KR102049688B1 - User-customized contents providing system using AI - Google Patents

Abstract

The present invention relates to a user-customized content providing system using an artificial intelligence (AI). More particularly, the present invention relates to a user-customized content providing system using an AI, which can present customized content that are optimized for a specific user by reflecting not only the past tendency of the user but also a current environment and using a result of tendency analysis of multiple users, and enable a user to find customized content quickly.

Description

User-customized contents providing system using AI

The present invention relates to a system for providing user-customized content using AI, and more particularly, the presentation of customized content optimized for a specific user by reflecting not only the user's past tendencies, but also the current environment, and the results of analyzing the tendency of a plurality of users. The present invention relates to a user-customized content providing system using AI that enables a user to find customized content in a short time.

With the development of information and communication technologies such as smart phones and communications, anyone can easily acquire various contents regardless of place. Accordingly, a system for storing and providing various contents, such as the following patent documents, has been widely developed. Furthermore, a system for providing differentiated contents for each user by grasping a user's disposition is being developed.

(Patent literature)

Korean Patent Publication No. 10-2009-0006464 (published Jan. 15, 2009) "User-customized content providing apparatus, method and recording medium"

However, the conventional system for providing differentiated content for each user only provides the content by analyzing the user's setting history, usage history, etc., that is, provides the content based on the user's past inclination, and is appropriate for the current situation. There is a problem that can not provide content.

The present invention has been made to solve the above problems,

The present invention provides a user-customized content providing system using AI that reflects not only the user's past inclinations but also the current environment and uses the results of the analysis of the propensity of a plurality of users. have.

In addition, an object of the present invention is to provide a user-customized content providing system using AI that allows a user to find customized content quickly.

The present invention is implemented by the embodiment having the following configuration to achieve the above object.

According to an embodiment of the present invention, a user-customized content providing system according to the present invention is a user terminal that receives content by accessing a content server, and a content server that provides personalized content to the user by exchanging information with the user terminal Includes, the content server is characterized in that to provide the content customized to the user by reflecting the user's information and usage history.

According to another embodiment of the present invention, in the user-customized content providing system according to the present invention, the content server includes an education content unit for providing education content when information requesting education content is output from the user terminal. The education contents unit checks the member information of the user when the information requesting the education content is output from the user terminal, and then selects the member information extraction module that selects the education contents matched among the stored education contents, and the user's interest. It is characterized in that it comprises a usage history extraction module for re-selecting the education content selected in the member information extraction module by checking the previous education content use result and reflecting it.

According to another embodiment of the present invention, in the user-customized content providing system according to the present invention, the education content unit checks the education content use result of the stored multiple users and reflects the education content selected by the user history extraction module. The method may further include a statistical reflecting module for reselecting and providing a statistical reflection module provided to the user terminal, and a statistical generation module for analyzing and using the results of using the educational contents of the plurality of users.

According to another embodiment of the present invention, in the user-customized content providing system according to the present invention, the content server further includes a video content unit for providing video content when information requesting video content from the user terminal is output. The video content unit generates a search condition matching a specific video by analyzing a video storage module for storing the video and its information transmitted from a plurality of user terminals and the storage unit, and analyzing the video and the information stored in the storage unit. And a video analysis module, wherein the video analysis module analyzes a video stored in the storage unit and its information, and extracts a keyword for a specific search item for each section according to the unit time of the video; Combine keywords by search item extracted from the generation module. A search condition generation module for generating a search condition by synchronizing with time information of the video, and storing the search condition for each section of the video generated by the search condition generation module and the keyword for each section of the video extracted from the keyword generation module in the storage unit. Characterized in that it comprises a storage module.

According to another embodiment of the present invention, in the user-customized content providing system according to the present invention, the video content unit compares keywords of videos stored in the storage unit at regular intervals to determine the identity of the video and to recognize the same. It further includes a video trimming module for deleting only one video and deleting the remaining video, wherein the video trimming module is generated by the video analysis module at regular intervals to collect and output keywords of each video stored in the storage unit. A comparison module for comparing the keywords outputted by the keyword identification module, categorizing videos recognized as equal, and setting the same category, and a keyword of one search item from each of the videos included in the same category. Extract in chronological order And a keyword sorting module, wherein the keyword sorting module compares the playing time of the videos included in the same category, selects the video having the longest playing time, and outputs the information by the play time module; The full keyword module extracts and outputs a keyword of one search item in chronological order for all videos selected as having the longest playing time, except for the videos extracted by the full keyword module among the videos included in the same category. And a partial keyword module for extracting and outputting a keyword of one search item in chronological order only for a partial section of the video for each video.

According to another embodiment of the present invention, in the user-customized content providing system according to the present invention, the video cleaning module does not recognize the sameness by comparing keywords associated with time for each of the videos output from the keyword alignment module. Value judging module that selects only one video determined to have the highest value by comparing each other with the judging module for removing the video from the same category and the videos included in the same category after the operation of the judging module. And a removal module for deleting only one video determined to have the highest value output by the value determining module and deleting the remaining videos in the same category from the storage unit.

According to another embodiment of the present invention, in the user-customized content providing system according to the present invention, the keyword identification module outputs a keyword of a specific video by combining keywords of each section, and the comparison module is a keyword of one video. It is judged that the sameness is recognized when the keyword of one video is included in the keyword of another video as well as when it is the same as the keyword of another video, and the adjustment judging module compares the keyword associated with time to determine the content of the keyword. In other words, the order of the keywords is also compared, and the value judgment module determines that the video having the longest playing time is the highest value among several videos, and when there are a plurality of videos having the longest playing time, the first video is created later. I think it's worth more than a video Screening the phase will be characterized in that it outputs the information.

The present invention can obtain the following effects by the configuration, combination, and use relationship described above with the present embodiment.

The present invention reflects not only the past propensity of the user but also the current environment, and using the results of the propensity analysis of the plurality of users, it is possible to present customized content optimized for a specific user.

In addition, the present invention has the effect of allowing the user to quickly find the customized content.

1 is a block diagram of a content providing system according to an embodiment of the present invention.
FIG. 2 is a block diagram illustrating a detailed configuration of the educational content unit of FIG. 1. FIG.
3 is a block diagram showing a detailed configuration of a medical content unit of FIG.
4 is a block diagram illustrating a detailed configuration of a video content unit of FIG. 1.
5 is a block diagram showing a detailed configuration of a video analysis module of FIG.
6 is a block diagram illustrating a detailed configuration of a video arranging module of FIG. 4.
7 is a block diagram illustrating a detailed configuration of a keyword sorting module of FIG. 6.

Hereinafter, exemplary embodiments of a user-customized content providing system using an AI according to the present invention will be described in detail with reference to the accompanying drawings. In the following description of the present invention, if it is determined that a detailed description of a known function or configuration may unnecessarily obscure the subject matter of the present invention, the detailed description thereof will be omitted. Throughout the specification, when a part is said to "include" any component, which means that it may further include other components, except to exclude other components unless specifically stated otherwise, also described in the specification The terms "... unit, ... module" and the like refer to a unit that processes at least one function or operation, which may be implemented by hardware or software or a combination of hardware and software.

Referring to Figures 1 to 7 the user-customized content providing system using the AI according to an embodiment of the present invention, the content providing system is connected to the content server 1 and the user terminal 100 receives the content and A content server 1 that exchanges information with the user terminal 100 to provide content customized for the user, and exchanges information with the content server 1, wherein the content server 1 is customized to the user. And an information providing system 200 for producing information used for providing the content.

The user terminal 100 is configured to receive content by accessing the content server 1, to transmit information of a user, and to receive and display content generated by the content server 1. 1) A terminal capable of mounting an app for exchanging information (hereinafter, referred to as an “information app”) may be used, for example, a smartphone, a tablet, a PDA, a PC, or the like may be used.

The content server 1 is a configuration for exchanging information with the user terminal 100 to provide content customized for the user, for example, to provide content customized for the user by reflecting the user's information and usage history. . The content server 1 includes a transceiver 11, a member 12, an education content 13, a medical content 14, a video content 15, a storage 16, and a controller 17. ), And the like.

The transceiver 11 is connected to the user terminal 100 and the information providing system 200 by the Internet, and transmits and receives information.

When the user terminal 100 outputs information for requesting the user's membership, the member registration unit 12 checks the information necessary for membership registration, completes the membership registration, and stores the member information.

The education content unit 13 is configured to provide education content when information requesting education content is output from the user terminal 100. The member information extraction module 131, the usage history extraction module 132, and statistics Reflecting module 133, statistics generating module 134 and the like.

The member information extracting module 131 checks the user's member information when the information requesting the educational content is output from the user terminal 100, specifies the member's interests, and then matches the education contents stored in the storage unit. It is a configuration that selects contents. For example, if the member information stored in the storage unit includes English as an interest, when the information for requesting education contents is output through the user terminal 100, the member information extraction module 131 may be stored in the storage unit. Select educational content for education.

The usage history extraction module 132 is configured to check the user's previous education content use result stored in the storage unit and to reselect the education contents selected by the member information extraction module 131 to reflect the result. For example, if it is determined that the basic level of English learning has been made as a result of confirming the user's previous education content use result stored in the storage unit, the usage history extraction module 132 may select English education selected by the member information extraction module 131. You can re-select educational content for intermediate level English language education.

The statistical reflecting module 133 confirms the results of using the education contents of the plurality of users stored in the storage unit, reflects this, and reselects the education contents selected by the use history extraction module 132 to the user terminal 100. It is a configuration to provide. For example, A, B, and C were selected as training contents for the intermediate level English education in the usage history extraction module 132. As a result of confirming the results of using the training contents (statistics) of the plurality of users stored in the storage unit, A, If it is confirmed that the use of B, C in order, the statistical reflecting module 133, the user terminal 100 will provide the educational content in the order of A, B, C.

The statistical generation module 134 is configured to analyze the results of using the educational content of a plurality of users to generate the results (statistics) and store them in the storage unit.

The medical content unit 14 is configured to provide medical contents to the user terminal 100, and includes a medical data generation module 141 and a medical information providing module 142.

The medical data generation module 141 is configured to generate disease information indicating a risk of a specific disease by analyzing the information output from the information providing system for providing medical information. The weather information extraction module 141a and the medical information extraction module 141b, matching information module 141c, and the like.

The meteorological information extraction module 141a is configured to extract and output weather information of a predetermined period of time from the information providing system 200 providing the meteorological information with a predetermined period, and the medical information extraction module 141b is a constant. It is a configuration for extracting and outputting medical information for the unspecified number of patients of the past predetermined period from the information providing system 200 for providing the medical information at regular intervals.

The matching information module 141c matches weather information extracted from the weather information extraction module 141a with medical information extracted from the medical information extraction module 141b to indicate a risk of a specific disease according to a specific weather condition. It is a configuration for generating disease linkage information.

The medical information providing module 142 is configured to provide medical contents by using the data generated by the medical data generating module 141. The user disease confirmation confirms and outputs past disease information of the user stored in the storage unit. Module 142a, a weather information check module 142b for checking current weather information and outputting the weather information, and the weather output information associated with the user disease check module 142a and the weather information check module 142b. The disease risk checking module 142c may be provided to calculate and provide a specific disease risk level of the current user.

The video content unit 15 is configured to provide video content when information requesting video content is output from the user terminal 100. The video storage module 151, the video analysis module 152, and the video trimming module are provided. 153, a video playback module 154, a download module 155, and the like.

The video storage module 151 is configured to store a video and its information transmitted from the user terminal 100 in a storage unit. A plurality of users can transmit their own video from the content server 1 and allow other users to use it, and when the video is transmitted, the introduction information of the video is transmitted together.

The video analysis module 152 is configured to generate a search condition that matches a specific video by analyzing the video stored in the storage unit and its information, and includes a keyword generation module 152a, a search condition generation module 152b, and a storage module. 152c and the like.

The keyword generation module 152a is configured to extract a keyword for a specific search item for each section according to a unit time of the video by analyzing the video stored in the storage unit and its information. The search item may include at least one of a photographing time, a photographing place, a person, and a conversation. For example, if there is an 8-minute video and the unit time is 2 minutes, the video is divided into 4 sections and a keyword for a search item is extracted for each section. The keyword of the search item may be extracted from a video being played through a known face recognition technology, an object recognition technology, a voice recognition technology, a caption recognition technology, or the like, and may include attributes of the video (digital data property) and introduction information of the video. It is also possible to extract from. For example, the unit time is 5 minutes and the search item is set to the character, the shooting date, the shooting location, and the conversation contents. In the 1 hour A movie, the characters B and C appear between 0 and 5 minutes and the shooting date is 2005. When the location is D and January 1, and the dialogue including the words E, F, G is made, the keyword generation module selects B, C as the keyword of the character item in the first section of the video. And extract January 1, 2005 as a keyword of the shooting date item, extract D as a keyword of the shooting place item, and extract E, F, and G as keywords of the dialogue content item.

The search condition generation module 152b is configured to generate a search condition by combining keywords for each search item extracted by the keyword generation module 152a and synchronizing with time information (section according to unit time) of the video. The search condition generation module 152b may generate various search conditions by synchronizing the keyword for each search item extracted by the keyword generation module 152a in various ways and synchronizing with the time information (section according to unit time) of the video. have. For example, by using the keywords extracted in the above example, the search condition synchronized to the first section of the A video is search condition 1 ((B or C) and (E or F or G)) and search condition 2 ((B and C ) and D and (E or F or G)).

The storage module 152c stores the search condition for each section of the video generated by the search condition generation module 152b and the keyword for each section of the video extracted by the keyword generation module 152a in the storage unit.

The video arranging module 153 is generated by the video analyzing module 152 at regular intervals and compares the keywords of the videos stored in the storage to determine the sameness of the videos, and only one video is left when the same is recognized. The video delete module includes a keyword checking module 153a, a comparison module 153b, a keyword sorting module 153c, an adjustment determination module 153d, a value determination module 153e, a removal module 153f, and the like. .

The keyword checking module 153a is configured to collect and output keywords of each video generated by the video analyzing module 152 at predetermined intervals and stored in the storage unit. Since the storage unit stores keywords for a search item for each section according to the unit time of the video, the keyword checking module 153a combines the keywords for each section to output the keyword of the specific video.

The comparison module 153b is configured to compare the keywords output by the keyword checking module 153a and to categorize moving images recognized by the same to set the same category. For example, assuming that the A, B, C, D, and E video are stored in the storage unit, the video A, C, and D are recognized as the same, and that the video B and E are recognized as the same. D forms the same category, and videos B and E form the same category. The comparison module 153b determines that the sameness is recognized not only when the keyword of one video is the same as the keyword of another video but also when the keyword of one video is included in the keyword of the other video. If a, b, and c, but the keywords of the B video is a, b, c, d, the comparison module 153b determines that the A video is identical to the B video. The comparison module 153b determines that equality is recognized not only when the keywords of the videos are the same but also when the keywords of one video are all included in the keywords of the other video, thereby effectively removing not only the same video but also a part of the overlapping video. You can do it.

The keyword sorting module 153c is configured to extract and output a keyword of one search item from each of the videos included in the same category in chronological order, and includes a playback time module 1531, a full keyword module 1532, and a part. Keyword module 1533 and the like.

The play time module 1531 is configured to compare the play times of the videos included in the same category, select a video having the longest play time, and output the information.

The full keyword module 1532 is configured to extract and output keywords of one search item in chronological order for the entire video selected by the playback time module 1531 to have the longest playback time. When there are a plurality of videos selected by the play time module 1531 as having the longest play time, the full keyword module 1532 extracts only one of the videos.

The partial keyword module 1533 sets a keyword of one search item only for a partial section of the video for each of the videos except the video extracted by the full keyword module 1532 among the videos included in the same category. It is the configuration to extract and output in order. For example, there are videos of A, B, and C. A video has the longest playing time, and in the case of video A, a, b, c, d, and e for the keyword of a conversation item (corresponding to one search item). , f, g, h, and i exist in chronological order, and c, d, e, f, g, h, and i exist in chronological order for keywords in the conversation content item of B movie. Assuming that a, b, c, d, e, f, and g exist in the keyword of, the play time module 1531 outputs information of the video A, and the full keyword module 1532 displays the video A. For a, b, c, d, e, f, g, h, i will be output in order, the partial keyword module 1533 will output c, d, e for the B video in order, A, b, c, d can be output in order for the C video.

The adjustment judging module 153d is configured to remove keywords from the same category by comparing the keywords associated with time for each of the videos output from the keyword sorting module 153c. Comparing the keywords associated with time by the adjustment determination module 153d means comparing not only the contents of the keywords but also the order of the keywords. For example, in the keyword sorting module 153c, the keywords of the A video are a, b, When the keywords of the video B are output in the order of time c, b, and a, the adjustment determination module 153d determines that the identity of the video A is not recognized. The adjustment judging module 153d may include not only the keywords listed in the chronological order of one video but the same as the keywords listed in the chronological order of another video, as well as the keywords listed in the chronological order of the one video. It is determined that the sameness is recognized even when included in all of the keywords listed in chronological order. When determining the sameness of the video only by the comparison module 153b, it is not possible to exclude the case where the video content is different but by chance the same video content is determined to be the same video, the keyword sorting module 153c is in chronological order. The keywords listed according to the output are output, and the adjustment judging module 153d compares the keywords associated with time to determine the sameness. Therefore, the contents of the keywords are the same but the keywords are different, that is, the contents of the videos are different but the keywords are the same. This can be effectively prevented from being treated as the same moving picture. In addition, the keyword sorting module 153c extracts the keyword from the entire video in the case of the video having the longest playback period when extracting the keyword of each video, but extracts the keyword of only a portion of the video for the remaining videos. The time taken to extract keywords can be effectively reduced.

The value judging module 153e compares videos included in the same category after the operation of the adjustment judging module 153d and selects only one video determined to have the highest value and outputs the information. For example, the value judging module 153e determines that a video having the longest playing time is the highest value among several videos, and when there are a plurality of videos having the longest playing time, the first video is judged to have a higher value than a later video. Selects only one video and outputs the information.

The removal module 153f is configured to delete the remaining videos in the same category from the storage unit, leaving only one video determined to have the highest value output by the value determining module 153e. The video arranging module 153 deletes a video that is recognized to be identical at predetermined intervals from the storage unit, so that a storage space of the storage unit can be secured and a time for searching a video can be shortened.

The video playback module 154 is configured to allow a video matching the search condition to be played in the user terminal 100 when information corresponding to a specific search condition is output from the user terminal 100. Check whether the video search information input from the user terminal 100 matches a specific search condition, and if the video search information matches a specific search condition, convert digital data of a specific section of the video synchronized to the specific search condition into streaming data. Is configured to transmit to the user terminal (100). The search condition generation module 152b generates a search condition by synthesizing the keyword for each search item extracted by the keyword generation module 152a and synchronizing with the time information (section according to the unit time) of the video (that is, specifying The part of the video that matches the search condition is the part corresponding to a specific section, not the whole), and only the specific section of the video is played instead of the whole specific video. It becomes possible to confirm in a short time.

The download module 155 includes information for requesting the download of the video (the whole video) from the user terminal 100 after a partial section of the video is played by the video playback module 154 on the user terminal 100. When the output is calculated, the cost required to download the video is calculated and the digital data for the video is provided to the user terminal 100 when the payment is confirmed.

The storage unit 16 stores information necessary for providing content and information generated during an operation of the content providing system. The controller 17 controls the overall operation of the content server.

The information providing system 200 is configured to produce information used by the content server 1 to provide content customized for a user, and a conventional system for generating and providing information such as medical care and weather may be used. Can be.

In the above, the Applicant has described various embodiments of the present invention, but these embodiments are merely one embodiment for implementing the technical idea of the present invention, and any changes or modifications may be made to the present invention as long as the technical idea of the present invention is implemented. It should be interpreted as falling within the scope of.

11: Transmitter / Receiver 12: Member Registration 13: Education Content Department
14: medical content unit 15: video content unit 16: storage unit
17: control unit 131: member information extraction module 132: usage history extraction module
133: statistical reflection module 134: statistics generation module 141: medical data generation module
142: medical information providing module 151: video storage module 152: video analysis module
153: video trimming module 154: video playing module 155: download module

Claims (7)

A user terminal receiving content by accessing a content server, and a content server exchanging information with the user terminal to provide contents customized to the user,
The content server includes an education content unit providing education content when information requesting education content is output from the user terminal, and a video content unit providing video content when information requesting video content is output from the user terminal. Include,
The education content unit checks the user's member information when the information requesting the education content is output from the user terminal, identifies the member's interests, and then selects the contents of the education contents stored in the education contents matching module. It includes a usage history extraction module for re-selecting the education contents selected from the member information extraction module to check and reflect the previous educational content use results of the,
The video content unit analyzes a video stored in the storage unit to store the video and its information transmitted from a plurality of user terminals, and the video analysis to analyze the video stored in the storage unit and its information to generate a search condition matching a specific video Module and a video arranging module for comparing the keywords of the videos stored in the storage unit at regular intervals to determine the sameness of the videos and deleting only one video and deleting the remaining videos when the same is recognized.
The video analysis module analyzes a video stored in the storage unit and its information, and generates a keyword for extracting a keyword for a specific search item for each section according to the unit time of the video, and the search item extracted from the keyword generation module. A search condition generation module for combining search keywords and synchronizing with time information of a video to generate a search condition, a search condition for each section of a video generated by the search condition generation module, and a section keyword of the video extracted from the keyword generation module. It includes a storage module for storing the storage unit,
The video trimming module is generated by the video analyzing module at regular intervals and collects and outputs keywords of each video stored in the storage unit, and compares the keywords output by the keyword checking module to recognize the sameness. And a keyword sorting module for extracting and outputting a keyword of one search item from each of the videos included in the same category in chronological order.
The keyword sorting module may be configured to compare the playing times of the videos included in the same category, select a video having the longest playing time, and output the information, and select the longest playing time by the playing time module. Part of the video for each of the videos except for the video extracted by the full keyword module among the videos included in the same category, and extracts the keyword of one search item in chronological order for the entire video It includes a partial keyword module that extracts and outputs keywords of one search item in chronological order only for the section.
The video arranging module compares keywords associated with time for each of the videos outputted from the keyword arranging module, and adjusts the decision judging module to remove videos that are not recognized from the same category from the same category. Compares the videos included in the category with each other and selects only one video determined to have the highest value and outputs the information, leaving only one video determined to have the highest value output by the value determining module. Further comprising a removal module for deleting the remaining videos in the same category in the storage unit,
The keyword checking module outputs a keyword of a specific video by combining keywords of each section, and the comparison module includes not only the keyword of one video but the keyword of one video as the keyword of another video. If the same thing is determined to be the same,
Comparing the keywords associated with the time means that the adjustment judgment module compares not only the contents of the keywords but also the order of the keywords. The value judgment module determines that the video having the longest playing time is the highest value among several videos. , If there are a plurality of videos having the longest playing time, it is determined that the first video is more valuable than the later created video, and the user selects one video and outputs the information. delete The method of claim 1, wherein the education content unit
Check the results of using the stored training content of the plurality of users, and reflects the statistical analysis module provided to the user terminal by re-selecting from the training content selected by the use history extraction module, and analyzes the results of using the training content of the plurality of users A user-customized content providing system, characterized in that it further comprises a statistics generation module for generating and storing the results. delete delete delete delete

Images