KR102243275B1 - Method, device and computer readable storage medium for automatically generating content regarding offline object - Google Patents

Abstract

An apparatus for automatically generating content according to an embodiment, comprising: a content collection module, configured to collect a plurality of online content related to an offline object; A content element production module comprising a plurality of artificial intelligence units (AI units) and configured to generate a plurality of content elements for a plurality of elements of the offline object, wherein the plurality of AI units include a first AI A unit and a second AI unit, wherein the first AI unit is configured to learn a first element related to the offline object and generate a first content element for the first element based on the plurality of online contents, , The second AI unit is configured to learn a second element related to the offline object based on the plurality of online contents and generate a second content element for the second element, and the first element of the offline object and A content element production module, wherein the second element is different; And receiving a plurality of content elements including the first content element and the second content element from the content element production module, combining at least some of the plurality of content elements, and using the combination, the offline object It may include a reproduction content generation module, configured to generate reproduction content related to. Other various embodiments are possible.

Description

Automatic content creation method for offline objects, device and computer-readable storage medium {METHOD, DEVICE AND COMPUTER READABLE STORAGE MEDIUM FOR AUTOMATICALLY GENERATING CONTENT REGARDING OFFLINE OBJECT}

The present disclosure relates to a method, an apparatus, and a computer-readable storage medium for automatically generating content related to an offline object. More specifically, the present disclosure relates to a method, apparatus, and computer-readable storage medium for automatically generating content based on an artificial neural network by utilizing existing content about places, goods, and people.

Unless otherwise stated herein, the content described in this section is not prior art to the claims in this application, and should not be admitted to be prior art for reasons described in this section.

Various online contents are overflowing due to the widespread use of smartphones, the growth of social media networks, and the personalization of contents creation tools, and modern people are constantly searching and contacting numerous online contents every day. In accordance with such changes in the market, an attempt is made to develop an application service that recycles some or all of the previously published online content.

However, it is difficult to get out of the level of simply redistributing online content, and if the image is recycled, there is room for infringement on the copyright of others. In addition, there is a problem that content consumers are not provided with reliable information due to the overflow of online content for the same offline content.

In this way, there is a need to develop a service that collects previously published online content and creates new online content based on it, but reproduces and provides improved quality content including images without the risk of copyright infringement.

An apparatus for automatically generating content according to an embodiment, comprising: a content collection module, configured to collect a plurality of online content related to an offline object; A content element production module comprising a plurality of artificial intelligence units (AI units) and configured to generate a plurality of content elements for a plurality of elements of the offline object, wherein the plurality of AI units include a first AI A unit and a second AI unit, wherein the first AI unit is configured to learn a first element related to the offline object and generate a first content element for the first element based on the plurality of online contents, , The second AI unit is configured to learn a second element related to the offline object based on the plurality of online contents and generate a second content element for the second element, and the first element of the offline object and A content element production module, wherein the second element is different; And receiving a plurality of content elements including the first content element and the second content element from the content element production module, combining at least some of the plurality of content elements, and using the combination, the offline object It may include a reproduction content generation module, configured to generate reproduction content related to.

According to an embodiment, the automatic content generation device further includes an input module for receiving an input regarding the offline object, and the content collection module may be configured to collect a plurality of online contents based on the input. have.

In an embodiment, the content element production module may be configured to determine an element plan corresponding to the offline object based on the input, and to generate the plurality of content elements according to the determined element plan.

In an embodiment, the first AI unit includes a plurality of first sub-AI units, and the plurality of first sub-AI units learn a plurality of first sub-elements of the first element to obtain a plurality of first sub-AI units. A sub content element may be generated, and the first element AI unit may be configured to generate a first content element for the first element by combining the plurality of first sub content elements.

In one embodiment, the first element is an image of the offline object, each of the plurality of first sub-elements is related to a part of the offline object, and the first AI unit It is configured to extract a plurality of images for an offline object, and each of the plurality of first sub-AI units learns a corresponding one of the plurality of first sub-elements from the extracted plurality of images to generate a first sub-content element. And, the first AI unit may be configured to combine and process a plurality of generated first sub-content elements to generate the first content element, which is an image of the offline object.

In an embodiment, the plurality of online content is an image of the offline object, and the reproduction content generation module may be configured to generate the reproduced content that is an image of the offline object.

In an embodiment, the reproduced content generating module may include a verification module that calculates the completeness of the reproduced content and returns a feedback on the result of the operation as an input of the reproduced content generating module.

In one embodiment, the reproduction content generation module may further include a verification module that calculates a similarity between each of the plurality of online contents.

A method for automatically generating content performed under the control of a computing device according to an embodiment, wherein the computing device includes a plurality of artificial intelligence units (AI units), and the method for automatically generating content includes: Collecting a plurality of online content; Generating a plurality of content elements by a plurality of AI units based on the plurality of online contents-Each of the plurality of AI units learns the plurality of online contents with respect to a predetermined element related to the offline object And, based on a result of the learning, a content element for the predetermined element is generated, wherein the predetermined element to be learned by the plurality of AI units is different from each other -; Combining at least some of the plurality of content elements; And generating reproduction content for the offline object based on the result of the combination.

A computer program stored in a computer-readable recording medium to automatically generate content related to an offline object according to an embodiment, wherein the computer program includes a plurality of artificial intelligence units (AI units) when executed. Collecting, by the computing device, a plurality of online contents related to the offline object; An operation of generating a plurality of content elements by a plurality of AI units based on the plurality of online contents-In the generating operation, the computing device causes each of the plurality of AI units to be With respect to the element, the plurality of online contents are learned, and based on the result of the learning, a content element for the predetermined element is generated, where the predetermined elements to be learned by the plurality of AI units are different from each other. -; Combining at least some of the plurality of content elements; And one or more computer-executable instructions for enabling execution of an operation of generating reproduction content for the offline object, based on the result of the combination.

The above brief summary and description of the effects are merely illustrative and are not intended to limit the technical matters intended in the present disclosure. By referring to the following detailed description and the accompanying drawings, in addition to the above-described exemplary embodiments and technical features, additional embodiments and technical features may be understood.

Features of the present disclosure described above and other additional features will be described in detail below with reference to the accompanying drawings. These drawings illustrate only a few embodiments according to the present disclosure, and should not be regarded as limiting the scope of the technical idea of the present disclosure. The technical idea of the present disclosure will be described in more detail and detail using the accompanying drawings.
1 is a block diagram showing a system for generating content related to an offline object according to an embodiment.
FIG. 2 is a diagram illustrating an exemplary form in which an element plan used to generate content about an offline object and information about an element are stored in an element database, according to various embodiments of the present disclosure.
3 is a block diagram illustrating an automatic content generation platform for generating content related to an offline object according to various embodiments of the present disclosure.
4 is a block diagram illustrating a technique of generating an image by the element AI unit of the content element production module 112.
5 conceptually shows a process of generating an image by an element unit of a content element production module.
6 shows an example in which reproduction content for an offline object is provided to a user.
7 is a flow diagram illustrating an exemplary process for generating online content for an offline object, in accordance with at least some embodiments of the present disclosure.
8 illustrates an exemplary computer program product that may be used to create content relating to an offline object, in accordance with at least some embodiments of the present disclosure.

Hereinafter, embodiments and embodiments of the present disclosure will be described in detail with reference to the accompanying drawings so that those of ordinary skill in the art may easily implement the present disclosure. However, the present application may be implemented in various different forms and is not limited to the embodiments and examples described herein.

1 is a block diagram showing an artificial intelligence system for automatically generating content related to an offline object according to an embodiment.

Referring to FIG. 1, the artificial intelligence system 10 according to an embodiment may include an intelligent server 100 and a user device 200.

The intelligent server 100 of an embodiment includes an automatic content generation platform 110, an element database 120, a front end 130, a big data platform 140, or It may include an analysis platform (analytic platform) (150).

The intelligent server 100 according to an embodiment may receive a signal for requesting creation of content related to an offline object from the user device 200 through a communication network and a signal related to the user device 200 (eg, status information). The signal for requesting the creation of content related to the offline object may include the name of the offline object and the nature of the offline object (eg, a tourist destination, a place, a product, a restaurant, a person, etc.).

The content automatic generation platform 110 according to an embodiment may generate content related to an offline object according to the user input and output the generated automatically generated content. The content automatic generation platform 110 may use an element plan corresponding to the characteristics of the offline object when generating content related to an offline object that is a requested target. The element plan may represent a data structure in which elements constituting content and a relationship between the elements are expressed in stages (or hierarchical). Accordingly, the content automatic generation platform 110 may generate and use a hazard plan or use the generated element plan. The element plan and each element included in the element plan may be stored in the element database 120.

The output automatically generated content may be transmitted to the user device 200 through the front end 130. Accordingly, the user device 200 may receive the automatically generated content and provide the received automatically generated content to the user.

The content automatic creation platform 110 according to an embodiment may manage information used in the intelligent server 100. The content automatic creation platform 110 according to an embodiment may collect data on an offline object. The big data platform 140 may store content related to the collected offline object, and information associated with the offline object and user information that has requested the automatic creation of the content. In addition, it is possible to collect and store a variety of information to provide an automatic content creation service. The analysis platform 140 of an embodiment may manage the quality of service (QoS) of the intelligent server 100. For example, the analysis platform 150 may manage components and processing speed (or efficiency) of the intelligent server 100.

The intelligent server 100 according to an embodiment may transmit the reproduced content generated by the content automatic generation platform 110 to the user device 200 or may transmit the reproduced content to the user device 200. According to an embodiment, the user device 200 may display reproduced content on the output device 230 such as a display. According to an embodiment, the user device 200 may output and provide reproduced content to the output device 230 through the app 251. The app 251 may reprocess the reproduced content for provision.

The front end 130 according to an embodiment may receive a user input received from the user device 200. The front end 130 may transmit a response (eg, automatically generated content for “Gwanghwamun”) corresponding to the user input (eg, a content playback request for “Gwanghwamun”).

The user device 200 according to an embodiment may be a terminal device (or electronic device) that can be connected to the Internet, for example, a mobile phone, a smart phone, a personal digital assistant (PDA), a notebook computer, a TV, a white appliance, It may be a wearable device, an HMD, or a smart speaker.

According to the illustrated embodiment, the user device 200 includes a communication interface 210, a processor 220, an output device 230 including a display and a speaker, a microphone, and an input including a touch sensor for detecting a user input to the display. It may include a device 240 or a memory 250. The components listed above may be operatively or electrically connected to each other.

The communication interface 210 of an embodiment may be configured to transmit and receive data by being connected to an external device. The microphone included in the input device 240 according to an embodiment may receive sound (eg, user speech) and convert it into an electrical signal. A speaker included in the output device 230 according to an embodiment may output an electrical signal as sound (eg, voice). The display included in the output device 230 of an embodiment may be configured to display an image or video. The display of an embodiment may also display a graphic user interface (GUI) of an executed app (or application program).

The memory 250 according to an embodiment may store at least one app 251, a software development kit (SDK) 252, and a client module 254. The client module 253 and the SDK 252 may configure a framework (or a solution program) for performing a general function. In addition, the client module 253 or the SDK 252 may configure a framework for processing various user inputs.

The plurality of apps 251 of the memory 250 according to an embodiment may be a program for performing a designated function. According to an embodiment, the plurality of apps 251 may include a first app 251-1 and a second app 251-2. According to an embodiment, each of the plurality of apps 251 may include a plurality of operations for performing a specified function. For example, the plurality of apps may include a travel guide app, a restaurant guide app, an alarm app, a message app, and/or a schedule app. According to an embodiment, the plurality of apps 251 may be executed by the processor 220 to sequentially execute at least some of the plurality of operations.

The processor 220 according to an embodiment may control the overall operation of the user device 200. For example, the processor 220 may be electrically connected to the communication interface 210, the output device 230, and the input device 240 to perform a specified operation.

The processor 220 of an embodiment may also execute a program stored in the memory 250 to perform a designated function. For example, the processor 220 may execute at least one of the client module 253 and the SDK 252 to perform the following operation for processing a user input. The processor 220 may control the operation of the plurality of apps 251 through the SDK 252, for example. The following operation described as the operation of the client module 253 or the SDK 252 may be an operation by the execution of the processor 220.

The client module 253 according to an embodiment may receive a user input. For example, the client module 253 may receive a signal corresponding to a user input detected through the input device 240 (eg, a user utterance requesting content reproduction for'Gwanghwamun' or an input to a touch sensor). I can. The client module 253 may transmit a signal corresponding to the received user input to the intelligent server 100. The client module 253 may transmit state information of the user device 200 to the intelligent server 100 together with the received voice input. The state information may be, for example, a location of the user device 200, a language used, and information on an execution state of an app.

The client module 253 according to an embodiment may receive a result corresponding to a signal indicating a user input transmitted to the intelligent server 100. For example, the client module 253 may receive a result corresponding to the user input calculated by the intelligent server 100. The client module 253 may display the received result on an output device (eg, a display).

According to an embodiment, the client module 253 may receive a signal for requesting information necessary to calculate a result corresponding to a user input from the intelligent server 100. According to an embodiment, the client module 253 may collect necessary information in response to the requested signal and transmit it to the intelligent server 100.

In the content automatic generation system 10 described above, the user device 200 may provide a content service related to an offline object to a user in response to a user input. The user input may include, for example, an input through a physical button, a touch input, or a voice input.

According to an embodiment, the intelligent server 100 (or the content automatic generation platform 110) generates, updates, and/or selects a series of element plans for automatically generating content corresponding to the received user input. I can.

According to one embodiment, the element plan may be generated by an artificial intelligent (AI) system. The artificial intelligence system may be a rule-based system, or a neural network-based system (e.g., a feedforward neural network (FNN)), a recurrent neural network (RNN). )), may be Generative Adversarial Networks (GAN), or may be a combination of the above or other artificial intelligence systems According to an embodiment, the element plan is of a predefined element plan. It may be selected from a set or generated in real time in response to a user request, for example, the artificial intelligence system may select at least one of a plurality of predefined element plans. Let's take a look.

FIG. 2 is a diagram illustrating an exemplary form in which an element plan used to generate content related to an offline object and information about an element are stored in an element database, according to various embodiments of the present disclosure. The element database 120 according to an embodiment may store an element plan corresponding to a characteristic of an offline object and a plurality of elements included in the element plan.

The element plan may represent a data structure in which a plurality of elements constituting a content and a relationship between the plurality of elements are shown in stages (or hierarchical). The plurality of elements may include a first element and a second element. The first element may include elements 1-1 and 1-2 as sub-concepts, and element 1-1 includes elements 1-1-1 and 1-1-2 as sub-concepts. can do. According to an embodiment, the element plan may further include association information (eg, ontology) between respective elements.

The element plan will be described by assuming an example of offline content. For example, offline objects may be classified according to their personality, such as'tourist spot','restaurant','shopping', and'culture', and may be pre-stored in an element plan suitable for each personality.

Referring to FIG. 2, when a user requests creation of content for an arbitrary restaurant (offline object), an element plan corresponding to a “restaurant” may be used for the requested offline object. Contents for'restaurant' can be classified into pictures (element 1) and services (element 2), and pictures (element 1) are external pictures (element 1-1) and interior pictures (element 1-2). Element). In addition, the second element (service) can be specifically classified into a 2-1 element (business hours), a 2-2 element (representative menu), and a 2-3 element (user rating), and the 2-2 The element (representative menu) can be further specifically classified into a 2-2-1 element (a picture of the representative menu) and a 2-2-2 element (a price of the representative menu).

For another example, if the nature of the offline object is'place', elements and sub-elements are the image of the place, the video of the place, the fee, the time of use, the address, the phone number, and the service available at the place. It may include people, etc. For another example, when the nature of the offline object is'product', the elements and sub-elements are the image of the product, the image of the product, the price of the product, the lowest price of the product, the type of the product, the expiration date of the product, and the The quality guarantee period, the country of origin of the product, the manufacturer of the product, etc. For another example, when the personality of the offline object is'person', the elements and sub-elements may include an image of the person, an image of the person, age, religion, occupation, place of residence, and the like.

In another embodiment, the intelligent server 100 may select one of a plurality of element plans stored in the database 120 in advance. In another embodiment, the intelligent server 100 may update a previously stored element plan. The update of the element plan may be performed by a developer tool. The intelligent server 100 may include the element the developer tool for updating the element plan. The developer may manually update the element plan by using the developer tool, or the content automatic generation platform 110 may determine a criterion for generating or updating the element plan.

In another embodiment, it can be created manually by the user. When the element plan for the offline content for which content creation is requested is not stored in the database 120, the intelligent server 100 may request the user to create an element plan.

3 is a block diagram illustrating an automatic content generation platform for generating content related to an offline object according to various embodiments of the present disclosure.

When receiving a request for automatic generation of online content for an offline object, the content automatic creation platform 110 collects a plurality of online contents related to the offline object, and based on the collected online contents, a plurality of online contents related to the offline object is received. A content element is generated, and reproduction content for an offline object may be generated by combining a plurality of generated content elements. Accordingly, referring to FIG. 3, the automatic content generation platform 110 according to an embodiment may include a content collection module 111, a content element production module 112, and a reproduction content generation module 113. The content automatic generation platform 110 may further include a verification module 114.

In an embodiment, the content collection module 111 may collect content related to an offline object. The content collection module 111 may collect content related to an offline object from the big data platform 140 included in the intelligent server 100 or directly through a network. In one embodiment, when a content reproduction request for an offline object is input, the content stored in the big data platform 140 may be collected using the offline object as a keyword, or online content may be searched and collected through a network.

In one embodiment, the content collection module 111 parses the HTML code constituting the Internet web page corresponding to the search result of the'offline object' and extracts the texts inserted through the tag associated with the text input in the HTML code. , It is possible to extract a plurality of images and texts existing on the web page. In addition, when a hyperlink tag exists on the HTML code, the content collection module 111 accesses the linked subpage through the hyperlink tag and uses a tag associated with text input from the HTML code of the subpage. By extracting a plurality of inserted images and texts together, extraction of the plurality of images and texts existing on the web page may be performed.

For example, the content collection module 111 may collect various online reviews and blog contents for the corresponding place when the nature of the offline object is related to a'place'. For another example, the content collection module 111 may collect various online reviews and blog contents for the product, contents related to the product in the open market, etc., when the nature of the offline object is related to'product'. have. For another example, the content collection module 111 may further collect SNS, USS, homepage, person search, person encyclopedia, and the like for the person when the personality of the offline object is related to'person'.

In one embodiment, the content element production module 112 may divide and reproduce a plurality of images and texts collected by the content collection module 111 for each element. A plurality of images and texts may be classified based on the element plan.

In one embodiment, the content element production module 112 may determine the element plan according to the nature of the requested offline object. For example, when the offline object is'Gwanghwamun', the content element production module 112 may use an element plan designed as a'tourist spot'. For another example, the content element production module 112 may use an element plan designed as a'restaurant' when the offline object is'Gwanghwamun Ramyun'. In FIG. 2, the case where the element plan is'restaurant' has been described.

In another embodiment, when the user inputs the offline object, the characteristics of the offline object may be input together, and the content element production module 112 may determine the element plan according to the characteristics of the input offline object.

In another embodiment, the content element production module 112 may generate an element plan based on a plurality of images and texts collected by the content collection module 111. For example, the content element production module 112 may select a key keyword using the number of appearances of the same object from a plurality of collected images and texts, and generate an element plan using the same.

The content element production module 112 may generate a content element by using an offline object and an element plan determined according to the characteristics of the offline object. Content element production module 112 The content element may be generated corresponding to each element and sub-elements of the element plan. The content element production module 112 may allocate each element of the element plan to each of the plurality of element AI units.

2 and 3, when the nature of the requested offline object is'restaurant', the content element corresponding to the first element (photo) may be generated by the first element AI unit 112-1, The content element corresponding to the second element (service) may be generated by the second element AI unit 112-2-. Although not shown, each of the first element AI units may further include sub-element AI units, and each of the sub-element AI units is a sub-element (eg, a 1-1 element, a 2-2-1 element, etc.) Sub content elements corresponding to can be created.

To this end, the content element production module 112 may classify a plurality of images and texts collected according to an element plan. Each of the element AI units 112-1 and 112-2 may generate a sub content element using a plurality of classified images and texts.

It can be created by an artificial intelligent (AI) system. The artificial intelligence system may be a rule-based system, or a neural network-based system (e.g., a feedforward neural network (FNN)), a recurrent neural network (RNN). )), may be Generative Adversarial Networks (GAN), or may be a combination of the above or other artificial intelligence systems.

Hereinafter, when the element (content) allocated to the element AI unit is an image, an operation of generating a new image will be described. The element AI unit may function as a generation model capable of learning using the collected images for offline objects and generating images for offline objects. The use of images generated by the element AI unit may not violate the copyright of the collected images.

FIG. 4 is a block diagram illustrating a technique of generating an image by the element AI unit of the content element production module 112, and FIG. 5 is a conceptual diagram illustrating a process by which the element unit of the content element production module generates an image.

As an example of what the element AI unit generates an image, the element AI unit can perform image translation. Image transformation refers to recreating the image by changing the image as the user wants, such as creating a black-and-white photo into a color photo and a simple illustration into a concrete photo.

Referring to FIG. 4, an image regeneration technique of Generative Adversarial Networks (GAN) will be described as an example of an element AI unit capable of regenerating an image.

GAN is an unsupervised learning-based generation model that learns two networks adversarially with a generator (Generator, 410) and a discriminator (Discriminator, 420). Input data may be input to the generator 410 and learned to generate a fake image 412 similar to the real image 411. A noise value may be input as the input data. The noise value can follow any probability distribution. For example, it may be data generated with Zero-Mean Gaussian. The discriminator 420 may learn to discriminate between an actual image and a fake image generated by the generator 410. More specifically, if the actual image 411 is input, a high probability may be obtained, and if a fake image 412 is input, the probability may be reduced. That is, the discriminator 420 may gradually learn to discriminate between the real image 411 and the fake image 412. When the generator 410 and the discriminator 420 sufficiently learn, the generator 410 may make the data distribution of the generated fake image 412 follow the data distribution of the actual image 411. In this case, the discriminator 420 may be correct or incorrect with a probability of 1/2 regardless of which image the generator 410 generates. In this state, the discriminator 420 cannot determine whether the fake image 412 is real or fake. The fake image 412 generated in this state can be used continuously for machine learning, and can be used as a regenerated image for an offline object.

In one embodiment, the element AI unit defines an image for an offline object as a first element, associates different parts of the image for an offline object as a plurality of first sub-elements, and then, a first element and a plurality of first elements. You can learn about sub-elements and recreate images for offline objects. In this case, the regenerated image may be defined as a first content element obtained by combining and processing a plurality of first sub content elements and a plurality of first sub content elements.

Referring to FIG. 5, it can be seen that the images 510 collected for'Gwanghwamun' as offline objects are learned using the GAN, and a regenerated image 520 is finally generated. In this case, the entire image may be the first element in'Gwanghwamun', and the building (or, element 1-1) and the background (or element 1-2) of'Gwanghwamun' among the images are each a first sub It can be defined as an element. In addition, the first sub-element defined as a building includes details such as the center of the building (or element 1-1-1), the roof (or element 1-1-2), and the door (or element 1-1-3). It can be further defined as an elemental element. That is, the element AI unit may divide and learn the images collected for Gwanghwamun into a first element and sub-elements, combine the generated sub-content elements, and process them again to reproduce an image for the entire'Gwanghwamun'.

In one embodiment, the reproduction content generation module 113 may combine and process the content elements (or content sub-elements) generated in the content element production module 112 to reproduce the content for the offline object. The reproduced content may be transmitted to the user device and provided to the user.

Hereinafter, when the element (content) allocated to the element AI unit is text, an operation of generating a new image will be described. For example, if the element assigned to the element AI unit is the user's'rating' for an offline object, the element AI unit uses the data related to the'rating' among the collected data (text) on the offline object to provide a comprehensive'rating'. You can create data for'. The collected data can include data based on a perfect score of 5, data based on a perfect score of 10, and data based on characters such as upper, middle, and lower, and the element AI unit synthesizes these and reproduces a comprehensive rating based on one standard. can do.

6 shows an example in which reproduction content for an offline object is provided to a user. Referring to FIG. 6, it can be seen that reproduction content for an offline object is provided through an app for providing a service introducing Seoul recommended travel destinations on a user device. For example, the content 610 for'Gwanghwamun' is generated by the content automatic creation platform 111 using an element plan including an image (first element) 611 and a rating (second element) 612. Can be reproduced. Here, the element plan only illustrates two elements (image and rating), but the element plan may include more elements and/or sub-elements (e.g. usage time, usage fee, etc.), and other elements not shown. Reproduction content may be provided in which the elements are included.

Although the example that is directly provided to the user has been described above, the reproduced content may be provided to a service provider, and the service provider may be reprocessed and used according to the service provided.

According to an embodiment, the verification module 114 may verify the reproduced content. For example, the verification module 114 may calculate the completeness of the reproduced content and transmit the feedback on the result of the operation to the content collection module 111, the content element production module 112, and/or the reproduction content generation module 113. have. The content collection module 111, the content element production module 112 and/or the reproduction content generation module 113 reflect the feedback of the verification module 114 to further collect content, re-produce content elements, or generate The elements can again be combined and/or machined.

For another example, the verification module 114 calculates the similarity between the reproduced content and the existing content, and provides feedback on the result of the operation, the content collection module 111, the content element production module 112, and/or the reproduction content generation module. Can be delivered to (113). The content collection module 111, the content element production module 112 and/or the reproduction content generation module 113 re-produces the content element by reflecting the similarity of the verification module 114, or the generated elements are combined and/or Or it can be processed.

7 is a flow diagram illustrating an exemplary process 700 for generating online content for an offline object, in accordance with at least some embodiments of the present disclosure.

For example, the process 700 may be performed under the control of a computing device such as the intelligent server 100 or the automatic content creation platform 110 disclosed in FIGS. 1 to 3. Alternatively, some of the processes of process 700 may be performed on an external computing device such as the user device 200 disclosed in FIG. 1.

Process 700 may receive a content creation request for an offline object at block 701.

Process 700 may collect a plurality of online content relating to the offline object at block 703. In one embodiment, the operation of collecting a plurality of online contents may be performed in response to receiving a request for content creation related to an offline object, or may be collected in advance and collected in a database (eg, 120 in FIG. 1) or a big data platform (eg: It may be stored in 140) of FIG.

For example, multiple online contents are extracted by parsing the HTML code that composes the Internet web page corresponding to the search result of the'offline object' and searching the images and texts inserted through the tags related to text input in the HTML code. It may include a plurality of images and texts.

Process 700 may continue at block 705 to determine an element plan based on the nature of the offline object. The element plan may be determined according to the nature of the offline object. The characteristics of the offline object may include'tourist spot','restaurant', and the like. The nature of the offline object may be automatically determined according to the content collection result, and may be input together when the user requests reproduction of the content for the offline object. The element plan may be previously stored in an element database (eg, 120 in FIG. 1). In another example, the element plan may be newly created according to the content collection result, or may be regenerated by updating an already stored element plan.

Process 700 may lead to generating a plurality of content elements for the offline object based on the element plan determined at block 707. An element AI unit is assigned to each of the elements and/or sub-elements included in the element plan, and the element AI unit may generate a content element using the collected content according to the nature of the element and/or sub-element.

Process 700 may generate reproduction content for an offline object by combining the plurality of content elements generated in block 709. In one embodiment, the reproduced content may be transmitted to the user device and provided to the user. In another example, the reproduced content may be provided to a service provider, and the service provider may reprocess and use it again according to a service that provides it.

Although not shown, the process 700 further includes the step of calculating a degree of completion by verifying the reproduced content, or calculating a degree of similarity by comparing the reproduced content with the existing content, and a feedback system for improving the quality of the reproduced content. This can be further implemented.

The invention is not intended or limited to the embodiments described in this flowchart. For example, one of ordinary skill in the art will appreciate that for the present process disclosed herein, the functions performed in the process and method may be implemented in a different order. On the other hand, the schematic operations illustrated in FIG. 7 are provided only as examples, and without departing from the essence of the disclosed embodiment, some of the operations may be optional, may be combined into fewer operations, or the operations described above may be Can be added and expanded.

8 illustrates an exemplary computer program product 800 that may be used to create content relating to an offline object, in accordance with at least some embodiments of the present disclosure. An exemplary computer program product 800 is provided using, for example, a signal bearing medium 802. In some embodiments, the signal bearing medium 802 of one or more computer program products 800 may include a computer readable medium 806, a recordable medium 808, and/or a communication medium 810.

The instructions 804 included in the signal-bearing medium 802 may be executed by, for example, a computing device such as the intelligent server 100 illustrated in FIGS. 1 to 3 and the content automatic creation platform 110. When executed, the instruction 804 may cause the computing device to perform a process for reproducing content related to an offline object.

For example, the instruction 804 may cause a computing device including a plurality of artificial intelligence units (AI units) to collect a plurality of online content related to an offline object; An operation of generating a plurality of content elements by a plurality of AI units based on the plurality of online contents-In the generating operation, the computing device causes each of the plurality of AI units to be With respect to the element, the plurality of online contents are learned, and based on the result of the learning, a content element for the predetermined element is generated, where the predetermined elements to be learned by the plurality of AI units are different from each other. -; Combining at least some of the plurality of content elements; And at least one command for executing an operation of generating reproduction content for the offline object based on the result of the combination.

The foregoing description of the present application is for illustrative purposes only, and those of ordinary skill in the art to which the present application pertains will be able to understand that it can be easily modified into other specific forms without changing the technical spirit or essential features of the present application. Therefore, it should be understood that the embodiments described above are illustrative and non-limiting in all respects. For example, each component described as a single type may be implemented in a distributed manner, and similarly, components described as being distributed may also be implemented in a combined form.

As described above, the subject to be claimed in the present disclosure was examined in detail. The subject matter claimed in this disclosure is not limited in scope to the specific implementations described above. For example, in some implementations, it may be in the form of hardware that is operably used on a device or a combination of devices, in other implementations it may be implemented in the form of software and/or firmware, and in another implementation, a signal bearing medium, It may include one or more articles such as storage media. Here, the storage medium, such as a CD-ROM, a computer disk, or a flash memory, is an instruction capable of causing execution of a corresponding processor according to the above-described implementation when executed by a computing device such as, for example, a computing system, a computing platform, or other system. Can be saved. Such computing devices may include one or more processing units or processors, one or more input/output devices such as displays, keyboards, and/or mice, and one or more memory such as static random access memory, dynamic random access memory, flash memory and/or hard drive. It may include.

On the other hand, whether to implement the system as hardware or software is a design choice that generally represents a cost-effective tradeoff. In this disclosure, there are various means (e.g., hardware, software and/or firmware) that can be influenced by processes, systems, and other technologies, and the preferred means is that processes and/or systems and/or other technologies are used. It will change according to the context of being. For example, if the implementer decides that speed and accuracy are paramount, the implementer can primarily choose hardware and/or firmware means, and if flexibility is paramount, the implementer can choose primarily software implementation; Or, as another alternative, the implementer may opt for some combination of hardware, software and/or firmware.

In the foregoing detailed description, various embodiments of an apparatus and/or process have been described through block diagrams, flow charts, and/or other examples. Such block diagrams, flow charts, and/or other examples will include one or more functions and/or operations, and those skilled in the art will understand that each function and/or operation in the block diagrams, flow charts, and/or other examples is hardware, software, firmware, Or it will be appreciated that they may be implemented individually or collectively by any combination thereof. In one embodiment, some parts of the object described in the present disclosure may be implemented through an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA), a Digital Signal Processor (DSP), or other form of integration. In contrast, some aspects of embodiments of the present disclosure include one or more computer programs running on one or more computers (e.g., one or more programs running on one or more computer systems), one or more programs running on one or more processors ( For example, one or more programs running on one or more microprocessors), firmware, or substantially any combination thereof, may be implemented in an integrated circuit wholly or partially evenly, and writing code for software and/or firmware And/or the design of the circuit is within the skill of those skilled in the art in light of the present disclosure. In addition, those skilled in the art will understand that mechanisms of the object of the present disclosure can be distributed to various types of program products, and examples of the object of the present disclosure are applied irrespective of the specific type of signal bearing medium used to actually perform the distribution. Will understand.

While certain exemplary techniques have been described and illustrated herein using various methods and systems, those skilled in the art can understand the possibility of substitution with various other modifications or equivalents without departing from the claimed subject matter. Additionally, many modifications can be made to adapt a particular situation to the claimed subject's teachings without departing from the central concept described herein. Thus, while claimed subject matter is not limited to the specific examples disclosed, it is intended that such claimed subject matter may also include all embodiments falling within the scope of the appended claims and their equivalents.

Throughout this disclosure, when a part is said to be "connected" with another part, this includes not only the case that it is "directly connected", but also the case that it is "electrically connected" with another element interposed therebetween. do. In addition, throughout the present disclosure, when a member is said to be positioned "on" another member, this includes not only a case where a member is in contact with another member, but also a case where another member exists between the two members. Furthermore, throughout the present disclosure, when a certain part "includes" a certain component, it means that other components may be further included, rather than excluding other components, unless specifically stated to the contrary. The terms "about," "substantially", etc. of the degree used in the present disclosure are used at or close to the numerical value when manufacturing and material tolerances specific to the stated meaning are presented, and to aid understanding of the present application In order to avoid unreasonable use by unscrupulous infringers of the stated disclosures, either exact or absolute figures are used.

The scope of the present disclosure is indicated by the claims to be described later rather than the above detailed description, and the meaning and scope of the claims, and all changes or modified forms derived from the concept of equivalents thereof are interpreted as being included in the scope of the present disclosure. It should be.

Claims (10)

As a content automatic generation device,
A content collection module, configured to collect a plurality of online content related to the offline object;
A content element production module comprising a plurality of artificial intelligence units (AI units) and configured to generate a plurality of content elements for a plurality of elements of the offline object, wherein the plurality of AI units include a first AI A unit and a second AI unit, wherein the first AI unit is configured to learn a first element related to the offline object and generate a first content element for the first element based on the plurality of online contents, , The second AI unit is configured to learn a second element related to the offline object based on the plurality of online contents and generate a second content element for the second element, and the first element of the offline object and A content element production module, wherein the second element is different; And
Receiving a plurality of content elements including the first content element and the second content element from the content element production module, combining at least some of the plurality of content elements, and using the combination, to the offline object A reproduction content generation module configured to generate reproduction content relating to
Including,
The first AI unit includes a plurality of first sub AI units,
The plurality of first sub-AI units learn a plurality of first sub-elements of the first element to generate a plurality of first sub-content elements, and the first AI unit combines the plurality of first sub-content elements Thereby, being configured to generate the first content element for the first element. The method of claim 1,
The content automatic generation device,
Input module for receiving an input regarding the offline object
Including more,
Wherein the content collection module is configured to collect a plurality of online contents based on the input. The method of claim 2,
The content element production module is configured to determine an element plan corresponding to the offline object based on the input, and to generate the plurality of content elements according to the determined element plan. delete The method of claim 1,
The first element is an image for the offline object,
Each of the plurality of first sub-elements is related to a part of the offline object,
The first AI unit is configured to extract a plurality of images for the offline object from the plurality of online contents,
Each of the plurality of first sub-AI units learns a corresponding one of the plurality of first sub-elements from the extracted plurality of images, and generates a first sub-content element,
The first AI unit is configured to combine and process the generated plurality of first sub-content elements to generate the first content element, which is an image for the offline object. The method of claim 1,
Wherein the plurality of online content is an image of the offline object, and the reproduction content generation module is configured to generate the reproduced content that is an image of the offline object. The method of claim 1,
The reproduced content generating module comprises a feedback unit that calculates the completeness of the reproduced content and returns a feedback on the result of the operation as an input of the reproduced content generating module. The method of claim 1,
A verification module that calculates a similarity between each of the plurality of online contents on the reproduced contents generated by the reproduced contents generating module
Content automatic generation device further comprising. A method for automatically generating content performed under the control of a computing device, wherein the computing device includes a plurality of artificial intelligence units (AI units), and the method for automatically generating content comprises:
Collecting a plurality of online contents related to the offline object;
Based on the plurality of online contents, generating a plurality of content elements by a plurality of AI units-The plurality of AI units include a first AI unit and a second AI unit, the first AI unit And learning a first element related to the offline object based on the plurality of online contents, and generating a first content element for the first element, and a second AI unit is configured to generate the offline object based on the plurality of online contents. Learning a second element relating to an object and generating a second content element for the second element, wherein the first element and the second element of the offline object are different;
Combining at least some of a plurality of content elements including the first content element and the second content element; And
Based on the result of the combination, generating reproduction content for the offline object
Including,
The first AI unit includes a plurality of first sub AI units,
The plurality of first sub-AI units learn a plurality of first sub-elements of the first element to generate a plurality of first sub-content elements, and the first AI unit combines the plurality of first sub-content elements Thereby being configured to generate the first content element for the first element. A computer program stored in a computer-readable recording medium in order to automatically generate content related to an offline object, wherein when the computer program is executed, a computing device including a plurality of artificial intelligence units (AI units) causes a ,
Collecting a plurality of online contents related to the offline object;
An operation of generating a plurality of content elements by a plurality of AI units based on the plurality of online contents-The plurality of AI units include a first AI unit and a second AI unit, and the first AI unit And learning a first element related to the offline object based on the plurality of online contents, and generating a first content element for the first element, and a second AI unit is configured to generate the offline object based on the plurality of online contents. Learning a second element relating to an object and generating a second content element for the second element, wherein the first element and the second element of the offline object are different;
Combining at least some of a plurality of content elements including the first content element and the second content element; And
An operation of generating reproduction content for the offline object based on the result of the combination
Containing one or more computer-executable instructions that make it executable,
The first AI unit includes a plurality of first sub AI units,
The plurality of first sub-AI units learn a plurality of first sub-elements of the first element to generate a plurality of first sub-content elements, and the first AI unit combines the plurality of first sub-content elements Thereby being configured to generate the first content element for the first element.

Images