Historical Query Ad Matching via User Profile Annotation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current search engine techniques fail to provide advertisements responsive to user queries that were generated previously, as they require advertisements to be available at the time of the query, missing opportunities to display relevant ads that become available later.
Innovation Solution
A system and method that annotates user profiles with historical queries, including time and frequency information, to identify and display advertisements previously associated with those queries, even if they were not available at the time of the original query, by selecting historical queries based on oldest, most frequent, or most recent criteria and generating a result set of new, non-previously displayed ads.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If advertisements are selected based on availability at query time, then ad selection is simple and fast, but relevant advertisements that become available later are missed
Solution Approach 1:
The system performs preliminary actions by storing user queries and their contexts in advance. When advertisements become available later, the system can match them against historical queries to identify relevant ads that were missed at the original query time, thus improving ad relevance without requiring complex real-time processing
Solution Approach 2:
The patent introduces an intermediary component (ad selection system) that acts as a mediator between the query processing system and the advertisement database. This intermediary stores query information and performs subsequent matching when ads become available, resolving the contradiction by adding a intermediate storage and matching layer that improves relevance while managing complexity
2Reliability
If the system stores and processes historical queries to identify later advertisements, then ad relevance improves, but system complexity and processing requirements increase
Solution Approach 1:
The system extracts only the essential elements needed for ad matching from historical queries (such as query terms, user identifiers, and timestamps) and stores them in a condensed format. This extraction approach maintains ad relevance by preserving key matching information while reducing the overall complexity and storage requirements compared to storing complete query contexts
Solution Approach 2:
Instead of storing and processing complete advertisement content historically, the system creates simplified representations or copies of query information that can be efficiently stored and matched. When ads become available, the system uses these copied query representations to quickly identify relevant advertisements, reducing processing complexity while maintaining matching accuracy
Data Source
AI summary
The present invention is directed towards systems and methods for identifying advertisements responsive to one or more historical queries generated by a given user. The method of the present invention comprises annotating a profile for a user with information associated with one or more historical queries generated by the user. The profile associated with the user is retrieved, the profile indicating one or more historical queries previously generated by the user. One or more advertisements are identified as responsive to a selected historical query maintained in the profile of the user. The one or more advertisements previously displayed to the user in response to the selected historical query are identified. The one or more advertisements identified as not previously displayed to the user are thereafter displayed.