Learnable Contextual Network for Enterprise Search Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional search engines are inaccurate in searching enterprise data as they fail to consider the semantic meaning of keywords, leading to incorrect results due to the lack of contextual understanding of business objects and relations.
Innovation Solution
A learnable contextual network is developed that generates semantic objects and relations from a meta-model semantic network, integrating a neural network with these objects and relations to perform statistical analysis and update the network, thereby improving the accuracy of search results by identifying stronger semantic relations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional search engines use keyword matching to search enterprise data, then the search process is simple and fast, but the search accuracy is poor due to lack of semantic understanding
Solution Approach 1:
The patent introduces a semantic network as an intermediary layer between the search engine and enterprise data. This semantic network contains semantic objects representing business entities (employees, customers, products) and their relationships. The search process queries this semantic network to understand the meaning and context of search terms, thereby improving search accuracy without directly complicating the core search engine architecture.
Solution Approach 2:
The patent segments the search system into distinct components: the conventional search engine, the semantic network with semantic objects and relations, and the integration layer. This segmentation allows the complex semantic understanding functionality to be added as a separate module, maintaining the simplicity of the original search engine while enhancing overall search accuracy through the semantic network.
2Measurement precision
If the system integrates neural network with semantic objects and relations, then the contextual understanding improves, but the computational resources and processing time increase
Solution Approach 1:
The patent pre-builds the semantic network with semantic objects and their relationships before the search process. This preliminary action organizes enterprise data into a structured semantic framework in advance, so that during search operations, the system only needs to query this pre-organized network rather than performing complex real-time semantic analysis, thereby reducing processing time while maintaining high contextual understanding accuracy.
3Measurement precision
If the semantic network is updated with stronger semantic relations identified through statistical analysis, then the search relevance improves, but the system maintenance complexity increases
Solution Approach 1:
The patent implements a feedback mechanism where the system performs statistical analysis on connections between semantic objects to identify stronger semantic relations. These identified strong relations are then used to update the neural network and refine the semantic network structure. This continuous feedback loop automatically improves search relevance based on actual usage patterns and relationships, reducing the need for manual system maintenance and configuration.
Data Source
AI summary
A method and apparatus for detection of relationships between objects in a meta-model semantic network is described. Semantic objects and semantic relations of a meta-model of business objects are generated from a meta-model semantic network. The semantic relations are based on connections between the semantic objects. A neural network is formed based on usage of the semantic objects and the semantic relations. The neural network is integrated with the semantic objects and the semantic relations to generate a contextual network. A statistical analysis of the connections between the semantic objects in the contextual network is performed to identify stronger semantic relations. The identified stronger semantic relations are used to update the neural network. The updated neural network is integrated into the contextual network.