Cascading Learning System for Enterprise Search Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional search engines are inaccurate in searching enterprise data due to their inability to consider the semantic meaning of keywords, leading to incorrect results when searching for business objects and documents.
Innovation Solution
A cascading learning system that utilizes a meta-model semantic network to analyze and classify search terms, integrating contextual information and business terminology to provide relevant results by determining relationships between semantic objects and managing terminology in a contextual network.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If conventional search engines search for keywords in enterprise data, then search coverage is improved, but search accuracy deteriorates
Solution Approach 1:
The patent introduces semantic network models and domain-specific terminologies as intermediaries between user queries and enterprise data. These intermediaries enable the search system to understand the semantic meaning of keywords in specific business contexts, thereby improving search accuracy while maintaining comprehensive coverage
Solution Approach 2:
The system dynamically adjusts search parameters based on the type of enterprise data being searched. By changing search strategies according to data domains (e.g., using different terminologies for HR data vs. financial data), the system achieves both broad coverage and high precision across diverse enterprise data types
2Speed
If conventional search engines return results based on keyword matching, then search speed is improved, but result relevance deteriorates
Solution Approach 1:
The system pre-builds semantic network models and terminologies for different enterprise data domains before actual search operations. This preliminary preparation enables fast, relevant search results by avoiding complex semantic analysis during runtime while ensuring high result relevance
Solution Approach 2:
The patent replaces traditional mechanical keyword-matching mechanisms with semantic understanding based on pre-built models. This substitution maintains search speed while dramatically improving result relevance by enabling the system to understand the meaning and context of search terms
3Measurement precision
If a meta-model semantic network is integrated to analyze search terms, then search accuracy is improved, but system complexity increases
Solution Approach 1:
The semantic network system is segmented into distinct, domain-specific modules (e.g., HR terminology, financial terminology, manufacturing terminology). Each module handles a specific type of enterprise data independently, which reduces overall system complexity by allowing modular development, deployment, and maintenance while maintaining high search accuracy across diverse domains
Data Source
AI summary
A ranking in cascading learning system is described. The cascading learning system has a request analyzer, a request dispatcher and classifier, a search module, a terminology manager, and a cluster manager. The request analyzer receives a request for search terms from a client application and determines term context in the request to normalize request data from the term context. The normalized request data are classified and dispatched to a corresponding domain-specific module with a request dispatcher ranking calibrator. Each domain-specific module of a search module generates a prediction with a trained probability of an expected output using a corresponding domain-specific ranking calibrator. The terminology manager receives normalized request data from the request dispatcher and classifier, and manages terminology stored in a contextual network. The cluster manager comprises a central ranking calibrator, a training and sot container, and a module generator configured to generate a pluggable module.