# US 2018/0322195 A1 · Method for recommending musing in playlist and apparatus using the same

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2018/0322195 A1
- Document ID: US20180322195A1
- Status: Published application
- Kind: A1
- Application number: 15/608,604
- Filed: 2017-05-30
- Published: 2018-11-08
- Earliest claimed: 2017-05-04
- Assignees: Buzzmuisq Inc.
- Applicants: Buzzmuisq Inc.
- Inventors: Jungsuk LEE; Keunwoo CHOI; Tae Hoon KIM; Gukhyeon GYEONG; Uk HER; Sungho LEE
- Web page: https://patentgrep.com/patent/US20180322195A1

## CPC

- G06F 16/637: {Administration of user profiles, e.g. generation, initialization, adaptation or distribution}
- G06F 16/48: Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F 16/639: {using playlists}
- G06F 16/686: {using information manually generated, e.g. tags, keywords, comments, title or artist information, time, location or usage information, user ratings}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Disclosed herein is a method for recommending music in a playlist. The method includes: receiving at least one selection track; determining at least one recommended track for the received selection track by using an engine, the engine configured to predict a track to be played next among potential tracks based on an order of tracks in a playlist; and providing the at least one determined recommended track.

## Claims

### Claim 1

A method for recommending music in a playlist, the method comprising:
receiving at least one selection track;
determining at least one recommended track for the received selection track by using an engine, the engine configured to predict a track to be played next among potential tracks based on an order of tracks in a playlist; and
providing the at least one determined recommended track.

### Claim 2 (depends on 1)

The method of claim 1, further comprising: before the determining the at least one recommended track,
receiving tracks in at least one playlist collected from a device,
wherein the determining the at least one recommended tracks comprises: determining at least one recommended track for the received selection track by using the engine, based further on an order in which the tracks are played or created in the at least one playlist.

### Claim 3 (depends on 2)

The method of claim 2, further comprising: mapping the tracks in the received, at least one playlist to the potential tracks.

### Claim 4 (depends on 3)

The method of claim 3, wherein if the tracks in the at least one playlist exist in the potential tracks, the tracks in the at least one playlist are mapped to the potential tracks, and wherein
the mapping comprises converting the tracks into embedded tracks and mapping between the embedded tracks.

### Claim 5 (depends on 3)

The method of claim 3, further comprising:
converting the tracks in the at least one playlist into vectors if the tracks in the at least one playlist do no exist in the potential tracks; and
mapping the tracks in the at least one playlist to some of the potential tracks that have vectors closest to the converted vectors.

### Claim 6 (depends on 5)

The method of claim 5, wherein the tracks to be converted are tracks stored in a cache or tracks received from a streaming server upon request.

### Claim 7 (depends on 1)

The method of claim 1, further comprising: before the determining the at least one recommended track,
receiving non-preferred tracks determined based on behavior pattern of a user,
wherein the determining the at least one recommended tracks comprises: determining at least one recommended track for the received selection track by using the engine, based further on the non-preferred tracks.

### Claim 8 (depends on 7)

The method of claim 7, wherein the behavior pattern of the user is one of removing a track from a playlist, skipping a track to a next track before a predetermined period of time elapses, and rating a track below a predetermined rating.

### Claim 9 (depends on 1)

The method of claim 1, further comprising:
converting the received selection track to an embedded audio,
wherein the determining the at least one recommended track comprises: calculating a distance between the converted embedded audio and each of embedded audios of the potential tracks, and
wherein at least one recommended track for the received selection track is determined by using the engine, based further on the calculated distance.

### Claim 10 (depends on 1)

The method of claim 1, wherein the determining the at least one recommended track comprises: calculating a distance between the selection track and each of the potential tracks in term of at least one of artist, genre and release year, and
wherein at least one recommended track for the received selection track is determined by using the engine, based further on the calculated distance.

### Claim 11 (depends on 1)

The method of claim 1, wherein in the engine configured to predict a track to be played next among the potential tracks based on the order of the track in the playlist, each of the potential tracks is represented by one vector, and the value of the vector is set to be closer if tracks appear consecutively in the playlist while the value of the vector is set to be more distant if tracks do not appear consecutively in the playlist.

### Claim 12 (depends on 1)

The method of claim 1, wherein the engine configured to predict a track to be played next among the potential tracks based on the order of the track in the playlist is learnt based on a plurality of playlists received from a music streaming server.

### Claim 13 (depends on 1)

The method of claim 1, further comprising: calculating a distance between the selection, track and each of
the potential tracks in term of at least one of artist, genre and release year; and
determining at least one additional recommended track with respect to the recommended track based on the calculated distance, wherein the providing the recommended track comprises providing the recommended track, along with the additional recommended track.

### Claim 14

A device for recommending music in a playlist, comprising:
a communications unit configured to receive at least one selection track;
a processor operably connected to the communications unit and configured to:
acquire the at least one selection track from the communications unit, and
determine at least one recommended track for the received selection track using an engine, the engine configured to predict a track to be played next among potential tracks based on an order of tracks in the playlist;
a display unit configured to display the at least one recommended track; and
a touch input unit configured to select the at least one recommended track displayed on the display unit.

### Claim 15 (depends on 14)

The device of claim 14, wherein the communications unit is further configured to receive tracks in at least one playlist collected from a device before the processor determines the at least one recommended track, and
wherein the processor determines the at least one recommended tracks by determining at least one recommended track for the received selection track by using the engine, based further on an order in which the tracks are played or created in the at least one playlist.

### Claim 16 (depends on 15)

The device of claim 15, wherein the processor is further configured to: map the tracks in the received, at least one playlist to the potential tracks.

### Claim 17 (depends on 14)

The device of claim 14, wherein the processor receives non-preferred tracks determined based on behavior pattern of a user before it determines the at least one recommended track, and
wherein the processor determines the at least one recommended tracks by determining at least one recommended track for the received selection track by using the engine, based further on the non-preferred tracks.

### Claim 18 (depends on 14)

The device of claim 14, wherein the determining the at least one recommended track is carried out by: calculating a distance between the selection track and the potential tracks in term of at least one of artist, genre and release year, and determining at least one recommended track for the received selection track by using the engine based further on the calculated distance.
