Optimized Data Tree for Webpage Attribute Querying
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The challenge lies in efficiently storing and querying large datasets of webpage attributes, which have numerous combinations of values, leading to difficulties in identifying desirable webpages for advertisers due to the enormous number of potential combinations.
Innovation Solution
The method involves building an optimized data tree by identifying and associating content item attributes with values, optimizing the tree by deleting nodes with fewer associated content items, and storing the optimized tree structure to facilitate efficient querying and storage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all possible attribute combinations are stored in the data structure, then complete query coverage is achieved, but storage requirements become enormous
Solution Approach 1:
The patent segments the attribute space into a hierarchical tree structure where each level represents a specific attribute (e.g., country, city, timezone). This segmentation allows the system to store only the segments that actually contain data rather than creating entries for all possible attribute combinations. The tree structure divides the enormous attribute space into manageable, sparsely-populated segments.
Solution Approach 2:
The patent applies local quality by making each node in the tree structure store information locally about the specific attribute combination it represents. Each node contains a count of impressions for its specific attribute path, allowing the system to store only locally-relevant data rather than globally-redundant information. This enables efficient storage where each part of the structure contains only the quality information necessary for its specific query path.
2Measurement precision
If the data structure includes all attribute nodes, then query accuracy is maintained, but query processing time increases
Solution Approach 1:
The patent performs preliminary action by pre-building the hierarchical tree structure during data ingestion, organizing all attribute relationships in advance. This preliminary organization allows queries to be resolved by simple tree traversal rather than complex searches through all possible attribute combinations. The preliminary structuring maintains query accuracy while dramatically reducing processing time.
Solution Approach 2:
The patent transforms the flat, multi-dimensional attribute space into a hierarchical tree structure with depth as an additional organizing dimension. Each attribute level becomes a dimension in the tree hierarchy, allowing the system to navigate through attributes sequentially from root to leaf rather than searching through all combinations simultaneously. This dimensional reorganization maintains query precision while reducing processing complexity.
3Manufacturing precision
If the system stores detailed impression data for all attributes, then advertising targeting precision is improved, but system complexity increases
Solution Approach 1:
The patent segments the complex attribute data into a hierarchical tree where each node represents a specific attribute segmentation (country, city, timezone, etc.). This segmentation maintains detailed targeting information while organizing it in a structured manner that reduces system complexity. The tree structure naturally handles the segmentation of multi-attribute impressions without requiring complex flat-data management.
Solution Approach 2:
The patent changes the parameter representation from flat attribute-value pairs to a hierarchical path structure. Each impression is represented by a path from root to leaf in the attribute tree, transforming the parameter organization from a complex multi-dimensional array to a simple hierarchical path. This parameter transformation maintains targeting precision while significantly simplifying the system's data management complexity.
Data Source
AI summary
Systems, methods and computer program products identify a plurality of webpage attributes, where each of the plurality of webpage attributes is associated with a value. A date tree is built, the data including a plurality of nodes identifying the values associated with the plurality of webpage attributes, and is optimized by deleting one or more of the plurality of nodes. The number of web pages associated with at least one of the plurality of nodes in the optimized data tree is then determined.