Browse Prior Art Database

A Method for Dynamically Recommending a User Cluster for Targeting of an Advertisement

IP.com Disclosure Number: IPCOM000233859D
Publication Date: 2013-Dec-24
Document File: 3 page(s) / 114K

Publishing Venue

The IP.com Prior Art Database

Related People

Pradhan Pattanayak: INVENTOR [+5]

Abstract

A method is disclosed for dynamically partitioning users into one or more user clusters, and recommending the one or more user clusters for targeting an advertisement. A user cluster can be created based on one or more attributes such as behavioral, social, geographic, temporal and demographic attributes. Similarly, advertisements can be segregated into one or more ad clusters. Accordingly, one or more user clusters are recommended for an ad cluster based on an affinity of the user cluster to the ad cluster.

This text was extracted from a Microsoft Word document.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 53% of the total text.

A Method for Dynamically Recommending a User Cluster for Targeting of an Advertisement

Abstract

A method is disclosed for dynamically partitioning users into one or more user clusters, and recommending the one or more user clusters for targeting an advertisement.  A user cluster can be created based on one or more attributes such as behavioral, social, geographic, temporal and demographic attributes.  Similarly, advertisements can be segregated into one or more ad clusters.  Accordingly, one or more user clusters are recommended for an ad cluster based on an affinity of the user cluster to the ad cluster.

Description

Disclosed is a method for dynamically partitioning users into one or more user clusters, and recommending the one or more user clusters for targeting an advertisement.  An exemplary application can be recommending one or more user clusters to an advertiser involved in creating an ad campaign for targeting of the advertisement.

In an embodiment, the one or more user clusters can be created based on one or more attributes.  The one or more attributes can include one or more of behavioral, social, geographic, temporal and demographic attributes.

The social attributes can include, but is not limited to, friends of a user, followers of a user, a community associated with the user, conversations involving the user and, a status update provided by the user.  The social attribute can be numerically described to capture interactions of the user with other users.

The behavioral attributes include, but are not restricted to, an interaction of the user with the advertisements and property pages.  The interaction can be characterized in terms of number of clicks, number of page visits and likes on pages on a social networking site.

Similarly, the demographic attributes can include personal information such as, gender, age, and personal income.  The demographic attributes can be used to collectively describe and differentiate users from one another.

The geographic features can include, but are not restricted to, location, city, state and country.  The geographic features can be used independently.  In an instance, the geographic attributes can be combined with the social attributes, to derive attributes such as, a relative spatial proximity of the user to the friends of the user.

The temporal features can include, but are not limited to, a time of page visit and a frequency of page visits at different categorical times of the year.  The temporal features can be used to build a user profile, and cluster users with similar profiles.

In another embodiment, an algorithm can be used for identifying the one or more user clusters of simil...