Context-Aware Query Suggestion Using Concept Data Structures

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current search engines face challenges in providing accurate query suggestions due to ambiguity in search queries and varying responses from different engines, leading to inefficiencies in information retrieval.

Innovation Solution

A context-aware query suggestion process that analyzes a sequence of previous queries to generate query suggestions by mining historical search data, creating a concept data structure, and matching the current query with relevant suggestion candidates based on user intent.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If query suggestions are generated using traditional methods (finding similar queries in search logs), then query suggestions can be provided to users, but the suggestions may be slow and fail to resolve ambiguities in search patterns

Engineering Contradiction:

Improveaccuracy of query suggestionsVSAvoidspeed of generating query suggestions

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system pre-processes search logs offline to build a concept data structure that organizes queries into conceptual groups and identifies co-occurring query patterns. This preliminary action stores processed relationships in advance, enabling fast online query suggestion generation without performing complex log analysis in real-time, thus resolving the speed-accuracy tradeoff

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent introduces a concept data structure as an intermediary between raw search logs and query suggestions. This intermediate representation organizes queries into conceptual groups and captures contextual relationships, allowing the system to provide accurate suggestions quickly by matching against pre-computed concepts rather than analyzing raw logs during online operations

Inventive Principle:

Principle #24Intermediary (Mediator)

2Reliability

If context analysis is performed by analyzing sequences of previous queries, then user intent can be better understood, but the complexity of the system increases

Engineering Contradiction:

Improverelevance of query suggestionsVSAvoidcomplexity of context analysis system

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system segments the complex task of context analysis into distinct components: (1) mining search logs to identify co-occurring query pairs, (2) organizing queries into conceptual groups, (3) building a concept data structure, and (4) matching current queries against the structure. This segmentation makes the overall system more manageable and implementable while maintaining high relevance

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

Context-related processing is performed in advance during offline log mining, where co-occurring query patterns and conceptual relationships are pre-computed and stored in the concept data structure. This eliminates the need for complex real-time sequence analysis during online operations, reducing system complexity while preserving reliability

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS9330165B2Context-aware query suggestion by mining log data

Publication Date: 2016.05.03 MICROSOFT TECHNOLOGY LICENSING LLC

AI summary

Techniques described herein describe a context-aware query suggestion process. Context of a current query may be calculated by analyzing a sequence of previous queries. Historical search data may be mined to generate groups of query suggestion candidates. Using the context of the current query, the current query may be matched with the groups of query suggestion candidates to find a matching query suggestion candidate, which may be provided to the user.