Universal Concept Graph for Social Network Recommendation Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Social networking services face challenges in generating accurate recommendations due to vocabulary mismatches between different content types, such as member profiles and job descriptions, which use different terminologies to refer to the same concepts, leading to failed matches and recommendations.
Innovation Solution
A universal concept graph is generated by combining internal and external concept phrases from social networking services with external datasets like Wikipedia or Freebase, leveraging linkage structures to create a standardized set of concept phrases for improved matching and recommendation accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If different terminologies are used in member profiles and job descriptions, then content diversity and user expression flexibility are improved, but matching accuracy and recommendation quality deteriorate
Solution Approach 1:
The patent introduces a universal concept graph as an intermediary layer between diverse content terminologies and the recommendation system. This graph maps various terminologies from different content types (member profiles, job descriptions, articles) to unified concept nodes, enabling accurate matching while preserving content diversity. The concept graph acts as a mediator that translates between different vocabularies without losing information.
Solution Approach 2:
The universal concept graph serves multiple functions: it standardizes terminology across different content types, enables semantic matching, supports recommendation generation, and maintains relationships between concepts. By creating a unified conceptual framework that works across diverse content types, the system achieves both adaptability to different terminologies and precision in matching.
2Measurement precision
If a universal concept graph is generated by combining internal and external datasets, then matching accuracy and recommendation relevance are improved, but system complexity and data processing requirements increase
Solution Approach 1:
The system segments the complex task of creating a universal concept graph into manageable components: extracting concepts from internal datasets, extracting concepts from external datasets (Wikipedia, Freebase), mapping relationships between concepts, and building the graph structure. This segmentation allows the system to handle complexity in a structured, modular way.
Solution Approach 2:
The patent performs preliminary actions by pre-processing and pre-mapping concepts from external datasets before integrating them with internal data. The universal concept graph is built in advance and maintained as a reusable structure, avoiding the need to perform complex mapping operations in real-time during recommendation generation.
Data Source
AI summary
A machine may be configured to determining key concepts in documents. For example, the machine accesses a universal concept graph that includes a first set of nodes that represent concept phrases derived from internal documents associated with a social networking service (SNS) and external documents that are external to the SNS, and a first set of edges that connect a plurality of nodes of the first set of nodes. The machine accesses a content object associated with the SNS. The machine generates an induced concept graph associated with the content object based on an analysis of the content object and the universal concept graph. The machine identifies one or more key concept phrases in the content object based on applying one or more key concept selection algorithms to the induced concept graph. The machine stores the one or more key concept phrases in a record of a database.