Entity Graph Filtering for Central Topic Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current search systems fail to effectively identify and prioritize central entities representing main topics of resources, leading to irrelevant additional content being presented to users, which detracts from the web browsing experience.
Innovation Solution
The method involves generating a filtered entity graph to isolate central entities by removing non-connected nodes and those without outgoing edges, then using these entities to generate additional content relevant to the resource, incorporating pointwise mutual information and query log data to determine centrality scores and relevance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If additional content is presented to users based on entities from resources, then user engagement is improved, but content relevance deteriorates due to inclusion of peripheral and irrelevant entities
Solution Approach 1:
The patent segments entities into different categories (central entities versus peripheral entities) based on their connectivity in the entity graph. Central entities are those that remain after filtering out nodes without outgoing edges and removing non-candidate entities, while peripheral entities are discarded. This segmentation allows the system to selectively use only central entities for generating additional content, thereby maintaining content relevance while still providing engaging additional content to users.
2Quantity of substance
If all entities from resources are used to generate additional content, then quantity of content is increased, but quality and relevance of content deteriorates
Solution Approach 1:
The patent extracts only the central entities from the set of all entities associated with a resource. This is achieved by constructing an entity graph, filtering it to remove nodes that do not correspond to candidate entities, and then removing nodes without outgoing edges. The remaining central entities are used to generate additional content, ensuring both sufficient quantity and high quality/relevance of the content provided to users.
3Loss of information
If entity graphs include all entities from multiple resources, then completeness of entity information is improved, but complexity of processing increases
Solution Approach 1:
The patent performs preliminary filtering of the entity graph before using it to generate additional content. The filtering process removes nodes that do not correspond to candidate entities and removes nodes without outgoing edges, retaining only central entities. This preliminary action simplifies the entity graph structure, reducing processing complexity while preserving the essential entity information needed for generating relevant additional content.
Data Source
AI summary
Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for identifying central entities. In one aspect, a method includes obtaining candidate entities for a first resource; filtering a first entity graph whose nodes represent different entities found in a plurality of resources to remove nodes that do not correspond to a candidate entity, wherein pairs of nodes in the filtered first entity graph that are connected by an edge correspond to pairs of candidate entities that are associated with the same resource; generating a second entity graph for the first resource from the filtered first entity graph, wherein the second entity graph does not include nodes from the filtered first entity graph that are not connected to other nodes in the filtered first graph; and identifying candidate entities that are represented by nodes in the second entity graph as being central entities for the first resource.