Music recommendation engine

Patent metadata

Field	Value
Publication number	US-10055411-B2
Application number	US-201514928084-A
Country	US
Kind code	B2
Filing date	Oct 30, 2015
Priority date	Oct 30, 2015
Publication date	Aug 21, 2018
Grant date	Aug 21, 2018

Abstract

A method and system implemented for generating a music recommendation intended to surprise and delight a user using data and computer analytics. The method and system collects and categorizes musical preferences using past preferences, user demographics, social media data and psychological variables which include the Big-5 personality traits of the user to generate a profile containing preferred musical parameters. The method and system categorize music based on different dimensions and compare music with a user generated profile containing preferred music parameters. The method and system may search for additional music across two vectors, parameters that closely resemble the preferred musical parameters, and outlying parameters that extend outside the preferred parameters, in order to locate and present musical recommendations that have similarities to the preferred music of the user, but variable enough to surprise and delight the user.

First claim

The invention claimed is:

1. A method for dynamically generating a music recommendation to a user in real-time, said method comprising:
granting, by a processor, permission to dynamically collect user information in real-time from one or more data sources;
dynamically analyzing, by the processor in real-time, the user information to identify psychological variables of the user;
storing, by the processor, the user information and the psychological variables of the user in a data store;
dynamically generating, by the processor in real-time, a user profile from the user information, said user profile comprising preferred music parameters and the psychological variables of the user;
dynamically selecting, by the processor in real-time, a plurality of music recommendation candidates from a repository, said plurality of music recommendation candidates each having a plurality of parameters within a threshold range of the preferred music parameters, wherein at least one parameter is outside the threshold range;
dynamically analyzing, by the processor in real-time, the preferred music parameters and the psychological variables;
dynamically computing, by the processor in real-time, a similarity value and a surprise value for each of the music recommendation candidates;
dynamically further computing, by the processor in real-time, a recommendation value for each music recommendation candidate that balances the similarity value and the surprise value, wherein the recommendation value is calculated by the processor using the general formula ((similarity value) x *(surprise value) y ) or (x*similarity value)+(y*surprise value), wherein x and v are integers selected between 1 and 100; and
dynamically updating, by the processor in real-time, a presentation to the user of the music recommendation candidate having the highest recommendation value.

2. The method of claim 1 , wherein the user information being collected dynamically in real-time includes data or metadata describing the user's music purchases, streaming habits, downloading habits, viewing habits, logs of viewed websites, published comments, ratings or reviews.

3. The method of claim 2 , wherein the user information being collected dynamically in real-time further comprises user demographic information selected from the group consisting of marital status, family members, education, employment, income, perceived socio-economic status and a combination thereof.

4. The method of claim 1 , wherein the plurality of parameters within the threshold range is at least 80% of the preferred music parameters.

5. The method of claim 1 , wherein the preferred music parameters include preferences of music selected from the group consisting of genre of music, tempo, time signature, recording date, musical key, pitch distribution, volume, instrumental accompaniment, sentimental analysis of lyrics and a combination thereof.

6. The method of claim 1 , wherein the threshold range is ±1-y % of each value of the preferred music parameters, where y is selected from the group consisting of 5, 10, 15, 25, 33, 50, and 75.

7. The method of claim 1 , further comprising providing at least one support service for at least one of creating, integrating, hosting, maintaining, and deploying computer-readable program code in a computer system, where the computer-readable program code in combination with the computer system is configured to implement the granting, the dynamic analyzing, the storing, the dynamic generating, the dynamic selecting, the dynamic computing, the dynamic further computing, the presenting, and the dynamic updating.

8. A computer program product, comprising one or more computer readable hardware storage devices having computer readable program code stored therein, said computer readable program code containing instructions executable by one or more processors of a computer system to implement a method for dynamically generating a music recommendation to a user in real-time, said method comprising:
granting, by the one or more processors, permission to dynamically collect user information in real-time from one or more data sources;
dynamically analyzing, by the one or more processors in real-time, the user information to identify psychological variables of the user;
storing, by the one or more processors, the user information and the psychological variables of the user in a data store;
dynamically generating, by the one or more processors in real-time, a user profile from the user information, comprising preferred music parameters and the psychological variables of the user;
dynamically selecting, by the one or more processors in real-time, a plurality of music recommendation candidates from a repository, said plurality of music recommendation candidates each having a plurality of parameters within a threshold range of the preferred music parameters, wherein at least one parameter is outside the threshold range;
dynamically analyzing, by the one or more processors in real-time, the preferred music parameters and the psychological variables;
dynamically computing, by the one or more processors in real-time, a similarity value and a surprise value for each of the music recommendation candidates;
dynamically further computing, by the one or more processors in real-time, a recommendation value for each music recommendation candidate that balances the similarity value and the surprise value, wherein the recommendation value is calculated by the processor using the general formula ((similarity value) x *(surprise value) y ) or (x*similarity value)+(y*surprise value), wherein x and y are integers selected between 1 and 100; and
dynamically updating, by the processor in real-time, a presentation to the user of the music recommendation candidate having the highest recommendation value.

9. The computer program product of claim 8 , wherein the user information being collected dynamically in real-time includes data or metadata describing the user's music purchases, streaming habits, downloading habits, viewing habits, logs of viewed websites, published comments, ratings or reviews.

10. The computer program product of claim 9 , wherein the plurality of parameters within the threshold range is at least 80% of the preferred music parameters.

11. The computer program product of claim 8 , wherein the preferred music parameters include preferences of music selected from the group consisting of genre of music, tempo, time signature, recording date, musical key, pitch distribution, volume, instrumental accompaniment, sentimental analysis of lyrics and a combination thereof.

12. The computer program product of claim 8 , wherein the threshold range is ±1-y % of each value of the preferred music parameters, wherein y is selected from the group consisting of 5, 10, 15, 25, 33, 50 and 75.

13. An interactive jukebox, said interactive jukebox comprising:
a processor, a memory coupled to the processor, and a computer readable storage device coupled to the processor, said computer readable storage device containing program code executable by the processor via the memory to implement a method for dynamically generating a music recommendation for a user in real-time, said method comprising:
granting, by the processor, permission to dynamically collect user information in real-time from one or more data sources;
dynamically analyzing, by the processor in real-time, the user information to identify psychological variables of the user;
storing, by the processor, the user information and the psychological variables of the user in a data store;
dynamically generating, by the processor in real-time, a user profile from the user information, comprisi

Assignees

• IBM

Inventors

• Cama Karl J
• Freed Andrew R
• Herman Norbert
• Ray Shubhadip

Classifications

• G06F16/686

  using information manually generated, e.g. tags, keywords, comments, title or artist information, time, location or usage information, user ratings · CPC title

• G06F17/30035

  Physics · mapped topic

• G06F17/3005 Primary

  Physics · mapped topic

• G06F17/3053

  Physics · mapped topic

• G06F16/637 Primary

  Administration of user profiles, e.g. generation, initialization, adaptation or distribution · CPC title