KR20200049193A - Method for providing contents and service device supporting the same - Google Patents

Abstract

According to an embodiment of the present invention, a service device capable of supporting activation of content consumption ecosystem may comprise: a memory; and a processor functionally connected to the memory. The processor may be set to check a user consumption history stored in the memory, extract metadata of content corresponding to the user consumption history, generate at least one subject word by considering text information included in the metadata of the content and time consumed for the content obtained through the user consumption history, and generate a content recommendation list based on the subject word to store the content recommendation list in the memory.

Description

Method for providing contents and service device supporting the same}

The present invention relates to a content recommendation, and more particularly, to a content recommendation method and a service device supporting the content that can recommend content closely related to the user based on the user's consumption history and metadata of the content.

In the conventional content viewing environment, the recommendation technology used is machine learning and statistical techniques for recommending content that the user may prefer. For example, the conventional content recommendation technology uses a recommendation technology of a collaborative filtering method that recommends content viewed by other users with similar viewing history to the user, and obtains the similarity between contents by using the information of each viewing user to view the user Techniques for recommending content similar to one content have been mainly used.

The above-mentioned conventional content recommendation results mainly consisted of list-type recommendations sorted by rank, and no reason or additional explanation for the recommendation result to the user was provided. Accordingly, the conventional content recommendation technique often recommends content that is less relevant to the user, and accordingly, the user reliability is very low.

Registered Korean Patent Publication No. 10-2009-033750, April 06, 2009 (Name: Content Play List Recommendation Method and Apparatus)

The present invention is to satisfy the above-described needs, and by classifying the content preferences of the user based on the subject word for each user who can define the viewing characteristics of the user, a content recommendation method and support for determining the more accurate user preference To provide a service device to.

In addition, the present invention provides a content recommendation list based on user topic words that can define a user's content consumption characteristics, thereby providing a basis for content recommendation and supporting a content recommendation method that can improve the reliability of content recommendation and support To provide a service device to.

The service device according to an embodiment of the present invention may include a memory and a processor functionally connected to the memory. The processor checks the user consumption history stored in the memory, extracts the metadata of the content corresponding to the user consumption history, and the text information included in the metadata of the content and the content obtained through the user consumption history It may be configured to generate at least one subject word in consideration of time spent, and generate a content recommendation list based on the subject word and store it in the memory.

Here, the processor assigns a score for each word based on the words included in the text information and the time spent of the content, and the words having a score higher than a reference value or a relatively high score are assigned. Words can be determined as the subject word.

In particular, the processor may be configured to assign a score obtained by multiplying the number of occurrences of the words by the number of times the content is related to the words as the score of each word. Moreover, the processor may be set to give a relatively high weight to a consumption time within a preset time among consumption times of the content.

Meanwhile, the processor may be configured to provide the at least one subject word to the terminal device, and to provide the terminal device with a list of at least one content related to the subject word selected by the terminal device.

The processor detects content included in the user consumption history in connection with the selected subject word, and generates the content recommendation list based on at least one content having a specified similarity or higher to the detected content to the terminal device It can be set to provide.

Further, the processor may group terminal devices having a consumption history similar to the user consumption history, collect similar subject words related to similar consumption history of the grouped terminal devices, and provide the similar subject words to the terminal device Can be set.

Content recommendation method according to an embodiment of the present invention, the service device, checking the user consumption history stored in the memory, extracting the metadata of the content corresponding to the user consumption history, included in the metadata of the content And generating at least one subject word in consideration of text information and time spent of content acquired through the user consumption history, and generating a content recommendation list based on the subject word and storing the content recommendation list in the memory. have.

Here, the generating of the at least one subject word includes extracting words from the text information, multiplying the extracted words by the content consumption time associated with the words, and scoring the words. And determining words within a predetermined rank as the at least one subject word based on the score scored.

Alternatively, the generating of the content recommendation list may include providing the at least one subject word to the terminal device, detecting content included in the user consumption history in connection with the subject word selected by the terminal device, and And generating the content recommendation list based on the detected content and at least one content having a similarity level higher than a specified size.

A computer recording medium storing at least one instruction executed by a processor according to an embodiment of the present invention includes: checking a user consumption history stored in a memory; extracting metadata of content corresponding to the user consumption history; Generating at least one subject word in consideration of text information included in the metadata of the content and time spent of content obtained through the user consumption history, and providing a content recommendation list based on the subject word It may include at least one instruction set to perform.

The present invention can support the activation of the content consumption ecosystem by clearly grasping the content consumption characteristics of users as subject words and discovering and recommending various contents based on the subject words.

1 is a view showing an example of a network environment according to an embodiment of the present invention.
2 is a view showing a schematic form of a service device according to an embodiment of the present invention.
3 is a diagram illustrating an example of a processor of a service device according to an embodiment of the present invention.
4 is a diagram illustrating an example of a method for generating a content recommendation list according to an embodiment of the present invention.
5 is a diagram illustrating an example of a content recommendation method according to an embodiment of the present invention.

Hereinafter, various embodiments of the present invention will be described with reference to the accompanying drawings. However, this is not intended to limit the present invention to specific embodiments, and it should be understood that the embodiments of the present invention include various modifications, equivalents, and / or alternatives. In connection with the description of the drawings, similar reference numerals may be used for similar elements.

1 is a view showing an example of a network environment according to an embodiment of the present invention.

Referring to FIG. 1, the network environment 10 according to an embodiment of the present invention may include a network 150, a terminal device 100 and a service device 200.

Net of the present invention described above? In the environment 10, when the terminal device 100 accesses the service device 200 through the network 150, the service device 200 may provide a content recommendation list designated to the terminal device 100. In this operation, the service device 200 selects a subject word related to the user preference based on the user consumption history associated with the terminal device 100 and the contents metadata related to the user consumption history, and based on the selected subject word A content recommendation list can be generated. The generated content recommendation list may have a high user preference according to the consumption history of the user, and accordingly, the user's confidence in the recommendation list information may be high.

The network 150 is not only an IP-based wired communication network such as an Internet network, but also a long term evolution (LTE) network, a mobile communication network such as a WCDMA network, various types of wireless networks such as a Wi-Fi network, and combinations thereof. It can be made. That is, the network environment 10 related to the content recommendation function according to the present invention can be applied to both wired and wireless communication networks without distinction. Specifically, the network 150 may form a communication channel between the service device 200 and the terminal device 100. For example, the network 150 may support at least one of 3G, 4G, and 5G wireless mobile communication methods that can be operated by the service device 200 or the terminal device 100. Alternatively, the network 150 may form a communication channel between the terminal device 100 and the service device 200 on a wired basis. Such a network 150 should be interpreted as a concept including various wired networks, wireless networks, and combination networks thereof that are currently developed and commercialized or are developed and commercialized in the future.

The terminal device 100 may access the service device 200 through the network 150. The terminal device 100 may receive a content word recommendation list based on a subject word provided by the service device 200 and receive a content lease that can be recommended according to specific item content. In this regard, the terminal device 100 is associated with a terminal communication circuit for accessing the service device 200, a display capable of receiving and outputting a web page or a content search screen provided by the service device 200, and related to a subject word It includes an input unit used to select content recommendation items or a content to be downloaded, a memory to store the received content, and a control unit to control the above-described components, for example, a terminal communication circuit, a display, an input unit, a memory, and the like. Can be. Meanwhile, the terminal device 100 of the present invention may access the service device 200 according to a user input, and perform registration based on the ID and login information in the service device 200. The terminal device 100 may purchase and use various contents provided by the service device 200.

The service device 200 may have a communication standby state so that the terminal device 100 can access through the network 150. When the terminal device 100 accesses, the service device 200 may provide the terminal device 100 with a virtual page that can use the content related service provided by the service device 200 to the terminal device 100. have. The service device 200 while the terminal device 100 performs a task such as purchasing, listening to, presenting, or retrieving content through the virtual page, information related to the terminal device 100 Fields (eg, user's consumption history) can be stored and managed in memory. When the user consumption history related to the terminal device 100 is a specified amount of data (for example, the content purchase history occurs more than a specified number of times, the content is retrieved or presented, or at least a part of the content is listened to) When a history occurs more than a specified number of times), a topic word for recommending content related to the corresponding terminal device 100 may be calculated. Alternatively, the service device 200 extracts the user content consumption history and stores the content consumption time and recent consumption date together, and measures the user's content consumption by measuring the user's word score in consideration of the content consumption time and recent viewing preference. Personalized subject words that can represent can be extracted. Additionally or alternatively, the service device 200 may provide the subject word providing terminal device 100 in a similar user group having a consumption history similar to the user consumption history. The service device 200 may generate a content recommendation list using personalized subject words based on user consumption content, and provide the content recommendation list to the terminal device 100. Alternatively, the service device 200 may provide a content recommendation list and provide specific content to the terminal device 100 in response to the selection of the terminal device 100.

The network environment 10 related to the content recommendation of the present invention described above can provide a scalability related to content recommendation by integrating and analyzing and presenting various analysis information related to content recommendation, and also similar user groups with similar user consumption history Based on, by providing the subject words in the similar user group included in the specific terminal device 100, it is possible to recommend the content closely related to the user's taste or preference. In addition, according to the present invention, when the terminal word 100 selects a subject word, contents corresponding to the subject word are provided for each user by reflecting a user's content consumption tendency, thereby enabling personalized content recommendation distinguished from other users. have.

2 is a view showing a schematic form of a service device according to an embodiment of the present invention.

Referring to FIG. 2, the service device 200 of the present invention may include a communication circuit 210, a memory 240 and a processor 250.

The communication circuit 210 may form a communication channel of the service device 200. For example, the communication circuit 210 may form a communication channel with the terminal device 100 in response to a request for access to the terminal device 100. The communication circuit 210 may provide a virtual page (eg, a content related service page) designated to the terminal device 100 in response to the control of the processor 250. The communication circuit 210 may search for various contents related to the use of a virtual page from the terminal device 100 (for example, search for content by purchasing the input unit of the terminal device 100, purchase, present, listen to at least a part, etc.) And an input message for requesting service use). When the communication circuit 210 receives various messages from the terminal device 100, response messages corresponding to various messages in response to control of the processor 250 (eg, content recommendation list, content list, content, or content streaming) Data, etc.) to the terminal device 100.

The memory 240 may store various data, programs, and algorithms related to the operation of the service device 200. For example, the memory 240 may include a user consumption history 241, content metadata 243, a term set 245, a content list 247, and content 249.

The user consumption history 241 may include various action logs generated when the terminal device 100 accesses the service device 200 and uses content-related services. For example, the user consumption history 241 may include a content usage history including at least one of a content purchase history, a content search history, a content gift history, and a content viewing history provided by the service device 200. . In this regard, the user consumption history 241 may include identification information for each terminal device 100. For example, the user consumption history 241 may include ID and password information corresponding to the terminal device 100. The user consumption history 241 may be classified and stored for each terminal device 100 identification information. Also, the user consumption history 241 may be grouped and classified for each terminal device 100 having a similar usage consumption history. For example, the user consumption history 241 may be grouped and classified according to various criteria such as genre, singer or actor, content usage age group, content production time period, and content usage frequency. Alternatively, the user consumption history 241 may be grouped and classified according to characteristics of the terminal device 100. For example, the user consumption history 241 may be grouped and classified according to a user age group, a user gender, a region, and a usage type (eg, purchase, search, gift, listening, etc.) of the terminal device 100.

The content metadata 243 may include metadata of each content included in the user consumption history 241. For example, the content metadata 243 may include a genre, content, name of a singer or actor, and a local name associated with the content of the content included in the user consumption history 241 of the specific terminal device 100. Alternatively, the content metadata 243 may include adjectives, nouns, adverbs, etc. extracted from subtitle information included in the content.

The term set 245 may include meaningful words among text information extracted from the content metadata 243. For example, the term set 245 may include a person's name, a local name, an adjective or adverb expressing a specific emotion, a verb or onomatopoeia describing a specific situation. Alternatively, the term set 245 may include terms repeatedly appearing a specified number of times or more among various text information included in the content metadata 243. The term set 245 is provided for indexing words, and may be omitted.

The content list 247 may include at least one subject word calculated based on scoring of terms included in the term set 245 and identification information of at least one content related to the subject word. have. The content list 247 includes identification information of subject words and contents related to a specific terminal device 100, as well as at least one subject word and subject related to terminal devices belonging to a similar group to which the terminal device 100 belongs It may include identification information of contents related to a word.

The content 249 may include at least one content data provided to the terminal device 100. For example, the content 249 may include a music file, a video file (picture file or video file), a text file, and the like. The content 249 may be searched, purchased, viewed, or presented to another terminal device 100 according to the request of the terminal device 100.

The processor 250 may perform data transfer or data processing related to the operation of the service device 200. For example, the processor 250 is related to the performance of the content recommendation function of the present invention, the connection of the terminal device 100, the provision of a virtual page, the collection of user consumption history of the terminal device 100, the user consumption history Content metadata collection, creation of term sets based on content metadata, creation of content lists based on term sets, and content recommendation and content provision based on content lists can be handled. In particular, the processor 250 of the present invention collects the user consumption history according to the service use of the terminal device 100 and stores it in the memory 240, and when the content recommendation list is requested, the user stored in the memory 240 Check the consumption history 241, calculate the subject word related to the terminal device 100 based on the metadata of the content corresponding to the user consumption history 241, and based on the subject word, the content recommendation list to the user It can be stored in the memory 240. The content recommendation list generation request is, for example, when the user consumption history 241 of the terminal device 100 is greater than or equal to a specified value (eg, when the service device 200 access is greater than or equal to a specified number of times), the content provided by the service device 200 May be used for a specified number of times or more) or when a content recommendation list request of the terminal device 100 is received.

The processor 250 provides a list of content recommendations stored in the memory 240 to the terminal device 100 upon request to the terminal device 100 or to the terminal device 100 according to the policy of the service device 200 Can provide. When providing the content recommendation list, the processor 250 identifies a similar user group associated with the terminal device 100 and provides a similar subject word related to the similar user group to the terminal device 100 or the similar user The similar content recommendation list generated based on the subject word related to the group may be provided to the terminal device 100. The processor 250 may include a configuration as shown in FIG. 3.

3 is a diagram illustrating an example of a processor of a service device according to an embodiment of the present invention.

Referring to FIG. 3, the processor 250 includes a user information collection unit 251, a metadata collection unit 253, a subject word generation unit 255, a content recommendation unit 257, and a group processing unit 259 can do.

The user information collecting unit 251 may collect a user's content consumption history (eg, the user consumption history 241) and store it in the memory 240. In this regard, the user information collection unit 251 performs a login request when the terminal device 100 accesses, and when login information is provided from the terminal device 100, log-in information (or terminal device identification information). Based on the service history of the service device 200 of the terminal device 100 may be collected. For example, the user information collecting unit 251 may collect at least one of the content information purchased by the terminal device 100, the searched content information, the presented content information, and the viewed content information as the user consumption history 241. have. At this time, the user information collection unit 251 may collect information about the user's consumption by content and recent content consumption time. For example, the user information collection unit 251 may collect a specific content viewing time or a specific content search, purchase, or gift count. Alternatively, the user information collection unit 251 may collect and organize time spent for each content used by the terminal device 100. For example, the user information collecting unit 251 accumulates the total consumption time for each content (for example, the consumption time for viewing / purchasing / gift / search, etc.) and the consumption time within a specified time range (for example, recent week, etc.) , User consumption history 241 for each content may be stored. Alternatively, the user information collection unit 251 may store the user consumption history 241 according to the frequency of the content used by the terminal device 100 (for example, viewing / purchasing / gift / search).

The metadata collection unit 253 may extract metadata information about the consumption content of the user. For example, the metadata collection unit 253 may collect text information (eg, tag information, content information, subtitle information, plot information, etc.) included in contents consumed by the user as metadata information. The metadata collection unit 253 may perform morpheme analysis on the extracted metadata information to extract nouns, adjectives, verbs, and the like. The metadata collection unit 253 may analyze content by prioritizing objective information on content such as genre information, director, performer, and date and time of broadcasting. The method of assigning priority may include a method of giving heuristic weight. The metadata collection unit 253 may generate a vocabulary set based on words extracted from content. The meta data collection unit 253 may convert nouns, adjectives, and verb words in meta data of content extracted for each user into an index based on a set of terms. For example, the metadata collection unit 253 may check words extracted from the content metadata corresponding to a certain index based on a set of terms organized for each index, and collect corresponding index values.

The subject word generation unit 255 checks the user consumption history 241 for each terminal device 100 in the memory 240 and scores words of words existing in contents corresponding to the identified user consumption history 241. It can be calculated based on the following equations (1) and (2).

In Equation 1, the denominator of R _Cr may be the user's total content viewing time. t _Ck may be a time when a user watches a specific content Ck. The molecule of R _Cr is the user's specific content Cr viewing time multiplied by log (1 + wt _Cr ). _Cr wt of log (1+ wt _Cr) is when to take the log value as a numerical representation of the date and time of viewing certain content value Cr, the more recent the content viewing may indicate a value. S _wi shown in Equation 1 may be a formula for calculating a score value for a specific word wi. Equation 1 may refer to the result of adding N * C to all contents of the total number K viewed by the user for all contents. That is, Equation 1 may be a value calculated by multiplying the number of occurrences of a specific word by the content importance value. In Equation 1, N _Ck wi means the number of times a specific word wi was found in the Ck content, and the method of calculating the score of the specific word wi based on the above-described equation is for K content viewed by the user, It may include a method of multiplying the number of times the word wi appears in the specific content _Ck by the specific content weight value R _Ck wi. R _Ck is a ratio value of a specific content Ck, and the R _Ck value may be a value obtained by applying a recent viewing history weighting value to a specific content Ck consumption time compared to a total content consumption time consumed by a user.

As described above, the subject word generation unit 255 analyzes metadata of content corresponding to the user's content consumption history (or user consumption history) and user content preferences reflecting the consumption time confirmed based on the user consumption history. By integrating the analysis, it is possible to generate a subject word that can represent the consumption characteristics of the user. The generated subject words can indicate specific information about the content consumption taste of a specific user, and support to expand the content of other domains based on the subject words.

The content recommendation unit 257 calculates word scores for each user and sorts them in order of highest score, and may provide top N topic words for each user to the user. When the user selects a subject word, the content recommendation unit 257 includes the selected subject word, and extracts contents having high importance for the subject word and provides it to the user. For example, the content recommendation unit 257 provides keywords such as “action” and “exciting” as the subject word to the terminal device 100, and when the keyword “action” is selected by the user of the terminal device 100 , N * R values (word appearance frequency * viewing time-recently considered for recent viewing) of Equation 1 described above among contents including the word “action” among user viewing contents of the terminal device 100, for example, words The content having the greatest value) multiplied by a value obtained by setting the weight of the recent viewing time relatively high among the appearance frequency and the total viewing time may be set as the representative content C of the “action”. With respect to the word “action” selected by the user, contents having the word “action” while having high similarity with the representative content C of the user's “action” word may be sorted and extracted in order of similarity. D, representative contents for a specific word may be different for each user, and thus, different contents may be extracted for each user, that is, even if the same word is selected by the users, the content recommendation unit 257 may have different representative contents. It may give different results.

The group processing unit 259 applies the distributed memory version of Paragraph Vector (PV-DM), Latent Dirichlet Allocation, etc. by inputting the identification information of the content used by each user among the data collected by the user information collection unit 251. To extract user vectors. The group processor 259 may form a similar user group based on the extracted user vector. The group processing unit 259 extracts and provides similar subject words in the similar user group, and when the user selects the similar subject words, the group processing unit 259 may provide content associated with the corresponding word. In relation to providing the similar subject words described above, the group processor 259 may calculate vector similarity based on the vector for each user, and process users having high similarity with a specific user as a similar user group. The group processor 259 sorts the highest multiplied by the score value of the subject words among the subject words of users in the similar user group and the similarity value with a specific user, and then sorts the similar subject words of a certain rank in the terminal device 100 ).

The score for each word for extracting the subject word in the similar user group can be calculated through Equation 3 and Equation 4.

The UG _target may mean a group of similar users of the target user. The wi may mean one of the subject words of users existing in the similar user group. The S _{wi, UGtarget} may mean a score value for the subject word wi existing in the similar user group of the target user. UG _Num is the total number of users in the similar user group, u _a refers to a specific user in the similar user group, and Sim _{ua, utarget} can mean the similarity between the target user and the u _a user. The group processing unit 259 may arrange subject words wi having a large value of S _{wi and UGtarget} and provide them as similar subject words in the similar user group. The Sim _{ua, utarget} may correspond to Sim _uAuB in Equation 3 above. The group processor 259 may calculate similarity by using various similarity measurement methods (Pearson Correlation Coefficient, Euclidean Distance, etc.) in addition to Equation (3).

4 is a diagram illustrating an example of a method for generating a content recommendation list according to an embodiment of the present invention.

Referring to FIG. 4, with respect to a method for generating a content recommendation list, in operation 401, the processor 250 of the service device 200 may check a user consumption history stored in the memory 240. For example, when the terminal device 100 accesses and uses the content, the processor 250 of the service device 200 may store the user consumption history according to the content use in the memory 240. At this time, the processor 250 may store a user consumption history based on identification information (eg, login information) of the corresponding terminal device 100. The processor 250 may check the user consumption history 241 stored in the memory 240 when a designated content recommendation list is requested to be generated.

In step 403, the processor 250 may perform consumption content purification. For example, the processor 250 may accumulate the consumption time for each content, but may give a relatively high weight to the recent usage history. Alternatively, the processor 250 may accumulate the identification information of the used content, but may give a relatively high weight to the recent use history.

In step 405, the processor 250 may collect content metadata. In this regard, the processor 250 may check the user consumption history, and collect metadata (eg, text information related to the content) of contents included in the user consumption history.

In step 407, the processor 250 may perform term set generation. For example, the processor 250 may perform a terminological arrangement on meta data obtained from contents, and assign indexes to each term to perform indexing on the terms. The above-described steps 401 and 403 and the steps 405 and 407 may be simultaneously performed in the service device 200.

In step 409, the processor 250 may perform scoring on terms. The processor 250 may perform scoring on the terms in consideration of the number of appearances of the terms and the consumption time of the content containing the terms.

In step 411, the processor 250 may generate a personalized subject word. For example, the processor 250 may determine terms that fall within a predetermined rank among terms scored through scoring, as personalized subject words.

In step 413, the processor 250 may configure a content list. The processor 250 may identify content having high similarity for each subject word, and construct a content list based on content titles having a high similarity. Alternatively, the processor 250 may construct a content list by constructing a content list corresponding to each subject word, and then collecting content having high similarity (or more than a specified similarity) with each content.

5 is a diagram illustrating an example of a content recommendation method according to an embodiment of the present invention.

Referring to FIG. 5, in connection with the content recommendation method, in step 501, the processor 250 of the service device 200 may check whether there is a connection to the terminal device 100. If there is no connection to the terminal device 100, the processor 250 may process the performance of the designated function in step 503. For example, the processor 250 may receive at least one content from an external electronic device or a content providing device and store it in a memory.

When the terminal device 100 is connected, the processor 250 may extract the subject word related to the terminal device 100 in step 505. For example, the processor 250 may check the user consumption history of the content of the terminal device 100 and collect metadata of the content corresponding to the user consumption history. The processor 250 multiplies the words included in the collected metadata with the total consumption time of the identified content based on the user consumption history, to score points for the words, and at least based on the assigned scores You can create one subject word. For example, the processor 250 may generate words above a specified score as the subject word. Alternatively, the processor 250 may determine words included in a relatively high rank (or predetermined rank) among the words as the subject word. In the application of the total consumption time, the processor 250 may assign a relatively high weight to the consumption time viewed within a designated time based on the current time point and apply it to the score calculation.

In operation 507, the processor 250 may provide the subject word-based list to the terminal device 100. For example, the processor 250 may provide the terminal device 100 with a virtual page including a list including at least one subject word.

In step 509, the processor 250 may check whether a message related to item selection is received from the terminal device 100. In this regard, the terminal device 100 may output the received list of subject words to the display and provide selection information regarding the specific subject word selected according to the user input to the service device 200.

When a message corresponding to the selection of a specific item is received from the terminal device 100, the processor 250 generates a content list based on at least one of the selected item and the user consumption history in step 511 and stores it in the memory 240 Can be. The processor 250 may provide a list of contents stored in the memory 240 to the terminal device 100. In this process, the processor 250 may check content related to the selected subject word. Here, since content related to the selected subject word is checked according to the user consumption history, even if the same subject word is selected, the checked content for each terminal device may be different. The processor 250 may detect the identified content and at least one content having a similarity level greater than or equal to a specified size. Here, the processor 250 may detect content having a similarity level greater than or equal to a specified size based on the identified genre, director, actor, content, and the like.

In step 513, the processor 250 may check whether a content request is received from the terminal device 100. When a content request is received, the processor 250 may provide the requested content to the terminal device 100 in step 515.

The processor 250 may check whether an event related to the termination of the content recommendation function occurs in step 517. If the end event related to the content recommendation function does not occur, the processor 250 may branch before a specific step among the above-described respective steps according to the function performed immediately before. For example, the processor 250 may branch to an item selection operation before, a content request reception step, or the like, and perform the following operation again. When the termination event related to the content recommendation function occurs, the processor 250 may terminate the content recommendation function. According to this operation, the processor 250 may release a communication channel with the terminal device 100 and enter a connection waiting state that the terminal device 100 can access.

The method according to the present invention may be implemented in a form of software readable through various computer means and recorded on a computer-readable recording medium. Here, the recording medium may include program instructions, data files, data structures, or the like alone or in combination. The program instructions recorded on the recording medium may be specially designed and configured for the present invention or may be known and usable by those skilled in computer software. For example, recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes (Magnetic Media), compact disk read only memory (CD-ROM), optical media such as DVD (Digital Video Disk), and optical media. Includes magneto-optical media, such as a floppy disk, and hardware devices specifically configured to store and execute program instructions such as ROM, random access memory (RAM), flash memory, etc. do. Examples of program instructions may include high-level language code that can be executed by a computer using an interpreter, etc., as well as machine language codes made by a compiler. Such a hardware device may be configured to operate as one or more software modules to perform the operation of the present invention, and vice versa.

In addition, implementations of the functional operation and subject matter described in this specification may be implemented in other types of digital electronic circuits, or computer software, firmware, or hardware including the structure and its structural equivalents disclosed herein, or these It can be implemented in combination with one or more of. Implementations of the subject matter described herein are one or more computer program products, that is, one for computer program instructions encoded on a tangible program storage medium to control or thereby execute the operation of the device according to the invention. It can be implemented as the above modules. The computer-readable medium can be a machine-readable storage device, a machine-readable storage substrate, a memory device, a composition of materials affecting a machine-readable propagated signal, or a combination of one or more of these.

In addition, this specification includes details of a number of specific implementations, but these should not be understood as limiting on the scope of any invention or claim, but rather on features that may be specific to a particular embodiment of a particular invention. It should be understood as an explanation. Certain features that are described in this specification in the context of separate embodiments may be implemented in combination in a single embodiment. Conversely, various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable subcombination. Further, although features may operate in a particular combination and may be initially depicted as so claimed, one or more features from the claimed combination may in some cases be excluded from the combination, and the claimed combination subcombined. Or sub-combinations.

Likewise, although the operations are depicted in the drawings in a particular order, it should not be understood that such operations should be performed in the particular order shown or in sequential order, or that all shown actions should be performed in order to obtain desirable results. In certain cases, multitasking and parallel processing may be advantageous. In addition, the separation of various system components of the above-described embodiments should not be understood as requiring such separation in all embodiments, and the described program components and systems will generally be integrated together into a single software product or packaged in multiple software products. You should understand that you can.

Specific embodiments of the subject matter described herein have been described. Other embodiments are within the scope of the following claims. For example, the operations recited in the claims may be performed in different orders while still achieving desirable results. As an example, the process illustrated in the accompanying drawings does not necessarily require that particular illustrated order or sequential order to obtain desirable results. In certain implementations, multitasking and parallel processing can be advantageous.

The described description presents the best mode of the present invention, and provides examples for explaining the present invention and for those skilled in the art to make and use the present invention. This written specification is not intended to limit the invention to the specific terms presented. Therefore, although the present invention has been described in detail with reference to the above-mentioned examples, those skilled in the art can make modifications, alterations and modifications to the examples without departing from the scope of the present invention.

Therefore, the scope of the present invention should not be determined by the described embodiments, but should be determined by the claims.

The present invention is applied to the field of content use, and particularly relates to content recommendation technology.

In particular, the present invention provides a content recommendation technology that provides at least one personalized subject word based on the content consumption history for each user, and provides a similar content list according to the user consumption history, thereby recommending content recommendation with high accuracy for user preference. It can be enabled.

10: network environment
100: terminal
150: network
200: service device
210: communication circuit
240: memory
250: processor

Claims (11)

Memory;
And a processor functionally connected to the memory,
The processor
Check the user consumption history stored in the memory,
Extract the metadata of the content corresponding to the user consumption history,
Generating at least one subject word in consideration of the text information included in the metadata of the content and the consumption time of the content obtained through the user consumption history,
And a content recommendation list based on the subject word. According to claim 1,
The processor
Based on the words included in the text information and the time spent of the content, a score is given for each word, and the subject words are words in which the given score is higher than a reference value or words with a relatively high score. Service device characterized in that it is determined to. According to claim 2,
The processor
A service device characterized in that the words are set to give a score obtained by multiplying the number of appearances and the time spent of content related to the words as the score of each word. According to claim 3,
The processor
Service device characterized in that it is set to give a relatively high weight to the consumption time within a predetermined time among the consumption time of the content. According to claim 1,
The processor
Providing the at least one subject word to the terminal device,
And providing a list of at least one content related to the subject word selected by the terminal device to the terminal device. According to claim 1,
The processor
Set to detect the content included in the user consumption history in relation to the selected subject word, and generate the content recommendation list based on the detected content and at least one content having a specified similarity or higher and provide the content to the terminal device Service device characterized in that. According to claim 1,
The processor
Group terminal devices having a consumption history similar to the user consumption history,
Collect similar subject words related to similar consumption history of the grouped terminal devices,
And providing the similar subject words to the terminal device. Service device,
Checking a user consumption history stored in the memory of the service device;
Extracting metadata of content corresponding to the user consumption history;
Generating at least one subject word in consideration of text information included in the metadata of the content and time spent of the content acquired through the user consumption history;
And generating a content recommendation list based on the subject word. The method of claim 8,
Generating the at least one subject word is
Extracting words from the text information, and multiplying the extracted words by the content consumption time associated with the words to score the words;
And determining words within a predetermined rank as the at least one subject word based on the given score. The method of claim 8,
The step of generating the content recommendation list
Providing the at least one subject word to a terminal device;
Detecting content included in the user consumption history in connection with a subject word selected by the terminal device;
And generating the content recommendation list based on the detected content and at least one content having a similarity of a predetermined size or more. A computer recording medium storing at least one instruction executed by a processor, comprising:
The at least one command
Checking a user consumption history stored in the memory;
Extracting metadata of content corresponding to the user consumption history;
Generating at least one subject word in consideration of text information included in the metadata of the content and time spent of the content acquired through the user consumption history;
And providing a content recommendation list based on the subject word.

Images