Browse Prior Art Database

Recommendation system based on crowd's profiles in a virtual world

IP.com Disclosure Number: IPCOM000170247D
Original Publication Date: 2008-May-14
Included in the Prior Art Database: 2008-May-14
Document File: 2 page(s) / 30K

Publishing Venue

IBM

Abstract

Disclosed is a mechanism for, especially in a virtual world, displaying a recommendation (for example, to recommend user groups, stores, places, creators, and brands to be referred), which is based on a measurement of other avatar’s preferences as a kind of collaborative filtering. Those preferences are extracted from the participating group information, favorite place information (called “Picks” etc), advertisement (called “Classified” etc), and the wearing clothes and equipments that are set for each avatar residing near my avatar in the virtual world. After the extraction, the mechanism measures a degree of coincidence of those preferences among other avatars and my avatar. When the result of measurement indicates that those avatars have the preference similar to myself, and when they have some information that myself have not referred to yet, but many of those avatars refer to, the mechanism extracts the information and recommends for myself as a recommended list of group, place, creators, etc. In other words, among a certain avatar group in which many avatars have similar attribute values, the mechanism can find out the information and inform for myself, which many other avatars have already known, but my avatar does not have or does not know. The group may consist of avatars that are at the moment and at the same place, or may consist of avatars that are passing by and moving near my avatar within a certain time frame.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 1 of 2

Recommendation system based on crowd 's profiles in a virtual world
Disclosed is a mechanism for, especially in a virtual world, displaying a recommendation (for example, to recommend user groups, stores, places, creators, and brands to be referred), which is based on a measurement of other avatar's preferences as a kind of collaborative filtering. Those preferences are extracted from the participating group information, favorite place information (called "Picks" etc), advertisement (called "Classified" etc), and the wearing clothes and equipments that are set for each avatar residing near my avatar in the virtual world. After the extraction, the mechanism measures a degree of coincidence of those preferences among other avatars and my avatar. When the result of measurement indicates that those avatars have the preference similar to myself, and when they have some information that myself have not referred to yet, but many of those avatars refer to, the mechanism extracts the information and recommends for myself as a recommended list of group, place, creators, etc.

In other words, among a certain avatar group in which many avatars have similar attribute values, the mechanism can find out the information and inform for myself, which many other avatars have already known, but my avatar does not have or does not know. The group may consist of avatars that are at the moment and at the same place, or may consist of avatars that are passing by and moving near my avatar within a certain time frame.

The detail of this mechanism is shown below;

Under a situation in which one or more avatars are near my avatar, the mechanism initiates the following processing with a trigger fired by manual operation for my avatar.

(1) Detect other avatars near my avatar. (For example, inside of a sphere or a circle with a radius of 10 meter, which is centered at the position of my avatar.)

(2) Acquire the following information set for each avatar;
- Group information he/she joined.- Favorite place information (for example, two or more stores, halls, and places registered as "Picks".)- Advertisement (for example, two or more stores and places registered as "Classified".)
- Clothing information he/she is wearing (fo...