KR20220145718A - Method for participating live broadcast using local information - Google Patents

Abstract

Disclosed is a live broadcast participation method using location information. The live broadcast participation method performed in a live broadcast system associated with a user terminal, a live broadcast terminal, and an affiliated store terminal comprises the steps of: receiving broadcast data for live broadcasting from the live broadcast terminal; receiving user data including user location information from the user terminal; selecting a recommended broadcasting to be provided to the user terminal using location information included in the user data; providing a recommendation list including the recommended broadcasting; receiving a participation signal for a specific broadcasting from the user terminal among a plurality of recommended broadcastings included in the recommendation list; displaying an image of the specific broadcasting and an offline request button for the specific broadcasting on the user terminal; receiving a selection for an offline application button from the user terminal and issuing an offline ticket for the specific broadcasting; and receiving a usage confirmation for the offline ticket from the affiliated store terminal. Accordingly, the present invention can encourage users to purchase.

Description

{Method for participating live broadcast using local information}

The present invention relates to a method of participating in a live broadcast. Specifically, the present invention relates to a method of participating in a live broadcast that provides a recommended broadcast to a user based on the user's location information and provides a service for online or offline participation in the recommended broadcast.

The content described in this section merely provides background information for the present embodiment and does not constitute the prior art.

Due to the recent development of Internet communication, the sales of online shopping, such as Internet shopping and home shopping broadcasting, where products can be purchased at home, are rapidly increasing.

However, in the case of online shopping, it is difficult for small business owners with low sales to enter the market due to high brokerage fees and advertising fees. In addition, in online shopping, consumers have to decide whether to purchase by using only pre-photographed images or videos of selling products. However, when the actual product is different from the expected product, there is a problem in that it is difficult to return or exchange the product.

Accordingly, there is a need for a system that allows users to receive publicity broadcasts or sales broadcasts of businesses operating online shopping and to directly use goods or services provided by the businesses in offline stores of the businesses.

In addition, for the operation of the above system, the need for a platform that can receive only online shopping-related broadcasts of businesses located in the vicinity of the user and receive related broadcasts is also increasing.

An object of the present invention is to provide a method of participating in a live broadcast that receives a user's application for a live broadcast of a business and issues an offline ticket for using a product or service disclosed in the live broadcast.

Another object of the present invention is to provide a method of participating in a live broadcast in which a recommended broadcast available in a region where a user is located is selected based on location information of the user and a recommendation list including the selected broadcast is provided to the user.

Another object of the present invention is to provide a method of participating in a live broadcast that recommends a broadcast that is highly likely to be preferred by a user using a deep learning module based on user data and past participation data.

The objects of the present invention are not limited to the above-mentioned objects, and other objects and advantages of the present invention not mentioned may be understood by the following description, and will be more clearly understood by the examples of the present invention. It will also be readily apparent that the objects and advantages of the present invention may be realized by the means and combinations thereof indicated in the appended claims.

In the live broadcasting participation method according to an embodiment of the present invention, the live broadcasting participation method is performed in a live broadcasting system associated with a user terminal, a live broadcasting terminal, and an affiliated store terminal, and broadcasting data for live broadcasting from the live broadcasting terminal receiving user data including the location information of the user from the user terminal, selecting a recommended broadcast to be provided to the user terminal by using the location information included in the user data; providing a recommendation list including a recommended broadcast; receiving a participation signal for a specific broadcast from the user terminal among a plurality of recommended broadcasts included in the recommendation list; an image for the specific broadcast to the user terminal; displaying an offline application button for the specific broadcast, receiving a selection for the offline application button from the user terminal, issuing an offline ticket for the specific broadcast, and using the offline ticket from the affiliated store terminal receiving an acknowledgment.

In addition, the step of selecting a recommended broadcast to be provided to the user terminal may include receiving preset area range setting information input to the user terminal, and applying the area range setting information to the location information so that the user's information about broadcasting is performed. It may include deriving a reception request area, receiving a participation area of a broadcast recommendation candidate being broadcast live, and selecting the broadcast recommendation candidate as the recommended broadcast when the reception request area includes the participation area have.

In addition, the step of selecting the recommended broadcast to be provided to the user terminal may include deriving the nearest subway station or bus stop to the location information, receiving the operation information of the subway station or the bus stop, and receiving the operation information based on the operation information. Calculating the estimated movement time required from the user's location to the transmission location of the broadcast recommendation candidate, and selecting the broadcast recommendation candidate as the recommended broadcast when the calculated estimated movement time is included within a preset reference time can

In addition, in the step of issuing the offline ticket, if the specific broadcast is a broadcast related to the sale of a product, checking the inventory of the product, issuing an offline ticket for exchanging the product, and performing the specific broadcast Alternatively, in the case of a broadcast related to an event, it may include checking the number of people allowed to enter the performance or event, selecting the seat number of the user based on the remaining seats, and issuing an offline ticket.

In addition, the step of receiving the usage confirmation for the offline ticket comprises:

The affiliate store for the specific broadcast recognizes the offline ticket through the affiliated store terminal, checks the validity period of the offline ticket, and checks whether the usable area of the offline ticket is included in the reception request area of the user terminal However, the usable area may include a participating area capable of receiving the specific broadcast.

In addition, the step of selecting a recommended broadcast to be provided to the user terminal, based on the user data and the user's past participation data, calculates an expected preference for a broadcast recommendation candidate using a pre-learned deep learning module, , selecting the broadcast recommendation candidate whose calculated expected preference is higher than a reference value or is included in a preset high rank number as the recommended broadcast.

In addition, calculating the predicted preference is to derive the first predicted preference of the specific broadcast for the user by using the deep learning module, and select a subject whose similarity between the user and the user data is higher than a reference value, , deriving a second expected preference of the specific broadcast for the selected target using the deep learning module, and deriving a final expected preference based on the first expected preference and the second expected preference have.

In addition, the deep learning module includes an input layer using the user data, the past participation data, and the broadcast data as input nodes, an output layer using the predicted preference as an output node, and between the input layer and the output layer. one or more hidden layers disposed in , and weights of nodes and edges between the input node and the output node may be updated by a learning process of the deep learning module.

In addition, receiving the evaluation of the user for the specific broadcast from the user terminal, and further comprising the steps of re-learning the deep learning module using the received evaluation, re-learning the deep learning module The step may include comparing the predicted preference and the evaluation for the specific broadcast, and re-learning the deep learning module when the comparison result is out of a predetermined error range.

In addition, in the step of re-learning the deep learning module, the deep learning module is configured such that the user data, the past participation data, and broadcast data in which the user participated are applied to the input node, and the evaluation is performed to the output node By being applied, the weights of the nodes and edges may be updated.

The live broadcasting participation method of the present invention may receive a user's application for live broadcasting and issue an offline ticket for using goods or services initiated in the live broadcasting. Accordingly, the present invention enables users to use products or services provided by live broadcasting at offline affiliates, so that tangible goods and intangible content (workshops and academies, etc.) can be sold online and offline simultaneously. have. The present invention can generate a virtuous cycle effect of increasing user convenience and increasing sales of affiliated stores.

Also, according to the present invention, it is possible to select a recommended broadcast available in the area where the user is located based on the user's location information. Subsequently, it is possible to provide a recommendation list including the selected recommended broadcast. Through this, the present invention can obtain a marketing effect that can lead to face-to-face sales by increasing users' accessibility to affiliate stores selling services and products by selecting a user-customized recommendation broadcast for region-based exposure.

In addition, the live broadcast participation method of the present invention derives an expected preference for each user using a deep learning module based on user data and past participation data, and generates a recommendation list for broadcasting based on the derived predicted preference. have. Through this, the user's convenience and satisfaction can be improved by recommending a customized broadcast reflecting the user's preference, and the user's purchase can be promoted.

The specific effects of the present invention in addition to the above will be described together while explaining the specific details for carrying out the invention below.

1 is a conceptual diagram illustrating a live broadcasting system according to an embodiment of the present invention.
FIG. 2 is a block diagram illustrating each component of the live broadcasting system of FIG. 1 .
3 is a flowchart illustrating a method of participating in a live broadcast according to some embodiments of the present invention.
4 is a diagram for explaining an example of a method of participating in a live broadcast according to some embodiments of the present invention.
5 is a flowchart illustrating a method of selecting a recommended broadcast using regional range setting according to an embodiment of the present invention.
FIG. 6 is a diagram for explaining an example of selecting a recommended broadcast using the regional range setting of FIG. 5 .
7 is a flowchart for explaining a method of selecting a recommended broadcast using public transportation operation information according to another embodiment of the present invention.
8 is a diagram for explaining an example of selecting a recommended broadcast using the public transportation operation information of FIG.
9 is a flowchart illustrating an offline ticket issuance method according to some embodiments of the present invention.
10 is a flowchart illustrating a method of receiving use of an offline ticket according to some embodiments of the present invention.
11 is a flowchart illustrating a method of selecting a recommended broadcast using a deep learning module according to another embodiment of the present invention.
12 is a block diagram schematically illustrating a deep learning module used in the method of selecting a recommended broadcast of FIG. 11 .
13 is a diagram illustrating the configuration of the deep learning module of FIG. 12 .
14 is a flowchart for explaining an example of calculating an expected preference for broadcasting in step S343.
15 is a flowchart illustrating an interface for receiving a user's evaluation in a method of participating in a live broadcast according to another embodiment of the present invention.
16 is a block diagram schematically illustrating a re-learning step of a deep learning module used in a method of participating in a live broadcast according to another embodiment of the present invention.

Terms or words used in this specification and claims should not be construed as being limited to a general or dictionary meaning. In accordance with the principle that the inventor can define a term or concept of a word to best describe his/her invention, it should be interpreted as meaning and concept consistent with the technical idea of the present invention. In addition, the embodiments described in the present specification and the configurations shown in the drawings are only one embodiment in which the present invention is realized, and do not represent all the technical spirit of the present invention, so they can be replaced at the time of the present application. It should be understood that there may be various equivalents and modifications and applicable examples.

Terms such as first, second, A, B, etc. used in this specification and claims may be used to describe various elements, but the elements should not be limited by the terms. The above terms are used only for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, a first component may be referred to as a second component, and similarly, a second component may also be referred to as a first component. The term 'and/or' includes a combination of a plurality of related listed items or any of a plurality of related listed items.

Terms used in this specification and claims are used only to describe specific embodiments, and are not intended to limit the present invention. The singular expression includes the plural expression unless the context clearly dictates otherwise. It should be understood that terms such as “comprise” or “have” in the present application do not preclude the possibility of addition or existence of features, numbers, steps, operations, components, parts, or combinations thereof described in the specification in advance. .

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

Terms such as those defined in a commonly used dictionary should be interpreted as having a meaning consistent with the meaning in the context of the related art, and should not be interpreted in an ideal or excessively formal meaning unless explicitly defined in the present application. does not

In addition, each configuration, process, process or method included in each embodiment of the present invention may be shared within a range that does not technically contradict each other.

Hereinafter, with reference to FIGS. 1 to 16 , a live broadcasting participation method and a live broadcasting system for providing live broadcasting that can be participated in online and offline based on the user's location information will be described in detail.

1 is a conceptual diagram illustrating a live broadcasting system according to an embodiment of the present invention.

Referring to FIG. 1 , a live broadcasting system according to some embodiments of the present invention includes a content management server 100 , a user terminal 200 , a location-based server 300 , an affiliated store terminal 400 , a communication network 500 , and a live broadcast system. and a broadcast terminal (LS).

The content management server 100 (hereinafter, the server) according to the present invention may receive broadcast data for a live broadcast (hereinafter, broadcast) through the live broadcast terminal LS and the location-based server 300 . In this case, the broadcast data may include first broadcast data and second broadcast data.

Specifically, when a product or event is broadcast at an affiliated store registered in an open market, etc., the server 100 may receive the first broadcast data, which is a live broadcast image, through the live broadcast terminal LS, and the location-based The second broadcast data may be provided from the server 300 . In this case, the second broadcast data may include the location of the affiliate store where the broadcast is performed, product information or service information of the broadcast, and the like.

Subsequently, the server 100 may receive data about the user (hereinafter, user data) through the user terminal 200 . Here, the user data may include the user's access IP information, age, gender, location information, user setting information, and interest category (eg, product and service types, etc.). The server 100 may select a recommended broadcast for each user from the broadcast recommendation candidates based on the user's location information and broadcast data.

In this case, the broadcast recommendation candidate may be any broadcast that is captured by the live broadcast terminal LS and transmitted to the server 100 or registered in the open market. Subsequently, the recommended broadcast may be selected based on a preset criterion from among a plurality of broadcast recommendation candidates.

For example, the server 100 may select a recommended broadcast using the user's reception request area derived using the location information. Specifically, the server 100 may receive regional range setting information set through the user terminal 200 . Here, the regional range setting information is information about a regional range in which a broadcast recommendation candidate that can be provided to a user is included, centering on the user's location information. Of course, such a regional range may be set for each administrative district or may be set by designating a specific radius based on the user's location information. However, this is only an example, and the present invention is not limited thereto.

Then, the server 100 may derive a reception request area in which the user can receive the broadcast based on the area range setting information set in advance by the user.

Subsequently, when the derived reception request area includes a participation possible area of the broadcast recommendation candidate, the server 100 may select the corresponding broadcast recommendation candidate as the user's recommended broadcast. Details of the selection of the recommended broadcasting using the regional range setting information will be described with reference to FIGS. 5 and 6 .

As another example, the server 100 may calculate the estimated travel time of transportation (eg, public transportation and private vehicle) required from the user's current location to the broadcast transmission location, based on the user's location information. The server 100 may select a broadcast recommendation candidate in which the calculated estimated movement time is included within a preset reference time as a recommended broadcast. The details of this will be described with reference to FIGS. 7 and 8 .

On the other hand, the server 100 may provide a recommendation list based on the recommendation broadcast. The server 100 may generate a recommendation list by listing the recommended broadcasts based on the distance between the user's location and the transmission location of the recommended broadcast or the estimated moving time.

Subsequently, the server 100 provides the generated recommendation list to the user terminal 200 .

Subsequently, when receiving a signal to select a broadcast displayed on the recommendation list through the user terminal 200 , the server 100 may display an image and an offline application button for the selected specific broadcast on the user terminal 200 . have.

Subsequently, when receiving the user's selection of the offline application button, the server 100 may issue an offline ticket for the specific broadcast for which the selection is received to the user terminal 200 . In this case, the offline ticket may include information for using an offline service, and the information included may be different according to a broadcasting category. For example, the offline ticket may include a barcode, a QR code, the validity period of the offline ticket, information on a store selling a product or operation information of a service, and the like.

Subsequently, the server 100 may receive a usage confirmation for the offline ticket.

For example, when a user purchases a product using an offline ticket at an affiliated store, the affiliated store terminal 400 may receive the offline ticket information by recognizing the offline ticket displayed on the user terminal 200 . Subsequently, the affiliated store terminal 400 may check whether the recognition date of the offline ticket is included within the validity period included in the offline ticket, and may check whether the product of the offline ticket and information on the purchased product of the user match. In addition, the affiliated store terminal 400 may check whether the available area of the offline ticket is included in the user's reception request area.

Subsequently, when all checks for the offline ticket are completed through the affiliate store terminal 400 , the affiliate store may provide a product or service related to the offline ticket, and the affiliate store terminal 400 may process the use of the offline ticket.

Subsequently, the server 100 may receive a usage confirmation for the offline ticket from the affiliate store terminal 400 .

In addition, the server 100 according to another embodiment of the present invention may select a recommended broadcast using a deep learning module.

For example, the server 100 calculates an expected preference using deep learning based on the user's data, past participation data for broadcasting, and broadcasting data, and selects a recommended broadcast based on the predicted preference to select a recommended broadcast for the user terminal 200 ) can be provided.

In this process, the server 100 may select a recommended broadcast using artificial neural networks learned based on each user's data and past participation data. A detailed description of the deep learning module using an artificial neural network will be described later in detail with reference to FIGS. 4 and 5 .

Subsequently, when the broadcast ends or the confirmation of use of the offline ticket for the broadcast is received, the server 100 may receive the user's evaluation of the broadcast selected by the user from the recommendation list.

Then, the server 100 may re-learn the deep learning module by adjusting the weight of the deep learning module by comparing the user evaluation received through the evaluation receiving module and the predicted preference calculated in advance. However, the present invention is not limited thereto

Meanwhile, the server 100 and the user terminal 200 may be implemented as a server-client system. The server 100 may transmit/receive data to and from the user terminal 200 through a wired/wireless network. For example, the server 100 may transmit/receive data through a terminal application installed in the user terminal 200 .

Here, the terminal application may be a dedicated application for providing the recommendation list and the evaluation receiving module, or a web browsing application for providing the recommendation list and the evaluation receiving module through a web page. Here, the dedicated application for providing the recommendation list and the evaluation receiving module may be an application embedded in the user terminal 200 or an application downloaded from an application distribution server and installed in the user terminal 200 .

That is, the user terminal 200 may provide the recommendation list and the evaluation receiving module provided from the server 100 to the user through the terminal application installed in the user terminal 200 .

Hereinafter, for convenience of explanation, it is assumed that the user terminal 200 displays the interface, the recommendation list, and the evaluation receiving module provided from the server 100 on the screen using a pre-installed terminal application (hereinafter, the application). let it do

In addition, the user terminal 200 refers to a communication terminal capable of operating a terminal application in a wired/wireless communication environment. The user terminal 200 may be a user's portable terminal. Although the user terminal 200 in FIG. 1 is illustrated as a smart phone, which is a type of portable terminal, the present invention is not limited thereto, and as described above, it can be applied without limitation to a device capable of mounting a terminal application. . For example, the user terminal 200 may include various types of electronic devices such as a personal computer (PC), a notebook computer, a tablet, a mobile phone, a smart phone, and a wearable device (eg, a watch-type terminal).

In addition, although only one user terminal 200 is illustrated in the drawing, the present invention is not limited thereto, and the server 100 may operate in conjunction with a plurality of user terminals 200 .

Additionally, the user terminal 200 includes an input unit for receiving a user's input, a display unit for displaying visual information, a communication unit for transmitting and receiving signals to and from the outside, and processing data and controlling each unit inside the user terminal 200, It may include a control unit for controlling data transmission/reception between units. Hereinafter, a command performed by the control unit inside the user terminal 200 according to a user's command is collectively referred to as a command performed by the user terminal 200 .

The location-based server 300 may transmit the second broadcast data received from the affiliated store terminal 400 to the server 100 . Here, a plurality of affiliated stores may pre-register information about a broadcast to be transmitted to the affiliated store terminal 400 . The location-based server 300 may provide the second broadcast data registered in the affiliate store terminal 400 to the server 100 in real time.

Also, the location-based server 300 may transmit the second broadcast data received through the affiliate store terminal 400 to the content management server 100 .

Specifically, the location-based server 300 may receive the second broadcast data for the broadcast selected by the user from the affiliate store terminal 400 . The location-based server 300 may transfer the received offline ticket application information to the affiliate store terminal 400 to complete the application for offline ticket issuance. Next, the affiliate store in which the affiliate store terminal 400 is installed provides information on the remaining seat such as the inventory status of goods or services (eg, events and performances) based on the received offline ticket application information to the location-based server 300 . can do.

Here, the affiliated store terminal 400 may be a terminal located in the affiliated store. The affiliated store terminal 400 may store and manage second broadcast data such as the location of the affiliated store, product information of broadcasting, service information of broadcasting, and price information of goods or services. The affiliated store terminal 400 may provide the second broadcast data to the location-based server 300 . In addition, when receiving the use of the offline ticket from the user, the affiliated store terminal 400 may determine whether the user uses the offline ticket based on the second broadcast data and the offline ticket information.

The live broadcasting terminal LS includes a photographing terminal that shoots live broadcasting. The live broadcasting terminal LS may transmit the first broadcasting data, which is a captured image of broadcasting, to the content management server 100 . In this case, the affiliate store terminal 400 and the live broadcasting terminal LS may be installed and used in the same affiliate store.

Meanwhile, the communication network 500 serves to connect the content management server 100 , the user terminal 200 , and the location-based server 300 . That is, the communication network 500 refers to a communication network that provides an access path so that the user terminals 200 can transmit and receive data after accessing the content management server 100 and the location-based server 300 . The communication network 500 is, for example, a wired network such as LANs (LoCAL Area Networks), WANs (Wide Area Networks), MANs (Metropolitan Area Networks), ISDNs (Integrated Service Digital Networks), wireless LANs, CDMA, Bluetooth, satellite communication, etc. may cover a wireless network, but the scope of the present invention is not limited thereto.

Hereinafter, a detailed configuration of the live broadcasting system will be described in detail.

FIG. 2 is a block diagram illustrating each component of the live broadcasting system of FIG. 1 .

Referring to FIG. 2 , the content management server 100 may include a data management unit 110 , an open market linkage unit 120 , and a deep learning unit 130 .

Specifically, the data management unit 110 may store and manage user data and past participation data. The data management unit 110 may store and manage user data, past participation data, and broadcasting data.

The user data may include the user's access IP information, age, gender, location information, setting information of the user, and interest categories (such as product and service types). In this case, the location information may include location analysis information including surrounding information about the user's location.

The data management unit 110 may inquire the user's past participation data based on the user data, and store and manage the inquired past participation data. The data management unit 110 may also inquire past participation data of broadcasts participated in through other affiliated stores or other applications. Here, the past participation data may include a broadcasting name for a previously participated broadcasting, a broadcasting category, offline ticket issuance history, participation time information, and the like.

In addition, the server 100 may receive the first broadcast data from the live broadcast terminal LS, and may receive the second broadcast data from the location-based server 300 . The broadcast data may include information such as a live image for a live broadcast, a broadcast name, a broadcast category, and offline ticket information.

The open market linkage unit 120 may store and manage data about affiliated stores registered in the server 100 . The open market interworking unit 120 may receive registration of an affiliated store for the open market, and receive a request for interworking with the server 100 for the affiliated store terminal 400 and the live broadcasting terminal LS. Subsequently, the open market linkage unit 120 may receive location information of the affiliated store, operating hours of the affiliated store, product information, and service information.

In this case, the product information may include the product name, product image, and product price displayed at the affiliated store, and the service information includes the service type (eg, performance, event or lecture), service name, service banner, and service introduction. and the like.

The deep learning unit 130 may also calculate an expected preference for broadcasting by using a deep learning module including user data, past participation data, and artificial neural networks learned based on broadcasting data.

For example, the deep learning unit 130 may receive user data, past participation data, and broadcasting data from the data management unit 110 and the open market linkage unit 120 , respectively.

Subsequently, the deep learning unit 130 may derive a first predicted preference for broadcasting by inputting the received user data, past participation data, and broadcasting data as inputs. In addition, the deep learning unit 130 may select a subject similar to the user based on the user data. In addition, the deep learning unit 130 may derive the second predicted preference of the broadcast for the selected target using the deep learning module.

Then, the deep learning unit 130 compares the received user evaluation and the predicted preference calculated through deep learning (eg, the first expected preference and/or the second expected preference) to re-learn the deep learning module. have. The details of this will be described with reference to FIG. 16 .

Hereinafter, a method of participating in a live broadcast according to some embodiments of the present invention will be described.

3 is a flowchart illustrating a method of participating in a live broadcast according to some embodiments of the present invention.

Referring to FIG. 3 , the server 100 receives broadcast data for a live broadcast ( S110 ).

Specifically, the server 100 may receive a live broadcast image in real time through the live broadcast terminal LS. In addition, the server 100 may receive from the location-based server 300 in real time the location, product information, or service information of the affiliated store where the broadcast is performed.

However, the present invention is not limited thereto, and the server 100 may receive a live broadcast image for a broadcast selected as a recommendation list or received participation by a user.

Subsequently, the server 100 receives user data through the user terminal (S120). The server 100 may receive the user's access IP information, age, gender, location information, user setting information and interest category, and the like.

In this case, the server 100 may receive user data directly from the user through the user terminal 200 , or may receive user data by linking an account of a user registered in SNS or an open market.

Next, the server 100 selects a recommended broadcast by using the location information included in the user data (S130).

Specifically, the server 100 may select a recommended broadcast for each user from the broadcast recommendation candidates based on the user's location information and broadcast data.

For example, the server 100 may derive the user's reception request area based on the user's location information, and determine whether the participation possible area of the broadcast recommendation candidate is included in the reception request area. A broadcast recommendation candidate whose participation possible area is included in the reception request area may be selected as a recommended broadcast.

As another example, the server 100 may calculate the estimated travel time of public transportation from the user's location to the affiliated store location. The server 100 may select a broadcast recommendation candidate in which the calculated estimated movement time is included within a preset reference time as a recommended broadcast.

As another example, the server 100 calculates the user's expected preference for each broadcast recommendation candidate using deep learning based on the user's data, past participation data, and broadcast data, and selects a recommended broadcast based on the predicted preference. have.

A detailed method of selecting a recommended broadcast will be described in detail with reference to FIGS. 5 to 8 and FIGS. 11 to 14 , respectively.

Based on the selected recommendation broadcast, the server 100 provides a recommendation list (S140). In this case, the server 100 may list the recommended broadcasts included in the recommendation list based on the latest order, popularity order, distance order, expected movement time order, or expected preference order.

Subsequently, the server 100 receives a participation signal for a specific broadcast included in the recommendation list (S150). The server 100 may receive a participation signal for a specific broadcast by receiving a touch or click on a specific broadcast included in the recommendation list through the user terminal 200 .

Next, the server 100 displays an image for a specific broadcast and an offline request button (S160).

Specifically, the server 100 may provide the user terminal 200 with a broadcast recorded at an affiliated store in real time. The server 100 may display affiliate store information including a phone number of the affiliate store or a signboard of the affiliate store on the broadcast based on the affiliate store setting information. In addition, the server 100 may display an offline application button on the broadcast.

The server 100 receives a selection for the offline application button displayed on the user terminal 200, and issues an offline ticket for a specific broadcast (S170).

Specifically, when receiving the selection of the offline application button, the server 100 may receive product information or service information for a specific broadcast. The server 100 may generate a barcode or QR code for a product or service. The server 100 may insert the generated barcode or QR code and product information or service information into the offline ticket and transmit it to the user terminal 200 .

Subsequently, the server 100 receives a usage confirmation for the offline ticket (S180).

For example, when a user uses an offline ticket at an affiliated store, the affiliated store terminal 400 may process the offline ticket as used after recognizing the offline ticket, determining the validity period and local information, and the like. The affiliated store terminal 400 may transmit the use completion processing information of the offline ticket to the server 100 .

Subsequently, the server 100 may receive a usage confirmation for the offline ticket based on the use completion processing information for the offline ticket.

4 is a diagram for explaining an example of a method of participating in a live broadcast according to some embodiments of the present invention.

Here, <a1> of FIG. 4 is a diagram displaying a recommendation list provided to the user terminal 200, <a2> is a diagram illustrating a broadcast provided to the user terminal 200, and <a3> is an issued offline It is a drawing for explaining a ticket.

Referring to <a1> of FIG. 4 , the server 100 may generate a recommendation list for a recommended broadcast based on the user's previously derived reception request area, the expected movement time, or the predicted preference using the deep learning module.

For example, the server 100 may provide an apple farm home shopping, a certificate lecture, and a Chinese language school lecture as a recommended broadcast of the recommendation list. The server 100 may provide popularity or predicted preference for each recommended broadcast together with the recommended broadcast. The server 100 may list the recommended broadcasts using the latest order, popularity order, distance order, movement expected time order, or expected preference order based on the user's setting. Hereinafter, it is assumed that the recommendation list is generated by selecting and listing recommended broadcasts based on the expected preference.

If the predicted preference is 4 stars, the Chinese lecture may be displayed as the 3rd priority recommendation broadcast, which is the lowest ranking among the 3 recommended broadcasts. In addition, when the predicted preferences are the same, such as apple farm home shopping and certification lectures, the server 100 may display recommended broadcasts with higher popularity as the highest ranking. That is, if the final expected preference of the apple farm home shopping and the certification lecture is the same, but the popularity of the apple farm home shopping is higher than the certification lecture, the server 100 may display the apple farm home shopping as the first priority recommendation broadcast.

As another example, the server 100 may display a recommendation broadcast having a higher first expected preference derived based on the user's preference as the highest priority. That is, if the final expected preference of the apple farm home shopping and the certification lecture is the same, but the first expected preference of the apple farm home shopping is higher than the certification lecture, the server 100 may display the apple farm home shopping as the first priority recommendation broadcast.

Although not shown in the present invention, the server 100 may display a recommended interest broadcast, a highest viewer rating broadcast, a popular reservation broadcast, live delivery content by region, or a last-order product of an affiliated store in the recommendation list. However, the present invention is not limited thereto.

Subsequently, the server 100 may receive whether the user participates in the recommended broadcast, apple farm home shopping.

Referring to <a2> of FIG. 4 , the server 100 may provide the user terminal 200 with an image and information about the apple farm home shopping received participation from the user.

The server 100 may provide a live broadcast image of the apple farm home shopping, a product price or a merchant phone number, and an offline application button (OB). In this case, the affiliate store information and interface displayed on the broadcast may be set by the affiliate store. The server 100 may or may not display the offline application button OB according to the setting of the affiliate store.

Subsequently, the server 100 may receive the user's selection for the offline application button OB of the apple farm home shopping.

Referring to <a3> of FIG. 4 , the server 100 may issue an offline ticket T1 (hereinafter, referred to as an apple offline ticket) for home shopping at the apple farm based on the user's selection.

Specifically, the server 100 may determine the apple farm home shopping as a product-related broadcast in a broadcasting category (eg, product or service) and issue an apple offline ticket T1 for the apple registered in the apple farm home shopping. . The apple offline ticket T1 may include a barcode for the product, the validity period of the offline ticket, the name of the affiliated store, the location of the affiliated store, and the operating time of the affiliated store.

In this case, the server 100 may issue the apple offline ticket T1 to the user after payment for the product is completed. However, the present invention is not limited thereto, and the offline ticket is a reserved ticket concept, and may be a deferred payment method in which a product reservation is made and payment is made at an affiliated store.

The server 100 may list the issued date of the offline ticket in the latest order for the issued offline ticket and display it on the user terminal 200 . The server 100 may display the previously issued English offline ticket T2 at the bottom of the recently issued apple offline ticket T1. The server 100 may list and display the available offline tickets in order of priority.

Here, the English offline ticket T2 may be an offline ticket issued for service broadcasting. Specifically, when receiving offline participation in a service broadcast (eg, a lecture), the server 100 may receive the remaining seats for a venue of the lecture. Subsequently, the server 100 may allocate a designated seat to the user based on the received remaining seat. The server 100 may issue an English offline ticket T2 including a barcode for the designated seat, the validity period of the offline ticket, the lecture name, the lecture host location, the lecture time, and the seat number.

By providing the issued offline ticket to the corresponding affiliate store terminal 400 , the user can exchange products related to the corresponding offline ticket or receive a related service. In this case, the affiliate store terminal 400 may review whether the offline ticket presented by the user is valid, and determine whether the offline ticket is used. Subsequently, the affiliated store terminal 400 may transmit information about the used offline ticket to the server 100 .

When the use of the offline ticket is completed, the server 100 may display the use completion mark on the offline ticket to process the use of the offline ticket.

That is, the live broadcasting participation method according to the present embodiment may receive a user's application for live broadcasting and issue an offline ticket for using goods or services started in the live broadcasting. Accordingly, the live broadcasting participation method enables the sales of small business type products and intangible contents (workshops and academies, etc.) A virtuous cycle effect can be expected to increase.

Hereinafter, a method of selecting a recommended broadcast based on regional range setting among methods of selecting a recommended broadcast in an embodiment of the present invention will be described.

5 is a flowchart illustrating a method of selecting a recommended broadcast using regional range setting according to an embodiment of the present invention.

Referring to FIG. 5 , the server 100 receives regional range setting information preset by the user ( S131 ).

Specifically, the server 100 may receive a set range (eg, 200 m) of a radius through which a broadcast can be received from the user's location through the user terminal 200 . Alternatively, the server 100 may receive a selection for an area (eg, Gangnam-gu and Seocho-gu, etc.) located in the vicinity of the user's location.

Based on the received regional range setting information, the server 100 derives a reception request area by applying the regional range setting information to the location information (S132).

For example, when receiving the regional range setting information of 200 m, the server 100 may display a radius corresponding to 200 m centered on the user's location. The server 100 may derive the area included in the displayed radius as the reception request area.

Then, the server 100 receives the participation area for each broadcast recommendation candidate from the location-based server 300 (S133). The participating area may be set by the affiliate that provides the broadcasting recommendation candidate.

Next, the server 100 determines whether the participation request area is included in the reception request area (S134). However, the present invention is not limited thereto, and the server 100 may receive the area coverage rate based on the user's input. The server 100 may determine whether the participation possible area has a higher inclusion rate than a preset area inclusion rate for the reception request area.

Next, if the reception request area does not include the participation area, the server 100 does not select the corresponding broadcast as a recommended broadcast (S135).

On the other hand, when the reception request area includes a participation area, the server 100 selects the corresponding broadcast as a recommended broadcast (S136). Subsequently, the server 100 may generate a recommendation list based on the selected recommendation broadcast.

FIG. 6 is a diagram for explaining an example of selecting a recommended broadcast using the regional range setting of FIG. 5 .

<b1> of FIG. 6 is a diagram illustrating a regional range setting interface, and <b2> is a diagram illustrating a recommended broadcast selection interface.

Referring to <b1> of FIG. 6 , the server 100 may receive setting information of a radius range through a regional range setting interface.

Specifically, the server 100 may display the user's location UP on the map based on the user's location information. Subsequently, the server 100 may receive the radius range setting information through the radius range setting interface AP (hereinafter, the radius interface). The radius interface AP may receive a set value of the radius range and a unit of the radius range from the user.

For example, the server 100 may receive a radius range of 100 m through the radius interface AP. The server 100 may derive a radius having a radius of 100 m based on the user location UP, based on the received radius range. Based on the derived radius, the server 100 may display the reception request area VA for the user location UP on the map.

Referring to <b2> of FIG. 6 , the server 100 may select a recommended broadcast for each user through a recommended broadcast selection interface.

The server 100 may receive the user's reception request area (VA) set through the area range setting interface. In addition, the server 100 may receive a participation area for each broadcast recommendation candidate from the location-based server 300 .

At this time, the server 100 may receive the radius range setting information of the affiliated store with respect to the affiliated store location information from the affiliated store, and may derive the participation possible area based on the received radius range setting information. However, the present invention is not limited thereto.

Subsequently, the server 100 may display the reception request area (VA) and the first to fourth available areas (AA, BA, CA, and DA) on the map of the recommended broadcast selection interface. The server 100 compares the displayed reception request area (VA) and the first to fourth possible participation areas (AA, BA, CA, and DA), and participates in each broadcast for the reception request area (VA) It is possible to determine whether a possible area is included or not.

For example, the area of the first available participation area (AA) set for the broadcast A may be all included in the reception request area (VA) of the user. Accordingly, the server 100 may select the broadcast A of the first participation possible area (AA) as a recommended broadcast. The user's reception request area (VA) may be entirely included in the area of the second possible participation area (BA) of the B broadcast. Accordingly, the server 100 may select the B broadcast as a recommended broadcast.

In addition, since the third participating area CA is not included in the reception request area VA, the server 100 may not select broadcast C as a recommended broadcast. The fourth available area DA may be determined to overlap the reception request area VA and a part of the area. That is, the server 100 may select broadcast D as a recommended broadcast. However, this is only an example, and the present invention is not limited thereto.

Hereinafter, a method of selecting a recommended broadcast based on an estimated moving time among methods of selecting a recommended broadcast in an embodiment of the present invention will be described. Hereinafter, portions overlapping with the above-described embodiments will be simplified or omitted.

7 is a flowchart for explaining a method of selecting a recommended broadcast using public transportation operation information according to another embodiment of the present invention. Hereinafter, for convenience of explanation, the user using public transportation such as subway or bus among transportation means will be described as an example.

Referring to FIG. 7 , the server 100 derives the nearest subway station or bus stop to the user's location information (S231).

In this case, the server 100 may derive a subway station or a bus stop (hereinafter referred to as a boarding location) located near the user by using the user's location information. The server 100 may calculate the walking time or straight-line distance required from the user's location to each subway station or bus stop. The server 100 may derive a subway station or bus stop having the shortest calculated walking time or linear distance as the nearest subway station or bus stop.

Then, the server 100 receives the derived operation information of the subway station or bus stop (S232).

Specifically, the server 100 may receive operation information for all subways or buses operating at a subway station or bus stop derived from a traffic server (not shown). Here, the operation information may include subway route information, subway departure time, subway transfer information, bus route information, each bus departure time or bus transfer information, and the like.

Based on the received operation information, the server 100 calculates an estimated movement time from the user's location to the broadcast transmission location (S233).

For example, the server 100 may set the start time of selection of the recommended broadcast as the user's departure time. The server 100 may calculate the estimated travel time by summing the boarding walking time from the user's location to the subway station or bus stop, the subway or bus operating time, and the getting off walking time walking to the broadcast location after getting off.

In this case, the server 100 may receive traffic information from a traffic server (not shown). The server 100 may calculate the bus travel time based on the traffic information. Here, the traffic information may include road information and driving speed.

Next, the server 100 checks whether the calculated estimated movement time is included within a preset reference time (S234). Here, the reference time may be a preset moving time that is a criterion for determining whether to select a broadcast recommendation candidate as a recommendation candidate. The reference time may include a unit of expected movement time, a type of movement time (eg, one-way time or round-trip time) for calculation of the expected movement time, and a preferred estimated movement time.

Subsequently, if the estimated movement time is not included within the reference time, the server 100 does not select the corresponding broadcast as a recommended broadcast, and if the estimated movement time is included within the reference time, the server 100 selects the broadcast as a recommended broadcast is selected (S235, S236).

Additionally, in another embodiment of the present invention, of course, the server 100 may select a recommended broadcast using the estimated travel time of the private vehicle. In this case, the server 100 may receive real-time traffic information from a traffic server (not shown). The server 100 may calculate the estimated travel time of the private vehicle from the user's location to the broadcast transmission location based on the traffic information. When the calculated estimated moving time of the private vehicle is included within the preset moving estimated time, the server 100 may select the corresponding broadcast as a recommended broadcast.

8 is a diagram for explaining an example of selecting a recommended broadcast using the public transportation operation information of FIG.

<c1> of FIG. 8 is a diagram illustrating an interface for setting an expected movement time, and <c2> indicates an interface for calculating an estimated movement time using a bus. In addition, <c3> represents an interface for calculating the estimated travel time using the subway.

Referring to <c1> of FIG. 8 , the server 100 may receive estimated movement time setting information from the user through an expected movement time setting interface (hereinafter, an estimated movement time interface).

Specifically, the server 100 may receive the movement means information, which is a calculation standard for the estimated movement time, through the estimated movement time interface. The server 100 may provide public transportation, a private car, and walking as a means of transportation, and may receive a selection of a means of transportation through the user terminal 200 .

When an input for public transportation is received, the server 100 may calculate an estimated travel time using a bus or subway as a means of transportation. When receiving an input for a private vehicle, the server 100 may receive traffic information from a traffic server (not shown) and calculate an estimated travel time. In addition, when receiving the selection for walking through the user terminal 200, the server 100 may calculate the shortest distance from the user's location to the broadcast transmission location, and calculate the estimated movement time for this.

Then, the server 100 may calculate the estimated movement time for the received movement means. In this case, the server 100 may receive reference time information from the user. Here, the reference time information may include the type of expected movement time (for example, one-way time or round-trip time), the unit of the expected movement time (hours or minutes, etc.), and the preferred estimated movement time. have. The server 100 may convert the estimated movement time calculated based on the received reference time information and display it on the user terminal 200 .

Hereinafter, when the server 100 receives public transportation as a means of transportation, an example of calculating the estimated travel time for a bus and a subway will be described. Hereinafter, it is assumed that the reference time is set to 30 minutes one way.

Referring to <c2> of FIG. 8 , the server 100 may calculate an estimated travel time for a bus during public transportation.

For example, the server 100 may calculate a walking time or a straight-line distance required to a nearby bus stop based on the user's location UP. The server 100 may derive the first bus stop BS1 from which the shortest walking time is calculated from the user location UP as the reference bus stop.

Subsequently, the server 100 may receive operation information for all buses operated at the first bus stop BS1. However, the present invention is not limited thereto, and only operation information for a bus including a bus stop that stops around a broadcast transmission location may be received.

Then, the server 100 may calculate the shortest expected movement time to the transmission location of the broadcast for each bus based on the received operation information.

Specifically, the server 100 may derive a bus stop located closest to the broadcast transmission location for each bus. The server 100 may derive the second bus stop BS2 as the bus stop closest to the transmission location EP of the E broadcast, and the third bus stop as the bus stop closest to the transmission location FP of the F broadcast. (BS3) can be derived.

The server 100 may calculate the bus operation time required from the first bus stop BS1 to the second bus stop BS2 and the third bus stop BS3, respectively, based on traffic information and transfer information. In addition, the server 100 can calculate the get-off walking time required from the second bus stop (BS2) and the third bus stop (BS3) to the transmission location (EP) of the E broadcast and the transmission location (FP) of the F broadcast. have.

That is, the server 100 may calculate the shortest expected travel time for each broadcast using the bus by summing the previously calculated boarding walking time, bus running time, and getting off walking time.

The server 100 may calculate 40 minutes one way as the estimated movement time for broadcast E, and may calculate 30 minutes one way as the estimated travel time for broadcast F. Based on the pre-received reference time of 30 minutes one way, the server 100 may not select the E-broadcast in which the estimated movement time exceeds 30 minutes one way as the recommended broadcast. In addition, the server 100 may select an F broadcast having an estimated travel time of 30 minutes or less one way as a recommended broadcast.

Referring to <c3> of FIG. 8 , the server 100 may calculate an estimated travel time for the subway during public transportation. Hereinafter, content overlapping with the content of the above-described embodiment of the present invention will be omitted and described.

For example, the server 100 may derive the first subway station SS1 for which the shortest straight distance or walking time is calculated based on the user location UP as the reference subway station.

Subsequently, the server 100 may receive operation information for all subways operated in the first subway station SS1. However, the present invention is not limited thereto, and it is possible to receive only operation information for a subway including a subway station that stops in the vicinity of a broadcast transmission location.

Then, the server 100 may calculate the shortest expected movement time to the transmission location of the broadcasting for each subway based on the received operation information.

Specifically, the server 100 may derive a bus stop located closest to the broadcast transmission location for each subway. The server 100 may derive the second subway station SS2 for the transmission location EP of the E broadcast, and may derive the third subway station SS3 for the transmission location FP of the F broadcast.

The server 100 may calculate the subway operation time required from the first subway station SS1 to the second subway station SS2 and the third subway station SS3, respectively. In addition, the server 100 may calculate the get-off walking time required from the second subway station SS2 and the third subway station SS3 to the transmission location of each broadcast.

The server 100 may calculate the shortest estimated travel time for each broadcast by adding up the boarding walking time, subway operating time, and getting off walking time.

The server 100 may calculate 28 minutes one way as the estimated movement time for the E broadcast, and may calculate 32 minutes one way as the expected movement time for the F broadcast. Based on the pre-received reference movement estimated time, the server 100 may select the E broadcast as the recommended broadcast and not select the broadcast as the recommended broadcast.

That is, according to the present invention, it is possible to select a recommended broadcast available in the area where the user is located based on the user's location information. Subsequently, it is possible to provide a recommendation list including the selected recommended broadcast. Accordingly, the live broadcast participation method selects a recommended broadcast based on the user's region-based exposure, thereby increasing users' access to merchants selling services and products, and obtaining a marketing effect that can lead to face-to-face sales.

Hereinafter, a method for issuing an offline ticket according to some embodiments of the present invention will be described.

9 is a flowchart illustrating an offline ticket issuance method according to some embodiments of the present invention.

Referring to FIG. 9 , in step S170 of FIG. 3 described above, the server 100 determines whether a specific broadcast is a broadcast related to a product or a broadcast related to a service ( S171 ).

Specifically, the server 100 may determine a specific broadcast as a product-related broadcast or a service-related broadcast based on the broadcast name or category information of the broadcast included in the second broadcast data registered by the affiliated store. In addition, the server 100 may determine based on information (product or service) sold or transmitted in a specific broadcast.

Next, when a specific broadcast is a broadcast related to a product, the server 100 checks the inventory of the product transmitted to the specific broadcast (S173a).

For example, the server 100 may check the stock based on product information previously registered in the affiliated store terminal 400 . As another example, the server 100 may transmit a “check whether stock” interface to the affiliated store terminal 400 and receive whether it is in stock from the affiliated store.

On the other hand, when a specific broadcast is a broadcast related to a service, the server 100 checks the number of people allowed to enter the service (S173b). The server 100 may calculate the number of people allowed to enter based on the number of occupants information and the number of people registered in advance in the affiliated store terminal 400 .

In this case, when a specific broadcast is a broadcast for a plurality of goods and services, step S173b may be performed in parallel with step S173a described above.

Subsequently, if there is an available number of people, the server 100 selects a seat number based on the remaining seats (S175).

In this case, the server 100 may derive the remaining seat information by using the total seat information and the seat information of the number of visitors. The server 100 may randomly select a seat number based on the remaining seat information. Alternatively, the remaining seat information may be provided to the user to receive a desired seat number from the user.

Then, the server 100 issues an offline ticket for the product or service (S177). The server 100 may issue to the user terminal 200 an offline ticket including a barcode or QR code in which purchase information for a product or a seat number for a service is written.

Hereinafter, a method of receiving product purchase or event participation using the issued offline ticket and processing the received offline ticket will be described.

10 is a flowchart illustrating a method of receiving use of an offline ticket according to some embodiments of the present invention.

Referring to FIG. 10 , in step S180 of FIG. 3 , the affiliate store terminal 400 recognizes an offline ticket ( S181 ).

Next, the affiliated store terminal 400 checks the validity period of the recognized offline ticket (S183). The affiliated store terminal 400 may compare the validity date included in the offline ticket with the recognition date of the offline ticket. When the recognition date of the offline ticket is included within the expiration date, the affiliated store terminal 400 may complete the validation of the expiration date.

Next, the affiliate store terminal 400 determines whether the available area of the offline ticket is included in the user's reception request area (S185).

In this case, the usable area of the offline ticket may be included in the available area of participation in a specific broadcast to which the offline ticket is issued. The affiliated store terminal 400 may determine whether to include the available area included in the offline ticket by comparing the preset user's reception request area. If the usable area is not included in the reception request area, the affiliated store terminal 400 may return to step S181 and repeat steps S181 to S185.

Subsequently, when the usable area is included in the reception request area, the affiliated store terminal 400 may process the offline ticket as used. The affiliated store terminal 400 may transmit the use completion processing information of the offline ticket to the server 100 .

Then, based on the received processing information of the use of the offline ticket, the server 100 receives the confirmation of use of the offline ticket (S187). The server 100 may recognize the use completion and process the use completion for the offline ticket for which the use confirmation is received.

Hereinafter, a method of selecting a recommended broadcast using a deep learning module according to another embodiment of the present invention will be described with reference to FIGS. 11 to 14 . Parts overlapping with the above-described embodiment will be simplified or omitted.

11 is a flowchart for explaining a method of selecting a recommended broadcast using a deep learning module according to another embodiment of the present invention.

Referring to FIG. 11 , in step S130 of FIG. 3 , the server 100 also calculates an expected preference for a broadcast recommendation candidate using a deep learning module based on user data and past participation data ( S341 ).

Specifically, the server 100 may inquire past participation data of the user, which is the details of participation in broadcasting of the live broadcasting system, based on the received user data. In addition, past participation data may include participation details for other systems.

Subsequently, the server 100 may derive the user's first expected preference for the broadcast recommendation candidate. Also, the server 100 may calculate a second predicted preference for each broadcast recommendation candidate for a target similar to the user. The server 100 may derive the final predicted preference by using the first predicted preference and/or the second predicted preference. The details of this will be described with reference to FIG. 14 .

Next, the server 100 selects a recommended broadcast based on the derived expected preference (S343). In this case, the server 100 may select, as a recommended broadcast, a broadcast in which the derived expected preference is greater than the preset standard, based on the preset standard of the expected preference.

Next, the server 100 provides a list of recommendations based on the selected recommended broadcast (S345).

Here, the server 100 may generate a recommendation list by arranging the selected recommended broadcasts in descending order based on the derived expected preference. The server 100 may display the generated recommendation list on the user terminal 200 .

Subsequently, the server 100 receives a participation signal for a specific broadcast included in the recommendation list (S347).

Next, the server 100 receives an evaluation for a specific broadcast, and re-learns the deep learning module based on the evaluation (S349). For example, the server 100 may compare the user evaluation with the predicted preference calculated in advance, and if the comparison result is out of a predetermined error range, the deep learning module may be re-learned.

That is, it is possible to derive an expected preference for each user using a deep learning module based on the live broadcasting participation method of the present invention, user data, and past order data, and generate a recommendation list for broadcasting based on the derived expected preference. . Through this, the present invention can select and recommend broadcasts reflecting the user's taste and interests, and improve user satisfaction.

12 is a block diagram schematically illustrating a deep learning module used in the method of selecting a recommended broadcast of FIG. 11 .

Specifically, referring to FIG. 12 , the deep learning module DM may receive user data, past participation data, and/or broadcast data, and output the user's expected preference as an output thereof. Here, the user data may include the user's age, gender, location information, or interest category. In addition, the broadcast data may be data for a live broadcast transmitted by a plurality of affiliated stores.

In this case, the deep learning module (DM) may store data in a memory based on parameters for user data, past participation data, and broadcast data, or classify similar data according to categories. However, the aforementioned user data, past participation data, and broadcast data are only examples of input parameters input to the deep learning module (DM), and the input data applied to the deep learning module (DM) are variously added or changed. can be used.

Hereinafter, for convenience of explanation, user data, past participation data, and broadcast data are applied to the input terminal of the deep learning module (DM), and an expected preference is derived as an output thereof.

Then, the deep learning module (DM) may derive the expected preference required for the live broadcast participation method using the artificial neural network learned based on big data.

To explain in more detail, Deep Learning (DeEP Learning) technology, a type of machine learning, learns by going down to a deep level in multiple stages based on data. That is, deep learning (DeEP learning) represents a set of machine learning algorithms that extract core data from a plurality of data while increasing the level.

The deep learning module (DM) may use various known deep learning structures. For example, the deep learning module (DM) may use a structure such as a Convolutional Neural Network (CNN), a Recurrent Neural Network (RNN), a DeEP Belief Network (DBN), or a GrAPh Neural Network (GNN).

Specifically, CNN (Convolutional Neural Network) is a human brain function created based on the assumption that when a person recognizes an object, it extracts the basic features of the object, then performs complex calculations in the brain and recognizes the object based on the result. It is a simulated model.

RNN (Recurrent Neural Network) is widely used for natural language processing, etc., and is an effective structure for processing time-series DAta that changes with time.

DBN (DeEP Belief Network) is a deep learning structure composed of multi-layered Restricted Boltzman Machine (RBM), a deep learning technique. When a certain number of layers is obtained by repeating Restricted Boltzman Machine (RBM) learning, a DBN (DeEP Belief Network) having a corresponding number of layers may be configured.

GNN (GrAPhic Neural Network, hereinafter referred to as GNN) represents an artificial neural network structure implemented in a way that similarities and feature points between modeling data are derived using modeling data modeled based on data mapped between specific parameters. .

On the other hand, artificial neural network learning of the deep learning module (DM) can be made by adjusting the weight of the connection line between nodes (and adjusting the bias value if necessary) so that a desired output is obtained for a given input. Also, the artificial neural network may continuously update a weight value by learning. In addition, a method such as back propagation may be used for learning the artificial neural network.

On the other hand, the memory of the content management server 100 may be loaded with an artificial neural network (Artificial Neural Network) pre-trained by machine learning.

That is, the deep learning module (DM) may perform an operation of outputting an expected preference for broadcasting by using user data, past participation data, and broadcasting data as input data based on machine learning.

In this case, both semi-supervised learning and supervised learning may be used as the machine learning method of the artificial neural network. In addition, the deep learning module (DM) may be controlled to automatically update the artificial neural network structure for outputting the predicted preference after learning according to the setting.

Additionally, although not clearly shown in the drawings, in another embodiment of the present invention, the operation of the deep learning module (DM) may be performed in the content management server 100 or a separate cloud server (not shown). Hereinafter, an example of the deep learning module (DM) described above will be described.

13 is a diagram illustrating the configuration of the deep learning module of FIG. 12 .

Referring to FIG. 13 , the deep learning module (DM) includes an input layer using user data, past participation data, and/or broadcasting data as input nodes, and an output layer using expected preferences of the user as an output node. and M hidden layers disposed between the input layer and the output layer.

Here, a weight may be set on an edge connecting the nodes of each layer. The presence or absence of such weights or edges may be added, removed, or updated during the learning process. Accordingly, through the learning process, weights of nodes and edges disposed between k input nodes and i output nodes may be updated.

Before the deep learning module (DM) performs learning, all nodes and edges can be set to initial values. However, when cumulative information is input, the weights of nodes and edges are changed, and in this process, parameters input as learning factors (that is, user data, past participation data and/or broadcasting data) and assigned to output nodes A match can be made between values (ie, expected preferences).

Additionally, when using a cloud server (not shown), the deep learning module (DM) may receive and process a large number of parameters. Therefore, the deep learning module (DM) can perform learning based on massive data.

In addition, the weights of nodes and edges between the input node and the output node constituting the deep learning module (DM) may be updated by the learning process of the deep learning module (DM). In addition, it goes without saying that the parameters output from the deep learning module (DM) can be further extended to various data in addition to the predicted preference.

The server 100 may generate a recommendation list using the predicted preference derived from the deep learning module (DM). In this case, the predicted preference may be post-processed and used in various ways.

Hereinafter, an embodiment of the present invention for post-processing the predicted preference for broadcasting will be described.

14 is a flowchart for explaining an example of calculating an expected preference for broadcasting in step S343.

Referring to FIG. 14 , the server 100 derives a first predicted preference of a broadcast recommendation candidate for a user by using a deep learning module (S343_1).

Additionally, the server 100 may calculate the first expected preference by reflecting the number of participation in broadcasting, the number of participation in each broadcasting category, and rating included in the user's past participation data. Also, the server 100 may derive the first predicted preference by weighting the expected preference of broadcasting corresponding to the interest category of the user data.

Next, the server 100 selects a subject having similar user data (S343_3). In this case, the server 100 may calculate and use the similarity for each subject with respect to the age, gender, and interest category included in the user information. For example, the server 100 may select subjects whose calculated similarity is higher than a preset reference value (eg, 70%).

Then, the server 100 derives the second predicted preference of the broadcast recommendation candidate for the selected target using the deep learning module (S343_5).

Specifically, the server 100 may derive the predicted preference for each selected subject using a deep learning module based on the user data and past participation data of the selected subject. The server 100 may derive the average value of the predicted preference for each selected subject as the second predicted preference.

Next, the server 100 derives a final predicted preference based on the first predicted preference and the second predicted preference ( S343_7 ).

In this case, the server 100 may apply different weights to the first expected preference and the second expected preference. For example, the server 100 may give a weight of 0.7 to the first predicted preference and apply a weight of 0.3 to the second predicted preference. In addition, weights applied to the first predicted preference and the second predicted preference may be changed by the user. The server 100 may derive the final predicted preference of broadcasting by adding the weighted first predicted preference and the second predicted preference. However, the present invention is not limited thereto, and the final expected preference for broadcasting may be derived using various methods.

Next, the server 100 selects a recommended broadcast based on the derived final predicted preference (S343_9).

That is, in the live broadcast participation method according to another embodiment of the present invention, the final predicted preference is derived using the first predicted preference reflecting the user's preference and the second expected preference of subjects similar to the user, thereby meeting the user's taste. It is possible to improve the accuracy of the selection of the recommended broadcasting.

15 is a flowchart illustrating an interface for receiving a user's evaluation in a method of participating in a live broadcast according to another embodiment of the present invention.

Referring to FIG. 15 , when the broadcast ends or the user's offline ticket issuance or use is completed, the server 100 may provide a user review interface to the user terminal 200 . The user review interface may display a user evaluation menu including an evaluation window for broadcast content, transmission quality, and ticket use.

Subsequently, when the user evaluation is completed, the server 100 may relearn the deep learning module based on the user evaluation.

Hereinafter, a re-learning method of a deep learning module according to some embodiments of the present invention will be described.

16 is a block diagram schematically illustrating a re-learning step of a deep learning module used in a method of participating in a live broadcast according to another embodiment of the present invention.

Referring to FIG. 16 , the server 100 may calculate a comparison result by comparing the user evaluation with the expected preference derived for a specific broadcast selected by the user. When the calculated comparison result is larger than a preset error range (eg, ±30%), the server 100 may classify the broadcast selected by the user as a learning factor that is a re-learning target.

Subsequently, the deep learning module (DM) may receive user data, past participation data, and broadcast data selected by the user as input to the input node, and receive user evaluation for this input to the output node.

Subsequently, the deep learning module DM may perform machine learning based on each parameter input as a learning factor. In this case, data used for machine learning and result data may be stored in the memory of the content management server 100 .

Accordingly, the content management server 100 may update the weights for the nodes and edges of the deep learning module (DM) through re-learning.

In summary, the live broadcasting participation method according to some embodiments of the present invention can improve the accuracy of the broadcasting preferred by the user by continuously re-learning the deep learning module by reflecting the user's broadcasting selection. Through this, the present invention can improve the reliability of the service provided to the user.

The above description is merely illustrative of the technical idea of this embodiment, and a person skilled in the art to which this embodiment belongs may make various modifications and variations without departing from the essential characteristics of the present embodiment. Accordingly, the present embodiments are intended to explain rather than limit the technical spirit of the present embodiment, and the scope of the technical spirit of the present embodiment is not limited by these embodiments. The protection scope of this embodiment should be interpreted by the following claims, and all technical ideas within the equivalent range should be interpreted as being included in the scope of the present embodiment.

Claims (10)

A method of participating in a live broadcast performed in a live broadcast system associated with a user terminal, a live broadcast terminal, and a merchant terminal, the method comprising:
receiving broadcast data for a live broadcast from the live broadcast terminal;
receiving user data including user location information from the user terminal;
selecting a recommended broadcast to be provided to the user terminal based on the user data;
providing a recommendation list including the recommended broadcast;
receiving a participation signal for a specific broadcast from the user terminal among a plurality of recommended broadcasts included in the recommendation list;
displaying an image for the specific broadcast and an offline application button for the specific broadcast on the user terminal;
receiving a selection for the offline application button from the user terminal, and issuing an offline ticket for the specific broadcast; and
Receiving a usage confirmation for the offline ticket from the affiliated store terminal
How to join a live broadcast.
The method of claim 1,
The step of selecting a recommended broadcast to be provided to the user terminal comprises:
Receives preset regional range setting information input to the user terminal,
Deriving a reception request area of the user for broadcasting by applying the area range setting information to the location information,
Receive the possible participation area of the broadcast recommendation candidate being broadcast live,
Selecting the broadcast recommendation candidate as the recommended broadcast when the participation request area is included in the reception request area
How to join a live broadcast.
The method of claim 1,
The step of selecting a recommended broadcast to be provided to the user terminal comprises:
Deriving the nearest subway station or bus stop to the location information,
Receive operation information of the subway station or the bus stop,
Calculating an estimated moving time from the user's location to the transmission location of the broadcast recommendation candidate based on the operation information,
When the calculated estimated moving time is included within a preset reference time, selecting the broadcast recommendation candidate as the recommended broadcast
How to join a live broadcast.
The method of claim 1,
The step of issuing the offline ticket includes:
If the specific broadcast is a broadcast related to the sale of a product, check the inventory of the product and issue an offline ticket for exchanging the product;
If the specific broadcast is a broadcast related to a performance or event, confirming the number of people allowed to enter the performance or event, selecting the seat number of the user based on the remaining seats, and issuing an offline ticket
How to join a live broadcast.
The method of claim 1,
Receiving the usage confirmation for the offline ticket comprises:
Recognizing the offline ticket displayed on the user terminal through the affiliated store terminal in the affiliate store for the specific broadcast,
Check the validity period of the offline ticket,
Comprising checking whether the available area of the offline ticket is included in the reception request area of the user terminal,
The usable area includes a participation area capable of receiving the specific broadcast.
How to join a live broadcast.
The method of claim 1,
The step of selecting a recommended broadcast to be provided to the user terminal comprises:
Based on the user data and the user's past participation data, a predicted preference for a broadcast recommendation candidate is calculated using a pre-trained deep learning module,
Selecting the broadcast recommendation candidate whose calculated expected preference is higher than a reference value or is included in a preset high rank number as the recommended broadcast
How to join a live broadcast.
7. The method of claim 6,
Calculating the predicted preference is
Derive a first predicted preference of the specific broadcast for the user using the deep learning module,
Selecting subjects whose similarity between the user and the user data is higher than a reference value,
Derive a second predicted preference of the specific broadcast for the selected target using the deep learning module,
and deriving a final predicted preference based on the first predicted preference and the second predicted preference.
How to join a live broadcast.
7. The method of claim 6,
The deep learning module,
an input layer using the user data, the past participation data, and the broadcast data as input nodes;
an output layer using the predicted preference as an output node;
one or more hidden layers disposed between the input layer and the output layer;
The weights of nodes and edges between the input node and the output node are updated by the learning process of the deep learning module.
How to join a live broadcast.
9. The method of claim 8,
Receiving the user's evaluation of the specific broadcast from the user terminal;
Further comprising the step of re-learning the deep learning module using the received evaluation,
The step of re-learning the deep learning module,
comparing the predicted preference and the evaluation for the specific broadcast;
When the comparison result is out of a predetermined error range, comprising re-learning the deep learning module
How to join a live broadcast.
9. The method of claim 8,
In the step of re-learning the deep learning module,
The deep learning module is configured to update the weights of the nodes and edges by applying the user data, the past participation data, and broadcasting data in which the user participated to the input node, and the evaluation is applied to the output node
How to join a live broadcast.

Images