KR20110076312A - Method of music recommendation using context information - Google Patents

Abstract

PURPOSE: A sound source recommending method using situation information is provided to recommend a sound source by the low costs, is provided not to be biased to a center of current popular music and is provided to recommend a sound source which is suitable according to the situation of a user. CONSTITUTION: A music recommendation request message including information and accident information is received from a user terminal(S300). A recommendation sound source is selected by using the location information, accident information, and user terminal information(S302). A message included in information about the selected sound source is transmitted to the user terminal(S304). The user terminal information includes information about weight about preferred item transmitted form the user terminal among a preference item including singer, genre, and title and history information including downloading time and sound source which the user terminal downloaded.

Description

Sound recording recommendation method using context information {Method of Music Recommendation using Context Information}

The present invention relates to a sound source recommendation method, and more particularly, to a sound source recommendation method using information such as user playback history, user preferences, user context information, and the like.

Digital music databases are growing and the demand for effective digital music retrieval, automatic music summaries, and music recommendation systems is growing, and in particular, the demand for automated music recommendation systems that can generate commercial value is growing. .

Common methods for recommending music are Pandora and Last.FM. Pandora adopts an item-based recommendation scheme. Music experts adopt a method of analyzing all service songs based on about 400 music attribute values in advance and DBizing them, and then replaying similar songs continuously when a consumer inputs sheet data.

Last.FM adopts a recommendation method using collaborative filtering (CF). The collaborative filtering method recommends music that other users with similar tastes listen to. Panda is an item-based recommendation method, whereas Last.FM recommends it based on user usage statistics.

In the case of Pandora, the similarity between music is calculated by extracting feature vectors of lower levels (for example, physical levels of signals such as energy and peak values) from the music signal. As described above, in the case of recommending similar music by finding similarity between music, there is an advantage that the user's preference can be reflected by recommending music similar to user's query, but the lower level (for example, energy, peak) Since a feature vector of a simple physical level of a signal such as a value is extracted and processed, the accuracy is lowered.

In the case of Last.FM, the music download list of users can be compared with each other to extract a user group in which downloads of the same songs overlap a lot. Based on this, a complete list of downloads / streamings of a similar user group is constructed, and then compared to this list, the song is not recommended to the user list. In this case, a group may be formed based on the user's music download list, but the signal distribution technology simply classifies the mood of the user's owned or downloaded music into joy, sadness, cheerfulness, etc. Can be calculated. Based on this, users' tastes may be grouped to recommend music held by a group having the same taste to other users of the same group.

 Such a method of recommending music through user grouping through a user's existing listening song or a list of downloaded songs is, for example, a user's possession of A or a distribution of existing listening songs is maintained by User B or distribution of existing listening songs. Assuming that is similar to, a new listening song of user B is recommended to user A, or simply classified based on the mood of music.

This may reflect some of the user's preferences for music, but ultimately, it is not recommended for the user's preference of the music (eg, rhythm, timbre, etc.) Since the music is selected by other users using the distribution of music, the accuracy is very low, and there is a problem in that the characteristics of the music preferred by the user (eg, rhythm, tone, etc.) cannot be properly reflected. In addition, the manner in which the user-owned music is simply classified into four or five categories and the users are grouped in the distribution thereof to recommend the music held by each other is highly accurate according to the ambiguity and diversity of the music atmosphere. There is a problem falling.

However, since the user's preferred intention is often based on the song characteristics itself, it may be more effective to grasp the music characteristics and to search for and recommend music similar to the characteristic patterns of the user's query music or the user's past queries.

In addition, Pandora, an item-based recommendation method, can analyze a lot of music by hand, so it is possible to build a highly sophisticated music metaDB and recommend various songs based on this. There is a limit that it is not easy to scale up.

Since Last.FM recommends based on user usage statistics, it is not necessary to build a music metaDB in advance, but it has a disadvantage in that recommended songs tend to be centered on current popular songs.

An object of the present invention is to propose a method for recommending a sound source suitable for a user in consideration of a user's situation.

The problem to be solved by the present invention is to propose a method for recommending a sound source more suitable to the user from a vast amount of information and services.

The problem to be solved by the present invention is to propose a method for recommending a sound source at a low cost.

The sound source recommendation apparatus of the present invention for achieving the above object is to the user's preferred items, including user profile metadata database (DB) that stores the user's personal information, artists, genres, titles, etc. A user preference metadata DB for storing weights, a user context information analyzer for analyzing situation information including time information for requesting sound source recommendation, location information, and event information, and the user profile metadata DB and the user preference metadata DB It includes a user database for storing information about the user-specific recommended sound source selected by the stored information and the information analyzed by the user context information analysis unit.

The sound source recommendation method of the present invention for achieving the above object comprises the steps of receiving a sound source recommendation request message including location information and event information from the user terminal, by using the location information and the event information and the information of the user terminal previously stored; Selecting a sound source to be recommended, and transmitting a message including information on the selected sound source to the user terminal.

The present invention recommends a sound source suitable for a user in consideration of a user's situation, and recommends a sound source more suitable for a user from a vast amount of information and services. In addition, the present invention can recommend a sound source at a low cost, there is an advantage that can recommend the appropriate sound source according to the user's situation without bias toward the current popular songs.

The foregoing and further aspects of the present invention will become more apparent through the preferred embodiments described with reference to the accompanying drawings. Hereinafter will be described in detail to enable those skilled in the art to easily understand and reproduce through this embodiment of the present invention.

The sound source recommendation method recommends a sound source that the user prefers in consideration of a user profile, a user preference, a history list of playing the user audio content, recommended music of experts, and the like. 1 is a diagram illustrating a configuration of a sound source recommendation system and a flow of content and metadata according to an embodiment of the present invention.

1, the solid line shows the flow of metadata, and the dotted line shows the flow of contents. 1 includes a content generating unit 100, a metadata editing unit 102, a metadata collecting unit 104, a content storing unit 106, and a user input unit 108.

The content creation unit 100 generates content or a program, and generates information about the content and the program. Metadata Editing 102 provides metadata editing for users who will use the generated content, and Metadata Aggregation 104 allows independent content creators and developers to maintain the system. Collect the provided content. The user input unit 108 may input a title, singer, composer, genre, etc. of a favorite audio at the time of user registration to obtain user preference information, and the content storage unit 106 may include a content storage unit 106. The selected program is stored for viewing at any time.

Referring to FIG. 1, the metadata moves to the content generating unit 100, the metadata editing unit 102, the metadata collecting unit 104, and the user input unit 108, and the contents are stored in the content generating unit 100 and the content storage unit. The unit 106 moves to the user input unit 108.

2 is a block diagram of a sound source recommendation database for recommending sound sources using context information according to an embodiment of the present invention. Referring to FIG. 2, the database manager 200 includes a metadata manager 210 and a user manager 220. The database manager 210 includes a content list file unit 212, a content metadata database 214, and an audio content database 216. The user manager 220 includes a user list file unit 222 and a user database 224, and includes a user context information analyzer 226, a user profile metadata database 228, and a user history metadata database. 230, user preference metadata database 232.

According to FIG. 2, the present invention enables quick searching by separately configuring a metadata DB of a user profile, a user preference, and a user history. In addition, the present invention provides a sound source suitable for a user by finally analyzing the user's situation according to time information, location information, and event information occurring to the user at the time the user requests audio content.

The user manager 220 manages a mapping table that stores user IDs, passwords, and device information. The mapping table stores connection relations so that users can easily access or search the user profile, user history, and user preference of the user DB 224 for each user, and manage user device information. In addition, the user manager 220 also manages an active user table that manages information on the situation of the currently connected user. The active user table includes a list of contents created by considering the time information, location information, and event information of the currently connected user, information related to the content being used by the user, and a stop point of content that the user has stopped watching. Record user status.

The metadata manager 210 manages a file that stores a connection relationship so that content metadata of the content DB 216 can be easily accessed or searched. The content DB separately configures an audio content database 216 for storing audio content files and a content metadata database 214 having metadata for each audio content. The metadata management unit 210 creates and manages a mapping table for easily accessing and searching the content metadata for quick search since the content metadata having the detailed information of the content is also a file consisting of xml.

The user profile metadata DB 228 stores and manages personal information such as age, age, and gender of the user in an XML file in the database. The user preference metadata DB 232 may include a favorite singer, composer, genre, title, keyword, and weights for each favorite item (Very High, High, Low, Very Low) that the user inputs when registering the server for the first time. Store and manage information in a database as an XML file. The user history metadata DB 230 stores and manages a list of audio contents recently streamed from the server by the user accessing the server and a time when the user used the audio contents.

The content database consists of a database that stores audio content files of various formats and a content metadata database that stores detailed information such as artists, composers, genres, titles, and keywords for each audio content.

The user context information analyzer 226 analyzes the time information, location information, and event information in consideration of the situation. The time information reflects the user's propensity over time for the music and cannot represent the time division numerically, but indicates any particular period or time. For example, a user may express a commonly used time such as morning, lunch, dinner, spring, and summer, and use the fuzzy set to convert a specific time to a specific time by using a fuzzy set. For example, spring maps from March to May.

Location information can use the location information of LBS, such as on the beach, mountains, roads, buildings, etc., and can distinguish and process meaningful place information such as restaurants, libraries, bathrooms, and living rooms. The event information links the user's personal calendar to allow the user to recommend sound sources suitable for the situation information in consideration of personal history such as anniversaries, events, business trips, and meetings.

It is possible to extend the sound source recommendation method using the user's profile information, preference, and history information to search multimedia contents, and to apply to medical service or navigation service using time, location, and personal event information.

3 illustrates a sound source recommendation database storing sound sources and sound source recommendation information and a terminal requesting sound sources to a sound source recommendation database according to an embodiment of the present invention. Hereinafter, a message transmitted and received between the sound source recommendation database and the terminal will be described with reference to FIG. 3.

In step S300, the terminal transmits a sound source recommendation request message requesting the transmission of the recommended sound source to the sound source recommendation database according to the user's request. The terminal transmits the location information of the current terminal is located in the sound source recommendation request message. To this end, the terminal receives the location information of the terminal from the GPS, and transmits the received location information to the sound source recommendation database. Alternatively, the terminal receives the schematic location information of the terminal from the base station currently connected, and transmits the received rough location information to the sound source recommendation database. In addition to the above-described method, the terminal receives the location information using another method and transmits the received location information to the sound source recommendation database. For example, the terminal displays information requesting current location information on the display unit. When the user inputs the current location information through the input unit, the terminal transmits the input location information to the sound source recommendation database.

The terminal transmits event information to the sound source recommendation database in addition to the location information. Of course, the sound source recommendation database may store the event information of the terminal in advance, and when the user inputs the event information to the terminal, the sound source recommendation database may receive the event information in real time.

The terminal transmits the event information related to the anniversary, event, business trip, meeting, etc. stored in the sound source recommendation request message. The terminal may transmit all of the event information such as an anniversary, an event, a business trip, a meeting, etc. stored therein to a sound source recommendation database, or may transmit event information such as an anniversary, an event, a business trip, a meeting related to the current day, to a sound source recommendation database.

In step S302, the sound source recommendation database selects a sound source to be transmitted to the terminal according to a stored process by using the received position information, event information, and time information of the terminal.

As described above, the user expresses a time that can be commonly used, such as morning, lunch, dinner, spring, and summer, and converts an unclear time concept into a specific time by using a fuzzy set. For example, spring maps from March to May.

Location information can use the location information of LBS, such as on the beach, mountains, roads, buildings, etc., and can distinguish and process meaningful place information such as restaurants, libraries, bathrooms, and living rooms. The event information links the user's personal calendar to allow the user to recommend sound sources suitable for the situation information in consideration of personal history such as anniversaries, events, business trips, and meetings.

In step S304, the sound source recommendation database transmits the selected sound source to the terminal. The sound source recommendation database may transmit only one sound source to the terminal or at least two sound sources according to the set method.

The sound source recommendation database may directly provide a sound source to the terminal or transmit only information about the sound source. The terminal receiving only information on the sound source may download the sound source through a predetermined procedure.

The terminal may play the downloaded sound source according to the user's setting, or use the ring tone or coloring, or other sound source required by the terminal.

In addition, the terminal may automatically transmit the sound source recommendation request message in a specific cycle according to the user's setting, or the sound source recommendation database may be automatically transmitted to the terminal in a specific cycle unit. As described above, when the terminal transmits the daily sound source recommendation request message, the terminal may update the ringtone or coloring daily by using the received sound source.

So far I looked at the center of the preferred embodiment for the present invention. Those skilled in the art will appreciate that the present invention can be implemented in a modified form without departing from the essential features of the present invention. Therefore, the disclosed embodiments should be considered in an illustrative rather than a restrictive sense. The scope of the present invention is shown in the claims rather than the foregoing description, and all differences within the scope will be construed as being included in the present invention.

 1 is a diagram illustrating a configuration of a sound source recommendation system and a flow of content and metadata according to an embodiment of the present invention.

2 is a block diagram of a sound source recommendation database for recommending sound sources using context information according to an embodiment of the present invention;

3 illustrates a sound source recommendation database storing sound sources and sound source recommendation information and a terminal requesting sound sources to a sound source recommendation database according to an embodiment of the present invention.

Claims (8)

A user profile metadata database (DB) that stores personal information of the user; A user preference metadata DB for storing a weight of a user's favorite item among preference items including a singer, a genre, and a title; A user context information analyzer configured to analyze context information including time information, location information, and event information that has requested sound source recommendation; And a user database that stores information about the user-specific recommended sound source selected by the information stored in the user profile metadata DB, the user preference metadata DB, and the information analyzed by the user context information analysis unit. Sound source recommendation apparatus to assume. According to claim 1, The sound source recommendation device, An audio content DB storing a sound source; And a content metadata DB for storing metadata about the sound source. According to claim 2, The sound source recommendation device, And a user history metadata DB for storing history information including a sound source downloaded by each user and a time for downloading the sound source. According to claim 3, The sound source recommendation device, It includes a metadata management unit for managing the audio content DB, the content metadata DB and a user management unit for managing the user DB, And the user manager divides and manages the situation information of the user who has requested the current sound source recommendation and the situation information of the user who has not requested the current sound source recommendation. Receiving a sound source recommendation request message including location information and event information from a user terminal; Selecting a sound source to be recommended using the location information, the event information, and previously stored information of the user terminal; And transmitting a message including information on the selected sound source to the user terminal. The method of claim 5, wherein the previously stored information of the user terminal, It includes information on the weight of the preferred item received from the user terminal among the preferred items including a singer, genre, title, etc., and history information including the sound source downloaded by the user terminal and the time when the sound source was downloaded. Sound source recommendation method characterized in that. The method of claim 6, wherein the event information, And at least one of an anniversary, an event, and a business trip stored in the user terminal, wherein only the event information related to the date of the event information is received. The method of claim 6, wherein the previously stored information of the user terminal, Sound source recommending method comprising the time information including the season, the time point for requesting the sound source recommendation.

Images