Relational Word List Generation for Search Coverage

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current information retrieval systems lack effective methods to enrich and improve the coverage of search results by identifying synonymous and near-synonymous relationships between search terms based on historical click logs, leading to suboptimal user search experiences.

Innovation Solution

A method and apparatus that acquire historical click logs, determine matching search results, establish synonymous relationships, and generate a relational word list using statistical machine translation, which enhances the content and coverage of search results by marking relevant terms in search queries.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If traditional information retrieval methods are used, then the search process is simple and fast, but the coverage and relevance of search results are insufficient

Engineering Contradiction:
Improvecoverage of search resultsVSAvoidcomplexity of information retrieval system
Core Design Contradiction:
Quantity of substanceVSDevice complexity

Solution Approach 1:

The system performs preliminary action by pre-processing historical click logs to extract synonymous relationships between search terms before actual search queries are executed. This pre-extraction of semantic relationships enables enhanced search result coverage without adding complexity during the actual search operation, as the relational word list is ready in advance for direct application.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent introduces an intermediary mechanism by creating a relational word list that mediates between the user's search query and the actual search results. This intermediary structure maps synonymous relationships from historical data onto current search queries, enabling broader result coverage while maintaining a relatively simple search interface and process.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If synonymous relationships are established based on historical click logs, then the relevance of search results is improved, but the processing time and computational resources increase

Engineering Contradiction:
Improverelevance of search resultsVSAvoidprocessing time for analyzing historical data
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The system performs the time-consuming task of analyzing historical click logs and extracting synonymous relationships in advance, before actual search queries are executed. By pre-processing and storing these relationships in a relational word list, the system eliminates the need to re-analyze historical data during each search operation, significantly reducing processing time while maintaining high relevance.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent creates a copied and simplified representation of historical search patterns by extracting only the essential synonymous relationships from the full historical click log data. This copied relational word list captures the semantic structure needed for improving relevance without requiring processing of the complete historical dataset during each search, thus reducing computational time and resources.

Inventive Principle:
Principle #26Copying

3Adaptability or versatility

If a relational word list is generated from historical data, then the content and coverage of search results are enriched, but the system complexity and data processing requirements increase

Engineering Contradiction:
Improveadaptability of search resultsVSAvoidcomplexity of data processing system
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system applies extraction by taking out only the essential synonymous relationships from the complex historical click log data. Instead of processing the entire historical dataset, the system extracts and stores only the meaningful semantic relationships in a relational word list. This extraction approach enriches search adaptability while minimizing the complexity of data processing by focusing only on the critical information needed for synonym matching.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The relational word list serves as an intermediary structure that simplifies the connection between historical data and current search operations. It mediates between the complexity of historical click log analysis and the simplicity of real-time search queries, enabling enhanced adaptability without requiring complex data processing during each search operation. The intermediary list pre-establishes the mapping between synonymous terms, making the system more versatile while keeping processing requirements manageable.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentUS11055373B2Method and apparatus for generating information
Publication Date: 2021.07.06 BEIJING BAIDU NETCOM SCI & TECH CO LTD
  • US11055373B2 patent drawing
  • US11055373B2 patent drawing
  • US11055373B2 patent drawing

AI summary

Embodiments of the disclosure disclose a method and apparatus for generating information. A specific embodiment of the method comprises: acquiring a historical click log, the historical click log comprising a historical search term and a clicked historical search result corresponding to the historical search term; determining whether matching clicked historical search results exist in the historical click log; establishing a synonymous relationship between historical search terms corresponding to the matching clicked historical search results, in response to determining the matching clicked historical search results existing in the historical click log; and generating a relational word list based on the established synonymous relationship. The embodiment helps to enrich the content of the relational word list, and improve the coverage of the relational word list.