KR101863672B1 - Method and apparatus for providing user customized multimedia contents based on multimedia contents information - Google Patents

Abstract

The present invention relates to a method and a device for providing user customized multimedia content based on multimedia content information. The method for providing user customized multimedia content based on multimedia content information comprises: a step in which a multimedia content service server receives a request for a multimedia content service application; and a step in which the multimedia content service server transmits the multimedia content service application directly or through a relay server. In addition, the multimedia content service application provides a music recommending service to a multimedia content receiver corresponding to a receiving terminal. The music recommending service recommends music which is matched with music listening taste information of the multimedia content receiver. The music listening taste information is able to be generated based on multimedia content information of music which is listened by the multimedia content receiver.

Description

[0001] The present invention relates to a method and apparatus for providing user-customized multimedia contents based on multimedia contents information,

The present invention relates to a method and apparatus for providing user-customized multimedia contents based on multimedia contents information. More particularly, the present invention relates to a method and apparatus for generating multimedia contents information for providing multimedia contents adaptively to a user by combining a real time music broadcasting service and a streaming radio service, and providing user-customized multimedia contents based on multimedia contents information .

The development of wireless Internet and mobile terminals has enabled users to easily provide various kinds of services such as games, news, music, digital broadcasts, etc. anytime and anywhere with the increase of the number of users, and the amount of information consumed exponentially have. In order to efficiently utilize such a wide variety of information, it is important that the user can easily access desired information anytime and anywhere.

It is important to quickly provide users with information that the user is interested and wants to find out of the overflowing information. When the information that the user is interested in is accurately conveyed to the user, the user's information searching time may be reduced.

Currently, many personalized broadcast services are being provided through networks. Even though it is not a large broadcasting station, a multimedia contents consumption structure in which many users directly generate multimedia contents and other users consume the generated multimedia contents is formed.

Currently, there are various technologies for providing customized multimedia contents to users. However, it can not be said that the recommendation accuracy of the present technology used to accurately provide custom broadcasting, customized multimedia contents, etc. to users is high. Therefore, the user does not utilize the recommendation system much, but often uses the multimedia contents through a separate direct search instead of the recommended multimedia contents through the recommendation system.

It is an object of the present invention to solve all the problems described above.

It is another object of the present invention to provide multimedia contents suited to a user's taste more precisely without selecting a user's separate multimedia contents in consideration of user information, multimedia content characteristic information, and the like.

In addition, the present invention provides a method and system for providing a multimedia streaming service in which a response of a multimedia content recipient to music provided through a broadcasting service and a radio streaming service and user information of a receiver of multimedia contents are fed back in real time to be reflected in multimedia content characteristics information and user information, It is another object of the present invention to provide a multimedia content recipient with a multimedia content suited to the taste of the multimedia content recipient.

In order to accomplish the above object, a representative structure of the present invention is as follows.

According to one aspect of the present invention, a method of providing user-customized multimedia content comprises the steps of: receiving a request for a multimedia content service application from a transmitting terminal and / or a receiving terminal, wherein the multimedia content service server directly Or transmitting the multimedia content service application to the transmitting terminal and the receiving terminal through the relay server, wherein the multimedia content application provides a music recommendation service to the receiver of the multimedia contents corresponding to the receiving terminal, Wherein the music recommendation service recommends music that matches the music listening preference information of the multimedia content recipient, and the music listening preference information is multimedia multimedia content of the music listener It can be generated based on the content information.

According to another aspect of the present invention, a multimedia content service server for providing customized multimedia content includes a communication unit implemented for communication with a transmitting terminal and / or a receiving terminal, and a processor operatively coupled to the communication unit, , The processor is configured to receive a request for a multimedia content service application from the transmitting terminal and / or the receiving terminal and to transmit the multimedia content service application to the transmitting terminal and the receiving terminal directly or through a relay server Wherein the multimedia content application provides a music recommendation service to a multimedia content recipient corresponding to the receiving terminal, and the music recommendation service matches a music listening preference information of the multimedia content recipient And the music listening preference information may be generated on the basis of the multimedia content information of the music listened to by the receiver of the multimedia contents.

According to the present invention, multimedia content suitable for a user's taste can be more accurately provided without selecting a user's separate multimedia contents considering user information, multimedia content characteristic information, and the like.

In addition, the response of the receiver of the multimedia contents to the music provided through the broadcasting service and the radio streaming service and the user information of the receiver of the multimedia contents are fed back in real time to be reflected in the multimedia contents information and the user information, The multimedia contents corresponding to the multimedia contents can be provided to the receiver of the multimedia contents.

In addition, music corresponding to the service providing environment of the receiver of the multimedia contents can be provided to the receiver of the multimedia contents without setting the receiver of the multimedia contents separately.

1 is a conceptual diagram of a multimedia content recommendation system according to an embodiment of the present invention.
2 is a conceptual diagram of a multimedia content recommendation system according to an embodiment of the present invention.
3 is a conceptual diagram illustrating a specific operation of the multimedia contents service server according to the embodiment of the present invention.
4 is a conceptual diagram illustrating a specific operation of the multimedia contents server according to the embodiment of the present invention.
5 is a conceptual diagram illustrating a recommendation operation for a specific broadcast service / streaming radio service of a multimedia content server according to an embodiment of the present invention.
6 is a conceptual diagram illustrating a recommendation operation for a specific broadcasting service / streaming radio service of a multimedia contents server according to an embodiment of the present invention.
FIG. 7 is a conceptual diagram illustrating a recommendation operation for a specific broadcasting service / streaming radio service of a multimedia contents service server according to an embodiment of the present invention.
FIG. 8 is a conceptual diagram illustrating a recommendation operation for a specific broadcasting service / streaming radio service of a multimedia contents service server according to an embodiment of the present invention.
9 is a conceptual diagram illustrating a method of generating multimedia content information according to an embodiment of the present invention.
10 is a conceptual diagram illustrating a method of generating multimedia content information according to an embodiment of the present invention.
11 is a conceptual diagram illustrating a method of generating multimedia content information according to an embodiment of the present invention.
12 is a conceptual diagram illustrating a learning method of a multimedia contents service server according to an embodiment of the present invention.
13 is a conceptual diagram illustrating a method of generating multimedia content information according to an embodiment of the present invention.

The following detailed description of the invention refers to the accompanying drawings, which illustrate, by way of illustration, specific embodiments in which the invention may be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the invention. It should be understood that the various embodiments of the present invention are different, but need not be mutually exclusive. For example, the specific shapes, structures, and characteristics described herein may be implemented by changing from one embodiment to another without departing from the spirit and scope of the invention. It should also be understood that the location or arrangement of individual components within each embodiment may be varied without departing from the spirit and scope of the present invention. Therefore, the following detailed description is not to be taken in a limiting sense, and the scope of the present invention should be construed as encompassing the scope of the appended claims and all equivalents thereof. In the drawings, like reference numbers designate the same or similar components throughout the several views.

Hereinafter, various embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art can easily carry out the present invention.

Hereinafter, in an embodiment of the present invention, music is disclosed as an example of multimedia contents for convenience of explanation. However, a recommendation method for a service based on multimedia contents other than music can also be performed by a method and apparatus for providing user-customized multimedia contents disclosed in embodiments of the present invention, .

1 is a conceptual diagram of a multimedia content recommendation system according to an embodiment of the present invention.

1, a multimedia content recommendation system for providing a user with a customized broadcast service (hereinafter referred to as a broadcast service) / customized streaming radio service (hereinafter referred to as a streaming radio service) is disclosed.

In the multimedia content recommendation system according to the embodiment of the present invention, the broadcasting service 100 and the streaming radio service 150 may be adaptively provided to the user.

The broadcast service 100 may be a personal music broadcast service. For example, the broadcasting service 100 may provide a broadcasting service for broadcast contents of a single creator or a medium-sized creator in the form of a multi-channel network (MCN) service. Specifically, a user (hereinafter referred to as a multimedia content provider) serving as a DJ (disc jockey) may exist for the broadcast service 100 and a user (hereinafter referred to as a multimedia content receiver) who listens to broadcasts transmitted by the multimedia content provider Can exist. The receiver of the multimedia contents may receive and provide at least one broadcast channel among a plurality of broadcast channels (or broadcastcasts) provided by each of the plurality of multimedia content providers.

The multimedia contents provider provides the selected music on the broadcasting service 100 to a terminal of the receiver of multimedia contents (hereinafter referred to as a receiving terminal) and transmits the personal multimedia contents combined with at least one of the messages, songs, ). ≪ / RTI > Also, a chat service between the multimedia contents provider and the multimedia contents receiver on the broadcasting service 100 may be provided.

The streaming radio service 150 may be a service for providing music to a multimedia content recipient by a multimedia content service server without a multimedia content provider. As will be described later, the multimedia content service server can adaptively provide music to the user based on the music selection algorithm through the streaming radio service 150. The receiver of the multimedia contents can receive the music corresponding to the taste of the multimedia contents receiver continuously through the streaming radio service 150. The multimedia content service server may provide different streaming radio services 150 through each of a plurality of streaming radio channels.

In other words, the broadcasting service 100 provides the multimedia contents information generated by the transmitting terminal to the multimedia contents receiver, and the streaming streaming service 150 provides the multimedia contents determined by the multimedia contents server to the multimedia contents receiver .

In the multimedia content recommendation system, the broadcast service 100 and the streaming radio service 150 are not provided as separate services, but may be combined and operated to provide a music recommendation service to the users. Hereinafter, a multimedia content recommendation method based on an organic operation between a broadcast service 100 and a streaming radio service 150 according to an exemplary embodiment of the present invention will be described in detail.

2 is a conceptual diagram of a multimedia content recommendation system according to an embodiment of the present invention.

2, a multimedia content recommendation system for providing a customized broadcast service (hereinafter referred to as a broadcast service) / customized streaming radio service (hereinafter referred to as a streaming radio service) to a user is disclosed.

Referring to FIG. 2, the multimedia content recommendation system may include a transmitting terminal 200, a multimedia content server 220, and a receiving terminal 240.

The transmitting terminal (or the multimedia content transmitting terminal) 200 may be a user equipment (UE) of a multimedia content provider that is implemented to provide a broadcasting service. The transmitting terminal 200 may receive a multimedia content service application from the multimedia content service server (or application relay server) 220. The multimedia content service application can provide a broadcasting service on the transmitting terminal 200. [ Specifically, the multimedia content service application can provide a music selection and transmission service, a chat service, and a personal multimedia contents transmission service on the transmitting terminal 200.

The receiving terminal (or the multimedia contents receiving terminal) 240 may be a user device of a multimedia contents receiver receiving a broadcasting service and / or a streaming radio service. Similarly, the receiving terminal 240 may receive the multimedia content service application from the multimedia content service server (or application relay server) 220. The multimedia content service application may provide a broadcast service and / or a streaming radio service on the receiving terminal 240. The multimedia content receiver can receive a specific broadcast or a specific streaming radio service through a multimedia content service application. The receiver of the multimedia contents may be provided with at least one streaming radio channel selected from at least one of the recommended specific broadcasting or specific streaming radio channels. Alternatively, the receiver of the multimedia contents may directly receive a specific broadcast channel or a specific streaming radio channel.

The multimedia contents service server 220 may be implemented to provide multimedia contents to the multimedia contents receiver. The multimedia content service server 220 may include a processor operatively coupled to a communication unit and a communication unit implemented for communication with the transmitting terminal 200 and / or the receiving terminal 240. The processor receives a request for a multimedia content service application from the transmitting terminal 200 and / or the receiving terminal 240 and transmits the request to the transmitting terminal 200 and the receiving terminal 240, either directly or through a relay server, As shown in FIG. Hereinafter, the operation of the multimedia content service server may be performed based on a processor.

The multimedia contents service server 220 may be implemented to transmit the personal multimedia contents generated by the transmitting terminal 200 to the receiving terminal 240 through a specific broadcasting channel based on the multimedia contents service application. In addition, the multimedia contents service server 220 may provide a streaming radio service for a multimedia contents receiver on a streaming radio channel based on a multimedia contents service application.

In addition, the multimedia content service server 220 may be implemented to perform a recommendation function of a broadcast channel recommendation / streaming radio channel. The multimedia content service server 220 can provide a customized broadcast service / customized streaming service to a user through a broadcast channel / streaming radio channel recommendation function.

That is, the multimedia content application provides recommendation service information to the multimedia content recipient, and the recommendation service information is provided to the multimedia content recipient through a plurality of broadcast service channels and / or a plurality of radio streaming service channels, A broadcast service channel, and / or a recommended radio streaming service channel. The recommended radio streaming service channel may be a personalized channel (or a channel for only a multimedia content recipient), and may provide music to the receiver of the multimedia contents by considering only the taste of the receiver of the multimedia contents.

The recommended broadcasting service channel and / or the recommended radio streaming service channel may be a broadcasting service and / or a recommended radio streaming service matching a music listening preference information of a multimedia contents receiver among a plurality of broadcasting service channels and / or a plurality of radio streaming service channels Information.

The multimedia content service server 220 may utilize user information and multimedia content information for recommendation of a broadcast channel / streaming radio channel.

Specifically, the user information may include information on a user age, a user's gender, a user environment, a user's existing service utilization history, and the like. User age, and user gender may be information on the age and gender of the receiver of the multimedia contents. The user environment information may include information about a location / situation / time / season etc. at which the multimedia content recipient currently receives the broadcasting service / streaming radio service. The user's existing service use history information may include information on the broadcast channel and / or the streaming radio channel, the broadcast channel and / or the streaming radio channel received by the receiver of the multimedia contents, and the response information to the music.

The multimedia content information may include emotions, feelings for multimedia content (e.g., music), information about images / ideas associated with multimedia content, and the like. For example, the multimedia content information of a particular music may include information about music, feelings about music, other objects associated with music, and the like. For example, the multimedia contents information of a specific music may include various feelings such as 'pleasant', 'happy', 'depressed', 'sad', 'calm', 'peaceful', 'fast' / Feelings / ideas.

The multimedia content information may be generated by synthesizing reaction information of the multimedia contents recipients, analysis information on the multimedia contents themselves, response information on multimedia contents of various users existing on other webs, and the like.

3 is a conceptual diagram illustrating a specific operation of the multimedia contents service server according to the embodiment of the present invention.

3, a method of generating multimedia content information of a multimedia content service server is disclosed.

Referring to FIG. 3, a method for generating multimedia content information 350 for a specific music is disclosed in detail.

The multimedia content information 350 for music may include information related to the characteristics / characteristics of the music, and may be formed through a plurality of paths.

Specifically, the multimedia content information may include information about the emotion of the music, the feeling of the music, and other objects related to the music as the sub characteristics information.

The emotion information for music may include information related to the emotion of a person listening to music such as pleasure, happiness, sadness, depression, serenity, crying, laughter, and the like.

The feeling information on music may include information related to a person's feeling of music such as peace, fastness, slowness, quietness, dreamy, noisy, progressive, quiet,

Other object information related to music may be information associated with music, and information about images / ideas associated with music such as independents, independents, Olympics, etc., when the national anthem is heard. Another example could include information about associated images / concepts such as vacation, sun, beaches, etc., when listening to songs related to summer. Other object information related to the music may be generated in association with the lyrics when the lyrics exist in the music. For example, when words such as 'snow' and 'winter' are included in the lyrics more than the threshold number, 'eye' and 'winter' may be selected as other object information related to music.

As a method of determining the multimedia content information 350, the multimedia content information 350 for music may be generated based on the music response information 300 provided on the multimedia content service application.

For example, if a broadcast service is provided, a chat service between the multimedia content provider and the multimedia content recipient may be provided on the multimedia content service application. In this chat service, conversation about music-related information (for example, feelings / emotions, associative images / notions, etc.) provided through the broadcast service can be conducted. The multimedia contents service server may collect the conversation including the information about the music and generate the multimedia contents information 350 for the specific music.

As a specific example, a specific multimedia content recipient can apply to the multimedia content provider for Music A to enter a specific broadcast channel and soothe the depressed mood. The multimedia content provider can select music A and broadcast it on the corresponding broadcast channel. In the chat window of the broadcasting channel, information related to music A (for example, "I am strong", "I am excited", "I feel good", "I am weak when I am depressed" . The multimedia content service server may collect / manage information related to the music and then generate / manage the multimedia content information 350 for the music A.

The response of the multimedia content recipient to music played on the broadcast channel can be reflected as user existing service usage history information, which is one of the user information of the multimedia content recipient. Hereinafter, the user's existing service use history information may be utilized for selecting music to be provided to the receiver of the multimedia contents in the streaming radio service for the receiver of the multimedia contents. That is, in the multimedia content recommendation system, the broadcast service and the streaming radio service may not be provided as separate services, but may be organically combined to provide a music recommendation service to the receiver of multimedia contents.

In addition, multimedia content information 350 for music may be generated based on information 320 collected from various other webs as an alternative method of determining multimedia content information 350. [ The multimedia contents service server can extract information related to specific music and / or feelings / feelings / thoughts on specific music based on a big data processing process. For example, multimedia content information 350 for music may be generated by collecting relevant information about a particular music on a social network service (SNS) and extracting an impression of a particular music associated therewith. Specifically, a music source and a music video of a specific music A are provided among users on the SNS / portal service, and many users can comment on the feeling of music A, the feeling of music, and the idea / image related to music . The multimedia contents service server collects comment information about the music A and extracts comments having high relevance to the music A among the comments after collecting the comment information. The multimedia content service server can generate the multimedia content information 350 for the music on the basis of the comments having high relevance to the extracted music A. [

In addition, multimedia content information 350 for music can be generated by analyzing characteristics of music itself as another method of determining multimedia content information 350. [ The multimedia content server can extract the characteristic information 340 contained in the music itself such as the composition of music, the melody of music, the speed of music, the musical instrument used, the presence of the singer, the meaning of the lyrics, The multimedia content information 350 for the music may be generated based on the characteristic information 340.

That is, the multimedia content information 350 for music may be generated / managed based on the reaction information of the multimedia content recipients, the information about the music collected on the web, and the characteristic analysis information about the music itself.

Since the response to the same music may vary depending on the user, there may be a variety of views on the same music A as well. For example, there may be a user who judges that music A is 'exciting music' while a user who judges music A as 'sad music'. If there is an inconsistency in the evaluation of the music A, the lower characteristic information indicating the emotion exceeding the emotion threshold ratio in proportion may be determined as the multimedia content information 350 for the music A. [

As a concrete example, if the sentimental rating of the exciting music is 80%, the impression rating of the sad music is 20%, and the emotion critical ratio for determining the multimedia content information is 70%, the impression that the music A is sad music is noise information And is not included in the multimedia content information 350 for the music A, and can be temporarily held. The characteristic information of 'exciting' and 'sad' as the multimedia content information 350 for the music A is the multimedia content information 350 for the music A in the case where the impression ratio of the music A is 60%, the impression of the sad music is 40%, and the emotion critical ratio is 70% It can be held in the content information 350 not yet included as sub information. If the specific emotion exceeds the emotion threshold ratio through the additional information collection, the emotion may be determined as the multimedia content information 350 for the music A. [

The feeling information about the incongruent music, the feeling information on the music, the idea / image information on the music coexist as sub property information for the music and can be included in the multimedia content information 350.

4 is a conceptual diagram illustrating a specific operation of the multimedia contents server according to the embodiment of the present invention.

4, a method of generating user information of a multimedia content server is disclosed.

As described above, the user information may include information on a user age, a user's gender, a user environment, a user's existing service utilization history, and the like.

The user environment information 440 may include information on a service providing environment to the multimedia content receiver. Specifically, the user environment information 440 may include information on the location / situation / time / season of the receiver of the multimedia contents.

The location information 400 of the receiver of the multimedia contents can be obtained by interworking with a location information module such as GPS (global positioning system) of the receiving terminal. The location information module of the receiving terminal transmits the location information to the multimedia contents service server, and the multimedia contents service server can determine the current location of the multimedia contents receiver based on the location information 400. The receiver of the multimedia contents may register the location information 400 for the place (for example, house, company, etc.) of the receiver of the multimedia contents. The multimedia content service server may determine whether the registered location matches the location information 400 of the receiver of the multimedia contents to determine the current location of the user.

The status information 410 of the receiver of the multimedia contents may include information on the current status of the receiver of the multimedia contents. For example, the multimedia content recipient context information 410 may include specific contextual information 410 regarding whether the current multimedia content recipient is currently working, resting, moving, sleeping, and so on. The context information 410 of the receiver of the multimedia contents may be determined by being interlocked with the location information 400 of the receiver of the multimedia contents at present or may be determined by further considering whether or not the receiver terminal is moved.

The time information 420 of the receiver of the multimedia contents may be determined in consideration of the synchronized time information 420 between the multimedia contents service server and the reception terminal.

The seasonal information 430 of the receiver of the multimedia contents may be determined in consideration of the synchronized season information 430 between the multimedia contents service server and the receiving terminal.

The user's existing service usage history information 460 may include information on a broadcast channel, a radio streaming channel, and the like received by a receiver of a conventional multimedia content. In addition, the user's existing service use history information may include information on the music that the receiver of the multimedia contents has so far listened to, information on the response of the receiver of the multimedia contents to the listened music, and the like.

According to the embodiment of the present invention, the music listening preference information 470 of the multimedia content recipient can be determined based on the user environment information 440 and the user existing service usage history information 460. If the multimedia contents information of the music mainly received by the receiver of the multimedia contents includes the sub-characteristics information indicating 'stillness', 'slow tempo', etc., the music listening preference information 470 of the multimedia contents receiver is silent and has a slow tempo You can direct music. The music listening preference information 470 may have an information format similar to the multimedia content information.

For example, if the multimedia content information includes 'quiet', 'quiet', 'slow tempo', and the like, and the multimedia contents receiver mainly listens to the music having such characteristics, When included in the information, the user's music listening preference information 470 may also include subordinate information such as 'quiet', 'quiet', 'slow tempo', etc. as the lower information.

In addition, the music listening preference information 470 of the receiver of the multimedia contents may be determined based on the user existing service usage history information 460 in the same manner. Music preference information 470 is classified according to the user's situation by classifying the preferred music when the receiver of the multimedia contents is in the room (or home), the music preferred by the recipient of the multimedia content at the time of moving, the music preferred by the receiver of the multimedia contents, May be determined.

For example, if the specific multimedia content information 450 that the receiver of the multimedia contents mainly listens at home includes 'low', 'quiet', 'low tempo', etc. as sub characteristics information and the multimedia contents receiver mainly listens The specific multimedia content information 450 may include 'high tempo', 'exciting', 'positive', etc. as sub characteristics information. In this case, the user's music listening preference information 470 is also stored as sub information ('quiet', 'quiet', 'slow tempo'), external (fast tempo, And so on.

The multimedia contents service server can provide a personalized music service to the receiver of the multimedia contents on the radio streaming service in consideration of the music listening preference information 470 of the receiver of the multimedia contents and the multimedia contents information of the music stored / managed in the multimedia contents service server .

According to the embodiment of the present invention, it is possible to recognize not only the music that is selected and reproduced on the broadcasting service but also whether or not the DJ is singing in the broadcasting service, Information 450 and user information may be set. That is, when the DJ is singing, the multimedia content information 400 and the user information can be generated based on the response to the music by considering the music to be reproduced in the selected music. In the broadcast service, when the DJ's moment is analyzed to recognize whether or not the DJ is performing the song, the DJ can recognize that the melody (or change in the pitch) is detected over a certain period of time.

5 is a conceptual diagram illustrating a recommendation operation for a specific broadcast service / streaming radio service of a multimedia content server according to an embodiment of the present invention.

5, a method of recommending a broadcasting service / streaming radio service in consideration of music listening preference information of a multimedia contents receiver is disclosed. In particular, it is assumed in FIG. 5 that the multimedia content recipient first accesses a specific broadcast channel or a specific streaming radio channel and the user information of the multimedia content recipient does not exist.

Referring to FIG. 5, a multimedia content receiver can access a specific broadcast channel or a specific streaming radio channel (step S500).

The receiving terminal of the receiver of the multimedia contents can access the multimedia contents service server based on the multimedia contents service application installed in the receiving terminal.

A specific broadcasting channel or a specific streaming radio channel is selected based on the multimedia content service application installed in the receiving terminal, and the multimedia contents service server provides a broadcasting service / streaming radio service through a specific broadcasting channel selected as the receiving terminal or a specific streaming radio channel can do.

The user information of the receiver of the multimedia contents is generated (step S510).

The multimedia content recipient may access the specific broadcast channel / specific streaming radio channel and listen to the music provided by the specific broadcast channel / specific streaming radio channel. In addition, a response to music provided by a particular broadcast channel / specific streaming radio channel may be provided. The service utilization history information of the receiver of the multimedia contents can be collected and generated through the process of using the broadcasting service / streaming radio service.

In addition, the user environment information of the receiver of the multimedia contents corresponding to the listener of the music can be collected in parallel and generated as user information. That is, the music (or the multimedia content information about the music) heard by the receiver of the multimedia contents may be generated / managed by matching or corresponding to the user environment information of the receiver of the multimedia contents.

The music listening preference information of the receiver of the multimedia contents is determined based on the user information of the receiver of the multimedia contents and the multimedia contents information (step S520).

The user's music listening preference information can be determined based on the user's existing service use history information and user environment information of the receiver of the multimedia contents. As described above, if the multimedia contents information of the music that the multimedia contents receiver mainly hears includes sub property information indicating 'pleasant', 'exciting', 'fast tempo', etc., the music listening preference information of the receiver of the multimedia contents You can direct music that is fun / exciting / fast.

In addition, since the location information of the user corresponding to the music, the time information of the user, and the user's situation information are collected in parallel and generated as the user information, the multimedia contents receiver can be stored in the room The music preference information of the user can be determined by classifying the preferred music when the user is present, the music preferred by the recipient of the multimedia content when moving, and the music preferred by the recipient of the multimedia content.

Information on the recommended broadcasting channel / recommended streaming radio channel is provided based on the music listening preference information of the receiver of the multimedia contents (step S530).

The multimedia contents service server can provide information on a recommended broadcasting channel / recommended streaming radio channel suitable for the receiver of the multimedia contents based on the music listening preference information of the multimedia contents receiver.

The multimedia content service server can define and manage attribute information on the provided broadcast channel and the streaming radio channel.

Specifically, the multimedia content server can define attribute information of a broadcast channel based on information on music played on the broadcast channel. In addition, the multimedia content server can receive streaming radio data based on information on music played on a streaming radio channel, Attribute information of the channel can be defined. The attribute information of such a broadcast channel / streaming radio channel may be similar to the above-described multimedia content information. The multimedia content information of music reproduced in the streaming radio channel is collected and the lower characteristic information occupying more than the threshold ratio can be defined as the attribute information of the broadcasting channel. For example, if music above a threshold rate (e.g., 80%) in a particular streaming radio channel is music with 'quiet' sub-characteristics information, the property information of the streaming radio channel may be 'quiet'.

Therefore, if the broadcast channel / streaming radio channel having the property information of the broadcast channel / streaming radio channel is defined and the property information matching the music listening preference information of the receiver of the multimedia contents is the recommended channel / Can be determined.

The recommendation of a streaming radio channel may mean that the recommended streaming radio channel provides a streaming radio channel only for the multimedia content receiver as described above. Or a streaming radio channel may mean recommending and providing a streaming radio channel by grouping a multimedia streaming radio channel with a multimedia content recipient and other multimedia content recipients having similar tastes to the multimedia content recipient.

6 is a conceptual diagram illustrating a recommendation operation for a specific broadcasting service / streaming radio service of a multimedia contents server according to an embodiment of the present invention.

6, a method of recommending a broadcasting service / streaming radio service in consideration of music listening preference information of a multimedia contents receiver is disclosed. In particular, FIG. 6 discloses a method in which user information of a multimedia content recipient exists and a user's response is reflected in real time after feedback to recommend a broadcast service / streaming radio service to a multimedia content recipient.

Referring to FIG. 6, information on a recommended broadcasting channel / recommended streaming radio channel is provided based on the first music listening preference information of the multimedia content recipient (step S600).

The receiver of the multimedia contents that is provided with the broadcasting service / streaming radio service through the multimedia contents service server can have the first music listening preference information. As described above, the first music listening preference information of the multimedia content recipient can be determined based on the first user information and the first multimedia content information of the receiver of the multimedia contents.

The multimedia content service server may recommend a broadcast channel / streaming radio channel to the receiver of the multimedia contents based on the first music listening preference information.

The first user information and / or the first multimedia content information is re-determined on the basis of the response to the music fed back by the receiver of the multimedia contents (step S610).

The receiver of the multimedia contents can reset the user's existing service usage history information among the user information of the receiver of the multimedia contents based on the response to the music provided on the connected broadcast channel / streaming radio channel. Specifically, when the receiver of the multimedia contents preferences the response to the still music, the information on the preference response of the receiver of the multimedia contents is reflected in the service usage history information of the first user information, and the first user information is set as the second user information .

In addition, the first multimedia content information for the music may be reset based on the response to the music provided on the broadcast channel / streaming radio channel accessed by the receiver of the multimedia content. For example, when the receiver of the multimedia contents responds to the music to be silent music, the response of the receiver of the multimedia contents may be reflected to the first multimedia contents information of the specific music and may be reset to the second multimedia contents information.

The second music listening preference information of the multimedia content recipient may be determined based on the second user information and the second multimedia content information of the receiver of the multimedia contents (step S620).

The second music listening preference information of the multimedia content recipient may be determined based on the second user information and the second multimedia content information of the receiver of the re-established multimedia content. The multimedia content service server may recommend a broadcast channel / streaming radio channel to the receiver of the multimedia contents based on the second music listening preference information.

FIG. 7 is a conceptual diagram illustrating a recommendation operation for a specific broadcasting service / streaming radio service of a multimedia contents service server according to an embodiment of the present invention.

FIG. 7 illustrates a method for providing a broadcast channel / streaming radio channel matching the current mood of a receiver of multimedia contents in consideration of the priority of user information. Even if existing music content preference information of the receiver of the multimedia contents exists, the current mood of the receiver of the multimedia contents is reflected most preferentially so that the broadcasting service and the streaming radio service can be provided to the user.

For example, the receiver of the multimedia contents may have the existing service usage history information of the user who listened to the silent music, but the receiver of the multimedia contents may want to listen to the music at the present time. In this case, it is possible to preferentially reflect today's musical taste of the receiver of the multimedia contents or to provide a broadcasting service / radio streaming service to the receiver of the multimedia contents by weighting.

Referring to FIG. 7, the multimedia content service server can collect the current reaction to the broadcast channel of the receiver of the multimedia contents, the current response to the streaming radio channel of the receiver of the multimedia contents, and determine the current music listening preference information 700. The current music listening preference information may be information determined based on the user's existing service use history information for a predetermined period (for example, one day) based on the current time point.

If the current music listening preference information 700 does not conflict with the music listening preference information 720, the multimedia content service server may provide the streaming radio service 740 by reflecting the music listening preference information 720.

Conversely, if the current music listening preference information 700 is inconsistent with the music listening preference information 720, the multimedia content service server may provide the streaming radio service 740 by reflecting the current music listening preference information 700.

Whether or not the current music listening preference information 700 and the music listening preference information 720 are inconsistent depends on the lower characteristic information indicated by the current music listening preference information 700 and the lower characteristic information indicated by the music listening preference information 720 And may be determined based on whether there is a contradiction between the information. For example, if the music listening preference information 720 indicates 'low tempo', 'quiet', and the current music listening preference information 700 indicates 'slow tempo', the current music listening preference information 700 ), The multimedia content service server may provide the streaming radio service 740 by reflecting the music listening preference information 720.

As another example, when the music listening preference information 720 indicates' quiet 'and' the music listening preference information 700 indicates' fast tempo 'and' exciting ' When there is contradictory lower information between the information 700 and the music listening preference information 720, the multimedia content service server preferentially reflects the current music listening preference information 700 to provide the streaming radio service 740 have. The inconsistency between the sub information can be determined based on the relation information between the sub information managed and set by the multimedia content server.

When a contradiction occurs between the current music listening preference information 700 and the music listening preference information 720, the multimedia content service server re-sets the user information of the multimedia content recipient through a separate determination, The music listening preference information can be reset.

FIG. 8 is a conceptual diagram illustrating a recommendation operation for a specific broadcasting service / streaming radio service of a multimedia contents service server according to an embodiment of the present invention.

8, when a contradiction occurs between the current music listening preference information and the music listening preference information, the multimedia content server resets user information (for example, user existing service usage history information) of the multimedia content recipient through a separate determination And resetting the music listening preference information of the user in accordance with the resetting of the user information.

Referring to FIG. 8, when the multimedia contents server determines that a contradiction occurs between the current music listening preference information 800 and the music listening preference information 820, re-determination of the user environment information may be performed.

The multimedia content server determines whether the current user environment information (user location / situation / time / season, etc.) 840 of the multimedia content recipient is the same as the user environment information 860 when generating the music listening preference information 820 Similarity) can be performed.

If the user environment information 860 and the current user environment information 840 are the same at the time of generating the music listening preference information 820, the user experiences an inconsistency with the existing music listening preference information 820 and the user hears The user can add information about the music to the existing service utilization history information of the user. The user's existing service use history information that the user has listened to the music in which such inconsistency has occurred can be managed as noise information as a temporary change.

In this case, the multimedia contents service server transmits music having the multimedia content information matching the current music listening preference information 800 to the receiver of the multimedia contents receiving the streaming radio service in a critical number of times, Can be clearly identified. If the user's existing service use history information is continuously accumulated and exceeds the threshold ratio, the multimedia content information of the music in which the inconsistency occurs is also reset to the music listening preference information 820 of the user Lt; / RTI >

For example, if the music content preference information 820 of the multimedia content recipient indicates music that is pleasant / exciting / fast in the first user environment, and that the multimedia content recipient in the first user environment currently has calm / . In this case, the record of listening to the still / sad music under the first user environment can be added as the existing service use history information of the user. If the still and / or sad music has been continuously added to the list of recorded music in the first user environment and exceeds the threshold ratio, the music content preference information of the multimedia content recipient is silent / pleasant / tempo-fast in the first user environment, It can be changed to pointing to sad music.

If the user environment information 860 and the current user environment information 840 are different when generating the music listening preference information, the music listening preference information 820 can be reset in consideration of the music listening preference information for the current user environment have.

For example, the multimedia content recipient's music listening preference information 820 may indicate music that is enjoyable / exciting / tempo-fast in a first user environment (e.g., a commute) The user at home) can hear quiet / sad music. In this case, the record of listening to the still / sad music under the second user environment is added as the existing service usage history information of the user, and the music listening preference information 820 under the second user environment of the multimedia receiver is used to store still / You can tell. Thereafter, the multimedia content service server may provide still / sad music to the multimedia content recipient in a second user environment.

Hereinafter, a method of generating multimedia content information will be described in more detail in the embodiment of the present invention.

As described above, the multimedia contents information can be generated based on the response information of the multimedia contents receivers, information about music collected from the web (or another server), characteristic analysis information about the music itself, and the like.

9 is a conceptual diagram illustrating a method of generating multimedia content information according to an embodiment of the present invention.

9, a method of generating multimedia content information based on response information of music content recipients is disclosed. In FIG. 9, for the sake of convenience, the generation of the multimedia content information based on the chat record collected on the broadcast service is started. According to an embodiment of the present invention, there may be a response of a multimedia content receiver to music played on a radio streaming service rather than a broadcast service, and multimedia content information about music may be generated based on the response.

Referring to FIG. 9, the response to the music of the multimedia contents receiver on the chat service provided on the broadcasting service can be collected.

The multimedia contents service server can collect the chat history by setting the reproduction start time (or the threshold time before the reproduction start time) based on the music reproduction time as the reaction collection start time and the reaction collection end time after the threshold time of the reproduction time have. Hereinafter, the section from the reaction collection start time to the reaction collection end time may be expressed by the term of the reaction collection section 900.

For example, in the case of music having a reproduction time of 3 minutes, the start time of the reproduction of the music is the reaction collection start time, and after the reproduction of 3 minutes of music, an additional critical time (for example, 1 minute) Collection end time so that the multimedia contents service server can collect the response to the music. In such a case, the total reaction collection interval 900 may be 4 minutes.

The multimedia contents service server collects chat information (910) including a plurality of conversation sentences generated during the entire response collection period (900) when music is reproduced, and generates response information (910) for music based on analysis of a plurality of conversation sentences Can be extracted. Hereinafter, in an embodiment of the present invention, a method of extracting information on feeling / feeling of music for generating multimedia content information is exemplarily disclosed for the sake of convenience of description. However, as an example of multimedia content information, A method of extracting information on the information can also be performed in the same manner.

Specifically, the multimedia contents service server can collect sentences including words expressing feelings / feelings (hereinafter, emotion / feel words) among a plurality of sentences. The database of the multimedia content service server may store information on feel / feel words. The multimedia content service server can determine whether information on emotion / feel words is included in the sentence based on the database. The multimedia contents service server can periodically update the information about emotion / feel words based on the machine learning.

The multimedia contents service server determines whether the object of emotion / feeling is the music to be reproduced in the sentence including emotion / feeling word (hereinafter, emotion expression sentence 920).

In the emotion expression sentence 920, the subject of the emotion / feel word may not be music. Accordingly, it can be additionally determined whether or not the object of the emotion / feel word is currently reproduced in the emotion expression sentence 920.

In the emotion expression sentence 920, a determination may be made on the subject of emotion / feel words in the emotion expression sentence to determine whether the subject of emotion / feeling is the music to be reproduced.

For example, when the subject of the emotional expression sentence 920 indicates music (or information related to music (singer, lyrics, etc.)), the emotion / feeling word in the emotional expression sentence 920 indicates the multimedia content information of the music Can be used to determine. When the subject of the emotion expression sentence 920 indicates music, such emotion expression sentence may be defined as a term of music emotion expression sentence 930. [

Conversely, when the subject of the emotional expression sentence 920 indicates an object not related to music, the emotion / feel words in the emotional expression sentence 920 may not be utilized to determine the multimedia content information of the music. If the subject of the emotional expression sentence 920 does not indicate music, such emotional expression sentence 920 may be defined with the term other emotional expression sentence 940. [

If the subject of the emotion expression sentence 920 does not exist, the emotion / feeling word in the emotion expression sentence 930 can be determined as multimedia content information based on the relevance to the emotion / feel word in the music emotion expression sentence 930 have. An emotional expression sentence in which the subject does not exist can be expressed by the term " non-subject emotional expression sentence "

For example, a phrase "Song is sad" may be present in the musical sentiment expression sentence 930 collected within the entire response collection section of Music A. In this case, the multimedia content information of the music A may be determined based on the 'emotion / feeling word' 'grief' included in the music emotion expression sentence 930. If there is no subject in the emotional expression sentence collected in the whole reaction gathering section like "It is depressed", emotion / feelings contained in sentence expression sentence without subject and emotion / feeling included in music emotion expression sentence The degree of association with the word 'sadness' can be judged. (E.g., sadness) included in the emotion / expression word (e.g., melancholy) included in the subject-less emotion expression sentence and the music emotion expression sentence 930 is greater than or equal to the threshold value , Emotion / feel words included in the pseudo emotional expression sentence 950 may be used to determine the multimedia content information of music A. [ The feel / feel words included in the pseudo emotional expression sentence 950 can be used to determine multimedia content information for music with relatively low weights.

The degree of association between emotion / feeling words can be determined based on information on words classified into a predetermined emotion category and a feeling category. For example, a classification for a primary category may be performed with a positive category / negative category, and a sub-category for negative feelings such as sadness or anger in a negative category may be classified. Further, the subcategories can be further classified hierarchically. Based on the hierarchical structure, the relevance between emotion / impression words classified into specific categories can be determined.

In the example described above, music emotion expression sentence 930, other emotion expression sentence 940, and non-sentence emotion expression sentence 950 are classified based on emotion for convenience of explanation. However, not only the emotion of music but also the sentences relating to information about feelings, images, and ideas can be distinguished in the same way and can be broadly classified into terms such as music expression sentences, other expression sentences, and non-subject expression sentences.

In other words, according to an embodiment of the present invention, the multimedia content application provides a music recommendation service to a multimedia content recipient corresponding to a receiving terminal, and the music recommendation service recommends music that matches music preference information of the multimedia content recipient . The music listening preference information can be generated based on the multimedia content information of the music listened to by the receiver of the multimedia contents.

The multimedia content information may include emotion information about music, feeling information about music, and information about objects related to music. The multimedia content information may be generated based on a chat record collected on a broadcast service provided on the basis of a multimedia content service application, and the chat record may be collected during a response collection interval set based on a play time of the music. The multimedia contents service server may divide a plurality of sentences included in the chat record into music expression sentences, other expression sentences, and non-subject expression sentences, and the multimedia contents service server may determine the multimedia contents information in consideration of the music expression sentences have.

The music expression sentence includes a first emotion information, a first feeling information, or a word related to information about the first object, wherein the first emotion information, the first feeling information, or the information about the first object (for example, Image, idea information) may be a musical sentence.

The other emotion expression sentence includes a word related to the second emotion information, the second emotion information, or the information about the second object, and the object of the information about the second emotion information, the second feeling information, It can be a non-sentence.

The non-subject expression sentence includes a word related to the third emotion information, the third feeling information, or the information about the third object, and the third emotion information, the third feeling information, It can be a sentence.

In addition, the multimedia contents service server determines whether the information about the third emotion information, the third feeling information, or the third object included in the non-subject expression sentence is for the music, and further considering the non-subject expression sentence, It is possible to determine the multimedia contents information about the multimedia contents. Whether the information on the third emotion information, the third emotion information, or the third object included in the non-subject expression sentence is for the music is determined based on the first emotion information, the first emotion information, Based on the degree of similarity with the information on < RTI ID = 0.0 >

10 is a conceptual diagram illustrating a method of generating multimedia content information according to an embodiment of the present invention.

FIG. 10 discloses a method for collecting multimedia content information for music in consideration of collected context / context information.

Referring to FIG. 10, there may be a situation where the multimedia contents receiver requests the multimedia contents provider to play music with a pleasant feeling while the multimedia contents receiver is depressed, and the multimedia contents provider plays the requested music. Alternatively, the multimedia content provider may say that he or she is going to play music that is feeling better, and then play the music. The multimedia content server may determine the multimedia content information of the music to be reproduced in consideration of the context information based on the context information (or the context information).

Learning for the music playback situation 1000 can be performed to determine context information. For example, after a first sentence such as 'Give a sad song a recommendation', which is generated by the receiver of the multimedia contents, the multimedia content provider may generate a second sentence indicating the meaning of acceptance And a situation in which the music is reproduced may occur. Information (e.g., 'pleasant' or 'sad') 1010 related to the music included in the first sentence is reflected in the multimedia content information 1020 for the music being played under the music reproduction state 1000 .

As another example, a situation may occur in which the music is reproduced after the third sentence such as 'I am playing a song with a good feeling' and 'I will give a sad song' generated by a multimedia content provider. Information (e.g., mood or sadness) 1010 related to the music contained in the third sentence under the above music reproduction status 1000 may be reflected in the multimedia content information 1020 for the music being played back.

The multimedia contents service server can continuously generate the multimedia contents information for the music by continuously learning the situation information (or the conversation situation information) generated to reproduce the music in a plurality of broadcasting channels.

Specifically, the multimedia contents service server can extract the multimedia content information for music in a specific sentence included in the corresponding situation by learning the situation, if there is a conversation situation which is mainly repeated before music reproduction.

11 is a conceptual diagram illustrating a method of generating multimedia content information according to an embodiment of the present invention.

11, a method for generating multimedia content information for music in an external server (or web) is disclosed.

Referring to FIG. 11, the multimedia content service server may collect multimedia content information 1120 for music from the external server 1100. [

For example, the multimedia content service server may search for information about music in a music playback service server (e.g., a YouTube server) or a social network service (SNS) server that provides a service for playing music, Information may be collected and multimedia content information 1120 may be generated.

Specifically, the reaction information for a specific music on the music reproduction service server may be executed in the same form as the comment. The multimedia content service server may collect comments including response information for music, and may generate multimedia content information 1120 for music through analysis of comments.

Likewise, on the SNS server, the response information about the music may be included in a format such as the original text and the comment. The multimedia content service server may collect the original text / comment including the response information for the music, and may generate the multimedia content information 1120 for the music through analysis of the comment.

11, the multimedia content service server analyzes the sentences of the comments on the basis of emotional expression words / feel expression words / concept expression words / image expression words included in the database, Feel expression / concept expression / image expression, and the like are included.

12 is a conceptual diagram illustrating a learning method of a multimedia contents service server according to an embodiment of the present invention.

12, a learning method for generating multimedia content information such as an emotional expression word, a feeling expression word, an idea expression word, an image expression word, and the like is disclosed.

Referring to FIG. 12, a sentence containing emotion expression words, a sentence including emotion expression words, a sentence including concept expression words, and a sentence including image expression words, which are primarily set in the multimedia contents service server, And the first learning 1210 can be performed.

Hereinafter, it is assumed that a sentence including an emotion expression word is inputted for convenience of explanation.

Sentences for the first learning 1210 can be classified and input into categories. For example, a sentence containing an emotional expression word may be classified primarily into sentences expressing positive emotions and sentences expressing negative emotions. Again, sentences that express positive emotions as subcategories can be classified and input into subcategories such as sentences expressing happy emotions, sentences expressing happy emotions, sentences expressing funny emotions, and the like.

Thereafter, the multimedia contents service server can perform learning on a new emotional expression word.

If the multimedia contents information can not be classified through the learning result based on the sentence input through the primary learning 1210, the multimedia contents service server can generate the multimedia contents information through the secondary learning 1220 have.

For example, it may be assumed that a new sentence expression such as 'melancholy' is included in a particular sentence. When a new emotional expression word is included in a specific sentence, the multimedia contents service server can not classify the emotion through the first learning 1210. [

In this case, the multimedia contents service server can perform learning on a new emotional expression word through the secondary learning 1220. [ Specifically, the multimedia content service server can determine whether a specific word is an emotional expression word.

For example, the multimedia content service server uses the dictionary database (or a separate search server) to search for the meaning of 'melancholy' and / or 'melancholy' to determine whether the word 'melancholy' And information about a similar word can be obtained.

If the meaning of 'melancholy' and / or a word similar to 'melancholy' is included in the emotion expression word learned through the first learning 1210, the multimedia content service server recognizes 'melancholy' as one of emotion expression words .

The multimedia content service server can determine whether to classify the new emotion expression word into new sub information of the new multimedia content information or to classify it as one of the existing emotion expression words based on the meaning of the new emotion expression word.

For example, it can be assumed that 'melancholy' means 'depression', and 'melancholy' exists as a sub information of the existing multimedia contents information. In this case, the multimedia content server can determine whether to generate new sub information such as 'melancholy' or to classify 'melancholy' as the same sub information as 'melancholy'.

The multimedia content service server first reserves classification of the new emotional expression word 'melancholy' as new sub information, and classifies it as being the same as 'depression', which is the most similar existing sub information, and generates multimedia content information for music can do. Thereafter, when the expression 'melancholy' is collected more than a threshold number of times as an emotion expression word, the expression 'melancholy' can be classified and managed as one separate lower-level characteristic information.

In this way, learning on emotion expression words frequently used for expressing emotions is performed, and multimedia contents information can be managed based on integrated sub property information for expressions that are not frequently used for emotional expression .

13 is a conceptual diagram illustrating a method of generating multimedia content information according to an embodiment of the present invention.

In Fig. 13, a method for generating characteristic analysis information on the music itself is disclosed.

Referring to FIG. 13, the multimedia content server may analyze the music itself to generate multimedia content information for music.

First, the multimedia content server can classify music composition 1310 and music speed 1320.

The multimedia content server may classify whether the music is major or minor for classifying music composition 1310 and classify the music according to the pace 1320. [

For example, when the music is the major, the multimedia contents information of the music can be set to be highly related to the feelings such as joy, happiness, and joy. Also, the faster the pace is, the stronger the emotion is classified, and the multimedia content information of the music can be determined.

On the other hand, when the music is forged, the multimedia content information of the music may be set to be highly related to the feelings such as sadness, depression, sadness, and the like. The slower the pace is, the stronger the emotion is classified, and the multimedia content information of the music can be determined.

In addition, the multimedia content server may determine the multimedia content information based on the lyrics 1330. [

Specifically, in the music in which the lyrics 1330 exist, the weight of the lyrics 1330 in determining the emotion of the receiver of the multimedia contents may be large. Accordingly, the multimedia content server can generate the multimedia content information of the music through the analysis of the lyrics 1330.

It is possible to classify the words constituting the lyrics 1330 for the analysis of the lyrics 1330. For classification of the words constituting the lyrics 1330, information on emotion expression words / feelings expression words / concept expression words / image expression words etc. obtained through learning as described above can be utilized.

For example, if the sentence "I miss the street I walked with when I walk on the way of rain" is included in the lyrics (1330), "I miss" is extracted as the emotional expression word, , 'Rain', and the like can be extracted. The extracted word can be managed as multimedia content information of music.

The multimedia content server may use various methods to reduce the noise of the words included in the lyrics 1330 to reduce the noise of the information classification. For example, if you have the lyrics (1330) of a song about farewell, "I do not know if it's in the dark or the sun is shining, I do not know if it's sunny" or "I'm always laughing" (1330) Although the main feeling of music may be sadness, the emotional expression word 'laugh' and the idea expression word / image expression word 'Hae' are extracted through the lyrics 1330 and generated as noise information of the multimedia contents information of music have.

In order to reduce the noise information, the multimedia contents service server may determine the subject of the lyrics through the context information about the lyrics 1330 and determine only the information having high relevance to the subject of the lyrics 1330 as the multimedia content information.

The multimedia content server may be analyzed for the lyric text to determine the subject of the lyric 1330. [ Specifically, the multimedia content server may extract repeated sentences / words from the lyrics 1330 and determine the subject of the lyrics considering the degree of association between repeated sentences / words. For example, if words such as 'I miss you', nostalgia, farewell, etc. frequently appear, the theme of the lyrics 1330 may be determined by nostalgia, parting, and the like.

When the subject of the lyrics 1330 is determined, only the emotional expression word / feel expression word / concept expression word / image expression word highly related to the subject of the lyrics 1330 can be utilized to determine the multimedia contents information for the music have.

The embodiments of the present invention described above can be implemented in the form of program instructions that can be executed through various computer components and recorded in a computer-readable recording medium. The computer-readable recording medium may include program commands, data files, data structures, and the like, alone or in combination. The program instructions recorded on the computer-readable recording medium may be those specifically designed and configured for the present invention or may be those known and used by those skilled in the computer software arts. Examples of computer-readable media include magnetic media such as hard disks, floppy disks and magnetic tape, optical recording media such as CD-ROM and DVD, magneto-optical media such as floptical disks, medium, and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions include machine language code, such as those generated by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like. The hardware device may be modified into one or more software modules for performing the processing according to the present invention, and vice versa.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed embodiments, but, on the contrary, Those skilled in the art will appreciate that various modifications and changes may be made thereto without departing from the scope of the present invention.

Accordingly, the spirit of the present invention should not be construed as being limited to the above-described embodiments, and all ranges that are equivalent to or equivalent to the claims of the present invention as well as the claims .

Claims (10)

A method of providing user-customized multimedia content based on multimedia content information,
Receiving a request for a multimedia content service application from a transmitting terminal and / or a receiving terminal; And
And transmitting the multimedia content service application to the transmitting terminal and the receiving terminal either directly or through a relay server,
Wherein the multimedia content service application provides a music recommendation service to a multimedia content recipient corresponding to the receiving terminal,
The music recommendation service recommends music that matches the music listening preference information of the multimedia content recipient,
Wherein the music listening preference information is generated based on multimedia content information of the music listened to by the multimedia content recipient,
Wherein the multimedia content information includes emotion information on the music, emotion information on the music, and information on objects related to the music,
The multimedia content service server receives first information on sentences including words expressing at least one of emotion, feeling, and objects as information classified into categories in order to generate the multimedia content information, performs first learning, If a sentence including a new word representing at least one of feelings, feelings, and objects that can not be classified through the first learning result is inputted, learning is performed on the new word through the second learning,
The second learning determines whether a specific word is a word expressing at least one of emotion, feeling, and object, classifies the specific word into the most similar existing lower-level characteristic information based on the meaning of the specific word, And if the words are collected more than a threshold number of times as the same expression words, the specific words are classified as one separate lower characteristic information and managed.
delete The method according to claim 1,
Wherein the multimedia content information is generated based on a chat record collected on a broadcast service provided on the basis of the multimedia content service application,
Wherein the chat record is collected during a reaction collection interval set based on a reproduction time of the music,
Wherein the multimedia content information is determined based on a word related to an emotion for the music included in the chat record, a feeling for the music, or information about an object related to the music. The method according to claim 1,
Wherein the multimedia content information is generated based on a chat record collected on a broadcast service provided on the basis of the multimedia content service application,
Wherein the chat record is collected during a reaction collection interval set based on a reproduction time of the music,
Wherein the multimedia content service server divides a plurality of sentences included in the chat record into music expression sentences, other emotional expression sentences, and non-
Wherein the multimedia content service server determines the multimedia content information by considering the music expression sentence,
Wherein the music expression sentence includes a first emotion information, a first feeling information, or a word related to information about the first object, wherein the object of the first emotion information, the first feeling information, It's a musical sentence,
The other emotional expression sentence includes a word related to information on the second emotion information, the second feeling information, or the second object, and the second emotion expression sentence includes a word related to the second emotion information, the second feeling information, This music is not a sentence,
Wherein the non-subject expression sentence includes a word related to third emotion information, third feeling information, or information about a third object, wherein the third emotion information, the third feeling information, Is an unspecified sentence. 5. The method of claim 4,
Wherein the multimedia content service server determines whether information on the third emotion information, the third feeling information, or the third object included in the pseudo-expression sentence is for the music,
Wherein the multimedia content service server determines the multimedia content information by further considering the pseudo-expression expression sentence,
Whether the third emotion information, the third emotion information, or the third object information included in the pseudo-expression sentence is for the music is determined based on the first emotion information included in the music expression sentence, 1 < / RTI > feeling information or information about the first object. A multimedia contents service server for providing user-customized multimedia contents based on multimedia contents information,
A communication unit implemented for communication with a transmitting terminal and / or a receiving terminal; And
And a processor operatively coupled to the communication unit,
The processor receives a request for a multimedia content service application from the transmitting terminal and / or the receiving terminal,
The multimedia content service application is transmitted to the transmitting terminal and the receiving terminal either directly or through a relay server,
Wherein the multimedia content service application provides a music recommendation service to a multimedia content recipient corresponding to the receiving terminal,
The music recommendation service recommends music that matches the music listening preference information of the multimedia content recipient,
Wherein the music listening preference information is generated based on multimedia content information of the music listened to by the multimedia content recipient,
Wherein the multimedia content information includes emotion information on the music, emotion information on the music, and information on objects related to the music,
The processor performs first learning by receiving information on sentences including words expressing at least one of feelings, feelings, and objects as information classified into categories in order to generate the multimedia content information, And a sentence including a new word representing at least one of emotion, feel, and object that can not be classified through the learning result is input, and performs learning on the new word through the second learning,
The second learning determines whether a specific word is a word expressing at least one of emotion, feeling, and object, classifies the specific word into the most similar existing lower-level characteristic information based on the meaning of the specific word, Wherein when the words are collected more than a threshold number of times as the same expression words, the specific words are classified and managed as separate lower-level characteristic information.
delete The method according to claim 6,
Wherein the multimedia content information is generated based on a chat record collected on a broadcast service provided on the basis of the multimedia content service application,
Wherein the chat record is collected during a reaction collection interval set based on a reproduction time of the music,
Wherein the multimedia content information is determined on the basis of a word related to an emotion for the music, a feeling for the music, or information about an object related to the music included in the chat record. The method according to claim 6,
Wherein the multimedia content information is generated based on a chat record collected on a broadcast service provided on the basis of the multimedia content service application,
Wherein the chat record is collected during a reaction collection interval set based on a reproduction time of the music,
The processor divides a plurality of sentences included in the chat record into music expression sentences, other emotional expression sentences, and non-subject expression sentences,
Wherein the processor determines the multimedia content information by considering the music expression sentence,
Wherein the music expression sentence includes a first emotion information, a first feeling information, or a word related to information about the first object, wherein the object of the first emotion information, the first feeling information, It's a musical sentence,
The other emotional expression sentence includes a word related to information on the second emotion information, the second feeling information, or the second object, and the second emotion expression sentence includes a word related to the second emotion information, the second feeling information, This music is not a sentence,
Wherein the non-subject expression sentence includes a word related to third emotion information, third feeling information, or information about a third object, wherein the third emotion information, the third feeling information, Is a non-designated sentence. 10. The method of claim 9,
Wherein the processor determines whether information on the third emotion information, the third feeling information, or the third object included in the pseudo-expression sentence is for the music,
Wherein the processor determines the multimedia content information by further considering the pseudo-expression expression sentence,
Whether the third emotion information, the third emotion information, or the third object information included in the pseudo-expression sentence is for the music is determined based on the first emotion information included in the music expression sentence, 1 < / RTI > feeling information or information about the first object.

Images