Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

Method and system for Determining Fine Grained User Interest from Search Advertisements and Display Advertisements

IP.com Disclosure Number: IPCOM000201719D
Publication Date: 2010-Nov-19
Document File: 3 page(s) / 40K

Publishing Venue

The IP.com Prior Art Database

Related People

Hastagiri Prakash: INVENTOR [+2]

Abstract

A method and system for determining fine grained user interest from search advertisements and display advertisements is disclosed. The method involves tracking interactions of a user with one or more search advertisements and one or more display advertisements. Further, the interactions of the user with the one or more search advertisements and the one or more display advertisements are used for determining the fine grained user interest. Based on the fine grained user interest, search retargeting is performed.

This text was extracted from a Microsoft Word document.
This is the abbreviated version, containing approximately 37% of the total text.

Method and system for Determining Fine Grained User Interest from Search Advertisements and Display Advertisements

Abstract

A method and system for determining fine grained user interest from search advertisements and display advertisements is disclosed.  The method involves tracking interactions of a user with one or more search advertisements and one or more display advertisements.  Further, the interactions of the user with the one or more search advertisements and the one or more display advertisements are used for determining the fine grained user interest.  Based on the fine grained user interest, search retargeting is performed.

Description

Disclosed is a method and system for determining fine grained user interest from search advertisements and display advertisements.  The search advertisements correspond to advertisements generated on a search results page based on user search queries.  The search advertisements are displayed on the search results page along with search results of the user search queries.  The display advertisements correspond to advertisements displayed on any webpage including the search results page.  The method involves tracking interactions of the user with one or more search advertisements and one or more display advertisements.  The interactions of the user may include, but are not limited to, clicking or hovering over the one or more search advertisements and the one or more display advertisements on the search result page.  Further, the interactions of the user with the one or more search advertisements and the one or more display advertisements are used for determining the fine grained user interest.  Based on the fine grained user interest, search retargeting is performed.

In order to determine the fine grained interest of the user, the method and system disclosed determines product/service information such as, but not limited to, a product model, a product category, a brand, pricing, and a company from one or more search advertisements and one or more display advertisements interacted by the user.   The product/service information is determined by analyzing text portions of the one or more search advertisements interacted by the user.  Additionally, the product/service information is determined by analyzing content of landing web pages corresponding to the one or more display advertisements interacted by the user.

In an embodiment, the fine grained user interest is determined using a machine learning algorithm.  The machine learning algorithm is initially trained and tested using data from the one or more search advertisements interacted by the user.  Based on the training and testing, the machine learning algorithm is fine tuned and applied to the one or more display advertisements interacted by the user.

In order to perform training/testing of the machine learning algorithm, product/service information is determined corresponding to training data.  The product/service information is dete...