may have a connection relationship with each other by simulating the synaptic activity of neurons in which neurons exchange signals through synapses. In an artificial intelligence learning model, multiple network nodes are located in layers of different depths and can exchange data according to convolutional connection relationships. The artificial intelligence learning model may be, for example, an artificial neural network model (Artificial Neural Network), a convolution neural network (CNN) model, etc. As an example, an artificial intelligence learning model may be machine-learned according to methods such as supervised learning, unsupervised learning, and reinforcement learning. Machine learning algorithms for performing machine learning include Decision Tree, Bayesian Network, Support Vector Machine, Artificial Neural Network, and Ada-boost. , Perceptron, Genetic Programming, Clustering, etc. can be used.

Among them, CNN is a type of multilayer perceptrons designed to use minimal preprocessing. CNN consists of one or several convolution layers and general artificial neural network layers on top of them, and additionally utilizes weights and pooling layers. Thanks to this structure, CNN can fully utilize input data with a two-dimensional structure. Compared to other deep learning structures, CNN shows good performance in both video and audio fields. CNNs can also be trained via standard back propagation. CNNs have the advantage of being easier to train and using fewer parameters than other feedforward artificial neural network techniques.

Convolutional networks are neural networks that contain sets of nodes with bound parameters. The increasing size of available training data and the availability of computational power, combined with algorithmic advances such as piecewise linear unit and dropout training, have led to significant improvements in many computer vision tasks. For extremely large data sets, such as those available for many tasks today, overfitting is not critical, and increasing the size of the network improves test accuracy. Optimal use of computing resources becomes a limiting factor. For this purpose, distributed, scalable implementations of deep neural networks can be used.

Figure 2 is a flow chart to explain the process of generating a corporate promotional article using a generative artificial intelligence model according to an embodiment.

Referring to FIG. 2, first, in step S201, the device 200 may obtain company information of the company from the company's terminal 100. Here, the company's corporate information may include the company's name, the company's industry, the company's history, the company's address, and the company's contact information.

Specifically, the device 200 may obtain company information including the company's name, company's industry, company's history, company's address, and company's contact information from the company's terminal 100.

In step S202, the device 200 may obtain product information about a product provided by the company from the company's terminal 100. Here, the product may include both tangible products and intangible services, and the product information may include the product name, product category, product characteristics, and product image corresponding to the product provided by the company. there is.

Specifically, the device 200 may obtain product information of the product provided by the company, including the name of the product, the category of the product, the characteristics of the product, and the image of the product, from the company's terminal 100.

Meanwhile, the device 200 may obtain product information of a product corresponding to a plurality of products provided by the company from the company's terminal 100, and may obtain product information of a product corresponding to one product that the company wishes to promote. You can also obtain it.

In step S203, the device 200 may collect online web pages related to the company based on company information and product information. Here, online may include a website, and the website may communicate with the device 200 wired or wirelessly.

Specifically, the device 200 may collect web pages related to a company online based on the company information and product information obtained from the company terminal 100. At this time, the device 200 generates a search term including company information and product information based on the company information and product information obtained from the company terminal 100 in order to collect web pages related to the company online. Based on the generated search term, you can search online and obtain web pages related to the company.

In step S204, the device 200 may analyze the text of the collected web page through natural language processing techniques and divide the text of the collected web page into useful phrases and unused phrases.

Here, Natural Language Processing (NLP) is a technique that allows computers to understand, generate, and manipulate human language, and natural language processing can be performed by an artificial intelligence model. At this time, the artificial intelligence model that performs natural language processing can be learned through machine learning to perform the target task through training on learning data. For example, an artificial intelligence model that performs natural language processing can be learned by sentiment analysis training data consisting of sentences and the emotions (positive, negative, or neutral) of the sentences, and natural language processing learned by sentiment analysis training data. The artificial intelligence model that performs may be an artificial intelligence model that returns the emotion of the sentence when a sentence is input. Meanwhile, the artificial intelligence model that performs natural language processing is not specifically written in the text, but can be implemented in the same form as the commonly used artificial intelligence model that performs natural language processing. Meanwhile, an artificial intelligence model that performs natural language processing may be provided in the device 200 or may communicate with the device 200 wired or wirelessly as a separate artificial intelligence model.

That is, the device 200 can recognize text included in a web page based on the collected web pages, and apply the recognized text to an artificial intelligence model that performs natural language processing, through natural language processing techniques. It can be analyzed. Specifically, the device 200 can perform a process of preprocessing the recognized text, the device 200 can classify the recognized text into sentences, and divide the classified sentences into word units again. , during this process, a preprocessing process can be performed to remove special characters or unnecessary spaces. Additionally, the device 200 may analyze the preprocessed text using natural language processing techniques. At this time, natural language processing techniques can predict the positive or negative meaning of a word by understanding sentence structure, vocabulary meaning, and context. In addition, the device 200 can classify words extracted through natural language analysis into positive words and negative words. Positive words are words with a positive meaning, and are words that indicate the advantages of the product or the effect of the product. Negative words are words with a negative connotation and may include words that indicate a product's shortcomings or areas that require improvement.

Additionally, the device 200 may distinguish text from a web page analyzed through natural language processing techniques into useful phrases and unused phrases. Specifically, the device 200 determines that sentences containing positive words, which are words with a positive meaning, can be used in corporate promotional articles, and can classify sentences containing positive words into phrases that are used, and sentences containing positive words with negative meanings. It is determined that sentences containing negative words, which are existing words, cannot be used in corporate promotional articles, so sentences containing negative words can be classified as unused phrases. Meanwhile, although the device 200 does not write in the text, it can also classify the text into useful phrases and unused phrases using commonly used natural language processing techniques.

In step S205, the device 200 may save content other than text from the collected web pages as online content. Here, online content may be photos and videos excluding text.

Specifically, the device 200 can recognize text included in the collected web pages, and the device 200 recognizes content other than text among the content included in the collected web pages, that is, photo content or video content as online content. It can be stored in a database provided in the device 200. At this time, in the process of storing the online content in the database, the device 200 may match and store the information of the online content and the web page that contained the online content. At this time, the information on the web page may include the link of the web page, the site where the web page is posted, the number of views of the web page, and the posting date of the web page.

In step S206, the device 200 may obtain a draft article by inputting company information and product information into a generative artificial intelligence model. Here, the generative artificial intelligence model is an artificial intelligence model that generates various types of data such as text, image, and voice for a given input. The generative artificial intelligence model is based on deep learning and natural language processing technology, and is a language generation model. and image generation model. Additionally, the generative artificial intelligence model may be provided in the device 200 or may communicate with the device 200 wired or wirelessly as a separate artificial intelligence model. Additionally, the generative artificial intelligence model may be implemented in the same form as a commonly used generative artificial intelligence model, and the generative artificial intelligence model may be a model of the same type as the ChatGPT model.

Specifically, the device 200 may obtain a draft article by inputting company information acquired from the company's terminal 100 and product information obtained from the company's terminal 100 into a generative artificial intelligence model.

For example, the device 200 can input “company information” AND “product information” AND “create a draft article using that information” into the generative artificial intelligence model, and the device 200 can input “company information” AND “create a draft article using that information.” You can obtain a draft article printed through a generative artificial intelligence model by entering “information” AND “product information” AND “create a draft article using the information.”

In step S207, the device 200 may correct the draft article using additional information including useful phrases and online content to create a corporate promotional article.

Specifically, the device 200 can correct a draft article obtained by inputting company information and product information into a generative artificial intelligence model using additional information including usage phrases and online content, and as a result of the correction, a corporate promotional article. can be created.

At this time, the process of correcting the draft article using additional information will be referred to Figure 4.

Additionally, the device 200 may provide a corporate promotional article generated through the above process to the corporate terminal 100.

Additionally, the device 200 may transmit a corporate promotional article to a web page of a media company that communicates with the device 200 wired or wirelessly.

Through this, the device 200 can quickly and efficiently generate corporate promotional articles by inputting information about the company and products provided by the company obtained from the company's terminal 100 into a generative artificial intelligence model.

Meanwhile, the device 200 can process product information to obtain a more efficient article draft through a generative artificial intelligence model.

Figure 3 is a flow chart to explain the process of processing product information by inputting it into a generative artificial intelligence model according to an embodiment.

Referring to FIG. 3, first, in step S301, the device 200 may generate a search term based on product information.

Specifically, the device 200 may generate a search term based on product information including the product name, product category, and product characteristics obtained from the company's terminal 100.

In step S302, the device 200 may search the generated search term on a website to obtain posted data within a critical period. Here, the website can communicate with the device 200 wired and wirelessly, and may include a web page. At this time, the web page may include a media company's web page, each company's web page, SNS, etc., and SNS. May include blogs, cafes, Instagram, Facebook, Twitter, YouTube, etc. Additionally, the critical period is a preset period and may vary depending on the embodiment.

Specifically, the device 200 can generate a search term based on the product information of the product obtained from the company's terminal 100, and send it to a website that communicates with the device 200 wired or wirelessly based on the generated search term. You can search and check the printed published data, and among the published data, you can obtain published data published within a preset threshold period.

In step S303, the device 200 may obtain information on products included more than the standard number of times based on the posted data. Here, the reference number is a preset number and may vary depending on the embodiment.

Specifically, the device 200 can search a website for a search term generated through product information of a product to obtain published data within a critical period, and based on the obtained published data, more than a preset standard number of times. You can obtain information about the included products.

That is, the device 200 can check words included more than the standard number of times, that is, keywords included more than the standard number of times, and obtain information about products included more than the standard number of times, based on the posted data within the critical period. can do. At this time, the keyword may be the name of the product included in the product information of the product obtained from the company's terminal 100, or may be a characteristic of the product included in the product information of the product. In addition, the device 200 checks keywords that are included more than the standard number of times based on the posted data, and if the keyword is confirmed to be the name of a product included in the product information, the company sells the product as its main product. You can also judge by the product that is available.

Figure 4 is a flow chart to explain the process of correcting a draft article and creating a corporate promotional article using additional information according to an embodiment.

Referring to FIG. 4, first, in step S401, the device 200 may extract a phrase related to a specific product mentioned in the draft article from among the phrases used as a related phrase. Here, the specific product may be a product confirmed based on product information obtained from the company's terminal 100, or may be a product mainly sold by the company identified through FIG. 3.

Specifically, the device 200 can collect web pages related to a company online based on company information and product information, and the device 200 can use natural language processing to convert the text of the web page based on the collected web pages. You can check the phrases used by analyzing them. In addition, the device 200 can check the draft article obtained through the generative artificial intelligence model and check the specific product mentioned in the article draft. In this case, the specific product is the product obtained from the company's terminal 100. It may be a product confirmed based on product information, or it may be a product mainly sold by the company identified through Figure 3. That is, the device 200 is based on the product information of the product acquired from the company's terminal 100 and the product information acquired through the process of FIG. 3, and the article mentioned in the draft obtained through the generative artificial intelligence model. You can check the product, and you can check the product as a specific product. In addition, when the specific product mentioned in the draft article is confirmed, the device 200 can check information about the specific product, including the name of the specific product, the category of the specific product, and the characteristics of the specific product included in the draft article. Based on information on a specific product, phrases related to a specific product among the phrases used can be identified and extracted as related phrases.

In step S402, the device 200 may search for a specific product online and set a keyword with a high frequency of appearance among keywords excluding the specific product as a trend keyword in the results output.

Specifically, when the device 200 confirms a specific product mentioned in the draft article, the device 200 can obtain a result by searching the name of the specific product on a website that communicates wired or wirelessly with the device 200. Based on the results output by searching for a specific product on a website, the keyword excluding the specific product, that is, the keyword with the highest frequency of appearance among keywords excluding the name of the specific product, can be set as a trend keyword. At this time, the reason for excluding specific products among the keywords in the process of setting them as trend keywords is that, for example, when searching for a specific product, “First Product,” the results output naturally contain a lot of “First Product” among the keywords. “First product” can be excluded.

In step S403, the device 200 may obtain a revised article by inputting the draft article, related phrases, and trend keywords into a generative artificial intelligence model so that the draft article is revised using the related phrases and trend keywords.

Specifically, the device 200 uses related phrases, which are phrases related to a specific product among the phrases used, and trend keywords, which are keywords set through the output of an online search for a specific product, so that the draft article is modified. You can obtain article revisions by entering phrases and trend keywords into a generative artificial intelligence model.

For example, the device 200 may input “related phrases” AND “trend keywords” AND “draft article” AND “edit the draft article through the information” into the generative artificial intelligence model, and the device (200) can obtain article revisions output through a generative artificial intelligence model by entering “related phrases” AND “trending keywords” AND “article draft” AND “edit the article draft using the relevant information.” there is.

In step S404, the device 200 may extract content related to a specific product from online content.

Specifically, the device 200 can store content other than text from web pages related to companies collected online as online content, and extract content related to a specific product from the stored online content.

At this time, the device 200 is based on the image of the specific product acquired from the company's terminal 100 in order to extract content related to the specific product from the online content, and if there is content that matches the image of the specific product with a higher than target rate. In this case, the content can be extracted as content related to a specific product.

Additionally, the device 200 can check text included in online content, and if there is content containing the name of a specific product based on the confirmed text, the device 200 can extract the content as content related to the specific product. . At this time, when the online content is a photo, the device 200 can recognize the text included in the photo through an image recognition artificial neural network and OCR to check the text included in the online content, and can recognize a specific product through the recognized text. You can check whether the name is included. Here, a device that performs an image recognition artificial neural network and OCR may be provided in the device 200 or may be a separate device that can communicate with the device 200 wired or wirelessly. Additionally, when the online content is a video, the device 200 can acquire voice data included in the video to check the text included in the online content, and convert the acquired voice data into text through the STT module. And you can check whether the name of a specific product is included through the converted text. Here, STT stands for Speech-to-Text, and the STT module is an offline learning stage that creates a recognition network model including an acoustic model, vocabulary/pronunciation dictionary, and language model from voice data, and an online exploration stage that recognizes the voice uttered by the user. The user's voice can be converted into second text. The STT module is provided in the device 200 or is a separate device and can communicate with the device 200 wired or wirelessly. Although not written in the text, it can convert voice into text through the tasks normally performed by the STT module. .

Meanwhile, in addition, the device 200 can extract content related to a specific product from stored online content using a commonly used method.

In step S405, the device 200 may check the original web page containing the extracted content and give priority to the extracted content according to the number of views of the original web page and the rating of the site on which the original web page was posted. .

Specifically, the device 200 can extract content related to a specific product from online content and obtain information about the original web page that contained the extracted content through a database. Additionally, the device 200 can check the number of views of the original web page and the site on which the original web page was posted through information on the original web page, and the rating of the site on which the original web page was posted through the site on which the original web page was posted. can confirm. At this time, a site and a rating corresponding to the site may be matched in advance in the database provided in the device 200. For example, Naver is matched with a rating of 1, Google is matched with a rating of 1, and Nate is matched with a rating of 1. It is matched to level 2, Daum is matched to level 2, ZUM is matched to level 3, Naoul is matched to level 4, and DreamWiz may be matched to level 4. Here, the rating corresponding to the site may be manually set by the device manager who manages the device 200. In addition, the device 200 acquires the average annual visit volume of the site through a website that communicates wired and wirelessly with the device 200, and if the obtained average annual visit volume is greater than a preset first threshold, a rating corresponding to the site is assigned. Set to level 1, and if it is less than the first threshold but more than the preset second threshold, set the level corresponding to the site to level 2. If it is less than the second threshold but more than the preset third threshold, set the level corresponding to the site to level 2. You can set the level to level 3, and if it is less than the third threshold, the level corresponding to the site can be automatically set to level 4. Meanwhile, the range of grades may vary depending on the embodiment.

Additionally, based on the extracted content, the device 200 can determine the number of views of the original web page that included the extracted content and the rating of the site on which the original web page was posted, and the number of views of the original web page that included the extracted content. Priority can be given to extracted content based on the number of views and the rating of the site where the original web page was posted. That is, according to the extracted content, the device 200 can first (quickly) give priority to the extracted content as the number of views of the original web page is higher and the rating of the site on which the original web page is posted is closer to level 1. In addition, the device 200 may grant priority to the extracted content later (slowly) as the number of views of the original web page is lower and the grade of the site on which the original web page is posted is further away from grade 1.

For example, extracting first content and second content as content related to a specific product, the number of views of the original web page on which the first content was posted is 10,000 views, and the number of views on the site on which the original web page on which the first content was posted is 10,000 views. If the rating is 1, the number of views of the original web page on which the second content is posted is 5,000 views, and the site on which the original web page on which the second content is posted has a rating of 2, the device 200 displays the first content The number of views of the original web page on which the second content was posted is higher than the number of views on the original web page on which the second content was posted, and the rating of the site on which the original web page on which the first content was posted is posted. By confirming that the level is closer to level 1 than the level of , the priority of the first content can be given faster than that of the second content.

Meanwhile, the device 200 determines that the number of views of the original web page on which the first content is posted is higher than the number of views of the original web page on which the second content is posted, and that the rating of the site on which the original web page on which the first content is posted is posted is lower than that of the original web page on which the first content is posted. 2 If the original web page on which the content is posted is closer to the level of 1 than the level of the posted site, the priority of the first content may be given faster than that of the second content, but the device 200 may not allow the first content to be posted. The number of views of the original web page on which the second content was posted is lower than the number of views of the original web page on which the second content was posted, and the rating of the site on which the original web page on which the first content was posted is posted is lower than the number of views on the original web page on which the second content was posted. If it is closer to level 1 than the level of , the priority of the first content may be given slower than that of the second content.

That is, if the number of views of the original web page on which the content is posted and the rating of the site on which the original web page on which the content is posted are both excellent, the device 200 quickly gives priority to the content as first, and the device 200 (200) If the view count of the original web page on which the content is posted is superior, and the rating of the site on which the original web page on which the content is posted is inferior, the priority of that content is given to the second fastest, and the device ( 200), if the number of views of the original web page on which the content is posted is inferior, and the rating of the site on which the original web page on which the content is posted is excellent, the priority of the content is given to the third fastest, and the device 200 ) may give the content the slowest priority if both the number of views of the original web page on which the content was posted and the rating of the site on which the original web page on which the content was posted are inferior.

In step S406, the device 200 may assign priority to each paragraph according to the number of times a specific product is mentioned in each paragraph of the article revision and the extent to which a phrase affirming the specific product is included.

Specifically, the device 200 classifies the article revisions by paragraph based on the article revisions obtained by inputting the article draft, related phrases, and trend keywords into a generative artificial intelligence model, and specifies a specific product in each classified paragraph. You can check the number of times it is mentioned and the number of times it contains phrases that affirm a specific product. At this time, the device 200 can determine the number of times a specific product is mentioned in each paragraph by checking how many times the name of a specific product is included in the paragraph. Additionally, the device 200 separates the paragraph into sentences and separates them. By analyzing the sentence using natural language processing techniques, you can check whether the sentence affirms a specific product. The device 200 can check the extent to which each paragraph contains a phrase affirming a specific product by checking how many times a phrase affirming a specific product is included in the paragraph.

In addition, the device 200 can generate a related index for a specific product corresponding to each paragraph by summing the number of times a specific product is mentioned and the degree to which a phrase affirming the specific product is included in each paragraph. , Based on the related index of the specific product corresponding to each created paragraph, the more related indexes, the more priority the paragraph can be given first (faster), and the smaller the related index, the later (slower) the priority can be given to the paragraph. can do.

In step S407, the device 200 may arrange the extracted content according to the priority of each paragraph and the priority of the extracted content.

Specifically, the device 200 can check the priority of each paragraph and place the content with the highest priority in the paragraph with the highest priority, and place the content with the highest priority in the paragraph with the next highest priority. You can deploy content quickly. That is, the device 200 can check the priority of the paragraph and the priority of the content and match and arrange the paragraph and content according to the priority.

Due to this, the device 200 can effectively generate corporate promotional articles by inserting online content including photos and videos as well as text, and place high-priority content in high-priority paragraphs to create corporate promotional articles. Users who encounter it can read the article efficiently.

Figure 5 is a flowchart for explaining the process of setting the size of extracted content according to an embodiment.

Referring to FIG. 5 , first, in step S501, the device 200 may confirm the release date of a specific product and create a confirmation period, which is the period between the release date of the specific product and the current time.

Specifically, the device 200 can check the release date of a specific product. At this time, the device 200 can access a website that communicates with the device 200 wired or wirelessly to check the release date of a specific product, and search the website based on a search term generated by combining the specific product and the release date. This allows you to obtain the release date of a specific product. Additionally, the device 200 may check the release date of a specific product through product information on the product obtained from the company's terminal 100. At this time, the product information may include the product name, product category, product characteristics, and product release date.

Additionally, when the release date of a specific product is confirmed, the device 200 may check the period between the release date of the confirmed specific product and the current time to create a confirmation period. At this time, the operation to check the current time point has not been specifically written, but the device 200 can check the current time point through a commonly used method. Meanwhile, a specific product may be a product that has already been released, or a product scheduled to be released that has not yet been released.

For example, if the release date of a specific product is August 10, 2023, and the current time is August 20, 2023, the device 200 detects the release date of the specific product, which is August 10, 2023, and the current time, August 8, 2023. You can create a confirmation period of 10 days, which is a period of 20 days per month.

In step S502, the device 200 may determine whether the confirmation period is longer than the first reference period. Here, the first reference period is a preset period and may vary depending on the embodiment.

If it is determined in step S502 that the confirmation period is longer than the first reference period, in step S503, the device 200 may set the size of the extracted content to the first size. Here, the first size may be a preset size and may vary depending on the embodiment.

Specifically, the device 200 may generate a confirmation period by checking the period between the launch date of a specific product and the current point, and compare the generated confirmation period with a preset first reference period, so that the confirmation period is the first reference period. If it is confirmed that it is longer than that, it is determined that the time of release of a specific product or the time of release of a specific product is far away, and the size of the extracted content can be set to a preset first size. At this time, the extracted content may be content related to a specific product among online content, and the size of the extracted content may be information about the width and height of the content to be included in the article.

For example, the release date of a specific product is May 20, 2023, the current point is August 20, 2023, the first reference period is one month, and the content related to the specific product, that is, the extracted content, is the first content. If the first size is 5cm By confirming that the period of 3 months is longer than the first standard period of 1 month, the size of the first content can be set to the first size of 5cmX5cm. That is, the device 200 may set the size of the first content included in the corporate promotional article to the first size of 5cmX5cm.

If it is determined in step S502 that the confirmation period is shorter than or equal to the first reference period, in step S504, the device 200 may determine whether the confirmation period is longer than the second reference period. Here, the second reference period is a period set shorter than the first reference period, and may be a preset period, and may vary depending on the embodiment.

If it is determined in step S504 that the confirmation period is longer than the second reference period, in step S505, the device 200 may set the size of the extracted content to the second size. Here, the second size is a size set larger than the first size, may be a preset size, and may vary depending on the embodiment.

Specifically, the device 200 may generate a confirmation period by checking the period between the launch date of a specific product and the current point, and compare the generated confirmation period with a preset first reference period, so that the confirmation period is the first reference period. If confirmed to be shorter or equal, the generated confirmation period can be compared to a second preset reference period. In addition, when it is determined that the confirmation period is longer than the second reference period, the device 200 determines that the time of launching a specific product or is close to the time of launching a specific product, and sets the size of the extracted content to a preset second size. It can be set to . At this time, the extracted content may be content related to a specific product among online content, and the size of the extracted content may be information about the width and height of the content to be included in the article.

For example, the launch date of a specific product is September 5, 2023, the current point is August 20, 2023, the first reference period is one month, the second reference period is 10 days, and the content related to the specific product That is, if the extracted content is the first content, the first size is 5cm 16 days, which is a period of days, can be created as a confirmation period, and by confirming that the confirmation period of 16 days is shorter than the first standard period of one month and longer than the second standard period of 10 days, the size of the first content is determined. It can be set to 2 sizes: 10cmX10cm. That is, the device 200 may set the size of the first content included in the corporate promotional article to the second size of 10cmX10cm.

If it is determined in step S504 that the confirmation period is shorter than or equal to the second reference period, in step S506, the device 200 generates a weight in inverse proportion to the confirmation period, applies the weight to the second size, and determines the third size. You can create and set the size of the extracted content to a third size. At this time, the third size may be larger than or equal to the second size. That is, the weight can be created greater than or equal to 1.

Specifically, the device 200 may generate a confirmation period by checking the period between the launch date of a specific product and the current point, and compare the generated confirmation period with a preset first reference period, so that the confirmation period is the first reference period. If confirmed to be shorter or equal, the generated confirmation period can be compared to a second preset reference period. In addition, when the device 200 determines that the confirmation period is shorter than or equal to the second reference period, the device 200 determines that the time of launching a specific product or the time of launching a specific product is very close, and generates a weight in inverse proportion to the confirmation period. , a third size can be created by applying a weight to the preset second size, and the size of the extracted content can be set to the third size. At this time, the extracted content may be content related to a specific product among online content, and the size of the extracted content may be information about the width and height of the content to be included in the article. Additionally, the device 200 multiplies the width corresponding to the second size by the weight to apply a weight to the second size to generate the third size, and multiplies the height corresponding to the second size by the weight to generate the third size. can do.

For example, the release date of a specific product is August 15, 2023, the current point is August 20, 2023, the first reference period is one month, the second reference period is 10 days, and content related to the specific product That is, if the extracted content is the first content, the first size is 5cm 5 days, which is a period of days, can be created as a confirmation period, and by confirming that the confirmation period of 5 days is shorter than the first standard period of one month and shorter than the second standard period of 10 days, the weight is inversely proportional to the confirmation period. 2, the third size can be created as 20cmX20cm by applying the weight of 2 to the second size, 10cmX10cm, and the size of the first content can be set to the third size, 20cmX20cm. That is, the device 200 may set the size of the first content included in the corporate promotional article to the third size of 20cmX20cm.

In addition, for example, the release date of a specific product is August 30, 2023, the current point is August 20, 2023, the first reference period is one month, the second reference period is 10 days, and the specific product and If the related content, that is, the extracted content is the first content, the first size is 5cm 10 days, which is a period of 20 days per month, can be created as a confirmation period. By confirming that the confirmation period of 10 days is shorter than the first standard period of one month and the same as the second standard period of 10 days, the weight is inversely proportional to the confirmation period. can be created as 1, and the third size can be created as 10cmX10cm by applying a weight of 2 to the second size, 10cmX10cm, and the size of the first content can be set to the third size, 10cmX10cm. That is, the device 200 may set the size of the first content included in the corporate promotional article to the third size, 10cmX10cm.

In addition, for example, the release date of a specific product is August 18, 2023, the current point is August 20, 2023, the first reference period is one month, the second reference period is 10 days, and the specific product and If the related content, that is, the extracted content is the first content, the first size is 5cm 2 days, which is a period of 20 days per month, can be created as a confirmation period, and by confirming that the confirmation period of 2 days is shorter than the first standard period of one month and shorter than the second standard period of 10 days, the confirmation period is inversely proportional to the confirmation period. The weight can be created as 5, the third size can be created as 50cmX50cm by applying the weight of 5 to the second size, 10cmX10cm, and the size of the first content can be set to the third size, 50cmX50cm. That is, the device 200 may set the size of the first content included in the corporate promotional article to the third size of 50cmX50cm.

That is, the device 200 can check the release date of a specific product and set the size of the content related to the specific product to be larger as the release date of the specific product is closer to the current time, and as the release date of the specific product is further away from the current time, the size of the content related to the specific product can be set larger. The size of content related to the content can be set small, allowing specific products to be promoted efficiently through corporate promotional articles.

Figure 6 is a flow chart to explain the process of supplementing a corporate promotional article according to an embodiment.

Referring to FIG. 6, first, in step S601, the device 200 may obtain an existing promotional article of the company. Here, the company's existing promotional article is an article that has been previously distributed in relation to the company. The company's existing promotional article may include information about the company, and the distributor, or author, may be a company.

Specifically, the device 200 may generate a search term based on company information obtained from the corporate terminal 100, and based on the generated search term, search on a website that communicates with the device 200 wired or wirelessly. You can obtain the company's existing promotional articles.

In addition, the device 200 can check the website operated by the company based on the company information obtained from the company's terminal 100, and can also obtain the company's existing promotional articles through the confirmed website. . At this time, the company information may include the name of the company, the company's industry, the company's history, the company's address, the company's contact information, and a link to a website operated by the company, and the device 200 may be connected to the website operated by the company. You can communicate with the site wired or wirelessly.

Additionally, the device 200 may obtain the company's existing promotional articles from the company's terminal 100.

In step S602, the device 200 may select an article written within a target period among the company's existing promotional articles as a reference article. Here, the target period is a preset period and may vary depending on the embodiment.

Specifically, the device 200 may check the date on which the company's existing promotional articles were written, extract articles written within a preset target period from among the company's existing promotional articles, and select them as reference articles.

In step S603, the device 200 may check the phrases used in the first part based on the reference article, and select the most used phrase among the identified phrases as the first phrase. At this time, the first part may be the front part of the reference article, such as the first paragraph.

Specifically, based on the reference article, the device 200 can check the phrases used in the first part of the reference article, such as the first part, that is, the first paragraph, and select the most used phrase among the identified phrases as the first phrase. You can select.

For example, based on the reference article, the device 200 changes the phrases used in the first part to 'The first company succeeded in task A last month' and 'The first company released the first report.' , 'The first company is a company located in the first region and performing the first business.' You can check, etc., and you can check the most used phrase among the phrases used in the first part. In addition, as a result of confirmation, the device 200 selects 'The first company succeeded in task A last month' as the most frequently used phrase, and then selects 'The first company succeeded in task A last month' as the first phrase. It can be selected.

In step S604, the device 200 may check the phrases used in the second part based on the reference article, and select the most used phrase among the identified phrases as the second phrase. At this time, the second part may be the latter part of the reference article, such as the last paragraph.

Specifically, based on the reference article, the device 200 can check the phrases used in the second part, that is, the latter part of the reference article, such as the last paragraph, and select the most used phrase among the identified phrases as the second phrase. You can select.

For example, based on the reference article, the device 200 changes the phrases used in the second part to 'Meanwhile, the first company is in the process of working on B', 'The first company will produce the first product within the second half of the year.' It is expected to be launched.’, ‘The first company is providing the first service to commemorate the launch of the second product.’ You can check, and you can check the most used phrase among the phrases used in the second part. In addition, if the most frequently used phrase is confirmed as 'The first company is expected to release the first product within the second half of the year,' the device 200 determines 'The first company is expected to release the first product within the second half of the year.' ‘It is expected that this will be done.’ can be selected as the second phrase.

In step S605, the device 200 may insert a first phrase into the first part of the corporate promotional article and insert a second phrase into the second part of the corporate promotional article.

Specifically, the device 200 may insert the first phrase obtained through the reference article into the first part of the corporate promotional article written through the generative artificial intelligence model, and the device 200 may insert the first phrase obtained through the reference article into the corporate promotional article written through the generative artificial intelligence model. In the second part, you can insert a second phrase obtained through reference articles.

Because of this, the device 200 can check recent corporate issues through reference articles and apply the issues to corporate promotional articles. Additionally, the device 200 may generate a more suitable corporate promotional article by applying the first phrase and the second phrase obtained through the reference article to the corporate promotional article written through the generative artificial intelligence model.

Figure 7 is a flowchart for explaining the process of adding a link related to a specific product according to an embodiment.

Referring to FIG. 7 , first, in step S701, the device 200 may identify an influencer who promoted a specific product through SNS and obtain the influencer's SNS account.

Specifically, the device 200 can communicate wired or wirelessly with a website including SNS, and the device 200 can access SNS to search for a specific product through the name of the specific product, and as a result of the search, the specific product and You can obtain related posts. Additionally, the device 200 can check the author who uploaded the post based on the post related to the specific product obtained, and check the influencer among the confirmed authors. At this time, the device 200 checks the number of times the post has been viewed, the number of times the post has been shared, and the number of followers who liked the post to determine whether the author who uploaded the post is an influencer. If the number of views, the number of times the post was shared, and the number of followers who liked the post are higher than the preset target number, it can be confirmed that the author who uploaded the post is an influencer, and the device 200 If the author who uploaded the post is included in the influencer database, it can be confirmed that the author is an influencer. At this time, the influencer database may be a database in which influencer information is stored for each influencer, and the influencer database may be provided in the device 200 or, if it is a separate device, communicate with the device 200 wired or wirelessly. You can.

Additionally, when the device 200 identifies an influencer among authors who uploaded posts related to a specific product to SNS, it can obtain the SNS account of the influencer.

In step S702, the device 200 may obtain a link containing a promotional post for a specific product as the influencer's product promotion link based on the influencer's SNS account.

Specifically, the device 200 can check posts related to a specific product, that is, promotional posts for a specific product, among posts uploaded by the influencer, based on the influencer's SNS account, and the confirmed promotional posts for the specific product. Based on this, you can obtain the link to the page where the promotional post for a specific product is posted, that is, the influencer's product promotion link.

In step S703, the device 200 may check a purchase link to purchase a specific product.

Specifically, the device 200 searches the website for the name of a specific product, checks the link corresponding to the sales site among the links on the printed web page, and identifies the identified link as a purchase link for purchasing the specific product. You can. At this time, the device 200 may store links to a plurality of sales sites in a database to check the links corresponding to the sales sites, and search the website for the links stored in the database and the name of the specific product to display the output web page. By comparing links, the name of a specific product can be searched on a website, and among the links on the printed web page, the link that has a higher match rate with the link stored in the database than a preset standard rate can be confirmed as the link corresponding to the sales site.

In addition, the device 200 can check the company's contact information and the link to the website operated by the company based on the company information acquired from the company's terminal 100, and the confirmed company's contact information and link to the website operated by the company. At least one of the links on the website can be created as a purchase link to purchase a specific product.

In step S704, the device 200 may add the influencer's product promotion link and purchase link to the company promotional article.

Specifically, the device 200 adds the influencer's product promotion link, which is a link containing a promotional post for a specific product, and a purchase link, which is a link to purchase a specific product, to the corporate promotional article among the posts uploaded by the influencer. can do.

As a result, when users who have read corporate promotional articles become interested in a specific product, they can easily obtain information about the product and easily purchase the product through the influencer's product promotion link and purchase link.

Meanwhile, the device 200 can manage the company's information by deleting or storing the company's information based on the time when the company's promotional article is transmitted to the company's terminal 100.

Figure 8 is a flow chart to explain the process of managing company information according to an embodiment.

Referring to FIG. 8, first, in step S801, if it is confirmed that a corporate promotional article has been sent at a first time, the device 200 may store the company's information in the first company database. Here, the company information may include company information obtained from the company terminal 100 and product information of products provided by the company obtained from the company terminal 100.

Specifically, the device 200 is provided with a plurality of corporate databases or is connected to a plurality of corporate databases, and when it is confirmed that a corporate promotional article has been transmitted to the corporate terminal 100 at the first time, the plurality of corporate databases Company information can be stored in one of the first company databases. At this time, if the device 200 generates a corporate promotional article in step S207, the device 200 may transmit the corporate promotional article to the company's terminal 100 after step S207.

In step S802, if it is confirmed that the company has confirmed the company promotional article, the device 200 may delete the company information stored in the first company database and store the company information in the second company database.

Specifically, when the device 200 receives a message from the company's terminal 100 that the company has confirmed the corporate promotional article, it can confirm that the company has confirmed the corporate promotional article. In addition, when it is confirmed that the company has confirmed the corporate promotional article, the device 200 deletes the company information stored in the first company database, which is one of the plurality of company databases, and the second company database, which is another one of the plurality of company databases. 2 Company information can be stored in the company database.

In step S803, the device 200 may set the storage period of information as the first storage period according to the corporate level of the company. To this end, the device 200 may store the history of creating corporate promotional articles in response to the company in a database provided in the device 200, and based on the history stored in the database, corporate promotional articles may be created in response to the company. You can check the number of times it has been done. The device 200 may set the company's corporate level higher as the number of corporate promotional articles is generated, and may set the company's corporate rating lower as the number of corporate promotional articles is generated. Additionally, the device 200 may check the set company and set the first storage period to be longer as the company's corporate level is higher.

For example, device 200 may set the first storage period to 2 years if the enterprise's enterprise rating is determined to be Excellent, which is the highest, and set the first storage period to 1 year if the enterprise's enterprise rating is determined to be Excellent. It can be set to .

In step S804, if it is confirmed that no additional corporate promotional articles have been sent to the company within the first storage period from the first time, the device 200 may delete the company information stored in the second company database.

For example, if the first storage period is set to 1 year, the device 200 can check whether additional corporate promotional articles have been sent to the company until 1 year has passed from the first time point, and If it is confirmed that no additional corporate promotional articles have been sent to the company by the end of one year, the company's information stored in the second company database can be deleted.

In step S805, if it is confirmed that an additional corporate promotional article has been sent to the company at a second time, the device 200 may set a second storage period by applying a second weight to the first storage period. Here, the second weight may be set to a value less than 1.

That is, if it is confirmed that an additional corporate promotional article has been sent to the company within the first storage period from the first time point, and if it is confirmed that an additional corporate promotional article has been sent to the company at a second time point, the device 200 stores the first storage period. A second storage period can be set by applying a second weight to .

For example, if the first storage period is 1 year and the second weight is 0.5, the device 200 may apply the second weight to the first storage period and set the second storage period to 6 months.

In step S806, if it is confirmed that no additional corporate promotional articles have been sent to the company within the second storage period from the second time, the device 200 may delete the company information stored in the second company database.

After step S805, if it is confirmed that additional corporate promotional articles have been sent to the company within the second storage period from the second time point, step S805 is performed again, and the device 200 sends additional corporate promotional articles to the company at the third time point. If it is confirmed that an additional corporate promotional article has not been sent to the company within the third storage period from the third point in time, the third storage period can be set by applying the second weight to the second storage period. 2 You can delete company information stored in the company database.

For this reason, the device 200 deletes the company's information based on the time when the corporate promotional article was last provided to the company, if the time when the corporate promotional article was last provided to the company is before the standard, and provides the company with a corporate promotional article. If the last time provided is later than the standard, by storing the company's information, only the information of the company that continuously creates corporate promotional articles can be stored.

Figure 9 is an exemplary diagram of the configuration of a device according to an embodiment.

Device 200 according to one embodiment includes a processor 210 and memory 220. The processor 210 may include at least one device described above with reference to FIGS. 1 to 8 or may perform at least one method described above with reference to FIGS. 1 to 8 . A person or organization using the device 200 may provide services related to some or all of the methods described above with reference to FIGS. 1 to 8 .

The memory 220 may store information related to the methods described above or store a program in which methods described later are implemented. Memory 220 may be volatile memory or non-volatile memory.

The processor 210 can execute programs and control the device 200. The code of the program executed by the processor 210 may be stored in the memory 220. The device 200 is connected to an external device (eg, a personal computer or a network) through an input/output device (not shown) and can exchange data through wired or wireless communication.

The embodiments described above may be implemented with hardware components, software components, and/or a combination of hardware components and software components. For example, the devices, methods, and components described in the embodiments may include, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, and a field programmable gate (FPGA). It may be implemented using one or more general-purpose or special-purpose computers, such as an array, programmable logic unit (PLU), microprocessor, or any other device capable of executing and responding to instructions. A processing device may execute an operating system (OS) and one or more software applications that run on the operating system. Additionally, a processing device may access, store, manipulate, process, and generate data in response to the execution of software. For ease of understanding, a single processing device may be described as being used; however, those skilled in the art will understand that a processing device includes multiple processing elements and/or multiple types of processing elements. It can be seen that it may include. For example, a processing device may include a plurality of processors or one processor and one controller. Additionally, other processing configurations, such as parallel processors, are possible.

The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded on a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, etc., singly or in combination. Program instructions recorded on the medium may be specially designed and configured for the embodiment or may be known and available to those skilled in the art of computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic media such as floptical disks. -Includes optical media (magneto-optical media) and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, flash memory, etc. Examples of program instructions include machine language code, such as that produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter, etc. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

Software may include a computer program, code, instructions, or a combination of one or more of these, which may configure a processing unit to operate as desired, or may be processed independently or collectively. You can command the device. Software and/or data may be used on any type of machine, component, physical device, virtual equipment, computer storage medium or device to be interpreted by or to provide instructions or data to a processing device. , or may be permanently or temporarily embodied in a transmitted signal wave. Software may be distributed over networked computer systems and stored or executed in a distributed manner. Software and data may be stored on one or more computer-readable recording media.

Although the embodiments have been described with limited drawings as described above, those skilled in the art can apply various technical modifications and variations based on the above. For example, the described techniques are performed in a different order than the described method, and/or components of the described system, structure, device, circuit, etc. are combined or combined in a different form than the described method, or other components are used. Alternatively, appropriate results may be achieved even if substituted or substituted by an equivalent.

Therefore, other implementations, other embodiments, and equivalents of the claims also fall within the scope of the following claims.

Claims (3)

In a method of generating corporate promotional articles using a generative artificial intelligence model performed by a device,
Obtaining corporate information of the company including the name of the company, the industry of the company, the history of the company, the address of the company, and the contact information of the company from the company's terminal;
Obtaining product information of the product including the name of the product provided by the company, the category of the product, the characteristics of the product, and the image of the product from the terminal of the company;
collecting web pages related to the company online based on the company information and the product information;
Analyzing the collected text of the web page using natural language processing techniques to classify the collected text of the web page into useful phrases and unused phrases;
storing content other than the text in the collected web pages as online content;
Obtaining a draft article by inputting the company information and the product information into a generative artificial intelligence model; and
Generating a corporate promotional article by correcting the draft article using additional information including the usage phrase and the online content;
In the step of obtaining a draft article by inputting the company information and the product information into the generative artificial intelligence model, the information input into the generative artificial intelligence model is:
A search term generated based on the product information is searched on a website and is a keyword that is included more than a preset standard number of times among posted materials within a preset threshold period,
The step of generating a corporate promotional article by correcting the draft article using additional information including the usage phrase and the online content,
Extracting phrases related to a specific product mentioned in the draft article from among the utilized phrases as related phrases;
In the results output by searching the specific product online, setting keywords with a frequency of appearance higher than a preset standard frequency among keywords excluding the specific product as trend keywords;
Obtaining a revised article by inputting the draft article, the related phrases, and the trend keywords into a generative artificial intelligence model so that the draft article is revised using the related phrases and the trend keywords;
Extracting content related to the specific product from the online content,
Giving priority to the extracted content according to the number of views of the original web page containing the extracted content and the rating of the site on which the original web page is posted;
assigning priority to each paragraph of the article amendment according to the number of times the specific product is mentioned in each paragraph and the extent to which phrases affirming the specific product are included; and
Arranging the extracted content according to the priority of each paragraph and the priority of the extracted content,
In the step of placing the extracted content,
Further comprising setting the size of the extracted content,
The step of setting the size of the extracted content is,
Confirming the release date of the specific product and creating a confirmation period, which is the period between the release date of the specific product and the current time,
determining whether the confirmation period is longer than a first preset reference period;
If it is confirmed that the confirmation period is longer than the first reference period, setting the size of the extracted content to a first size;
If it is confirmed that the confirmation period is shorter than or equal to the first reference period, determining whether the confirmation period is longer than a preset second reference period;
If it is confirmed that the confirmation period is longer than the second reference period, setting the size of the extracted content to a second size larger than the first size;
If it is confirmed that the confirmation period is shorter than or equal to the second reference period, a weight is generated in inverse proportion to the confirmation period, the generated weight is applied to the second size to set a third size, and the extracted Including setting the size of the content to the third size,
How to create corporate promotional articles using a generative artificial intelligence model. delete delete