Dynamic Faceted Search via Query Vector Space

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional facet generation methods for search systems require consistent document tagging, produce overwhelming numbers of facets, and often generate facets unrelated to the query, lacking dynamic management capabilities.

Innovation Solution

A query-focused faceted structure generation method that ingests a document corpus, generates vector space representations, and produces a dynamic structure of relevant categories and facets using two-vector space representations, allowing for dynamic facet generation based on user queries and improving navigation by filtering relevant facets from search results.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If conventional facet generation methods are used, then facets are generated from document corpus, but the number of facets becomes overwhelming and difficult to navigate

Engineering Contradiction:

Improvenumber of facetsVSAvoidnavigation difficulty

Core Design Contradiction:

Quantity of substanceVSEase of operation

Solution Approach 1:

The patent extracts only the most relevant facets from the document corpus by computing term frequencies and selecting top-N facets based on query relevance. This extraction principle filters out unnecessary facets, reducing the overwhelming number while maintaining navigation ease.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent dynamically adjusts facet parameters based on query-specific term frequencies rather than using fixed corpus-wide frequencies. This parameter change allows the system to adapt the number and relevance of facets to each query, preventing information overload while improving navigability.

Inventive Principle:

Principle #35Parameter changes

2Stability of the object's composition

If static facets are generated from existing hierarchy, then consistent structure is maintained, but the facets are not based on matching documents or query

Engineering Contradiction:

Improvefacet structure consistencyVSAvoidquery relevance

Core Design Contradiction:

Stability of the object's compositionVSAdaptability or versatility

Solution Approach 1:

The patent transforms static facets into dynamic facets by computing term frequencies specific to each query and document set. The facet structure adapts to each query while maintaining hierarchical organization, achieving both consistency and query relevance through dynamic term frequency calculations.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The patent creates a universal facet generation mechanism that works across different queries and document corpora. The same algorithm dynamically generates relevant facets for any query while maintaining structural consistency, making the system adaptable to various search scenarios without requiring query-specific customization.

Inventive Principle:

Principle #6Universality (Multi-functionality)

3Stability of the object's composition

If document tagging with existing hierarchy is required, then consistent faceted structure is achieved, but overhead in content curation and management increases

Engineering Contradiction:

Improvefaceted structure consistencyVSAvoidcontent curation time

Core Design Contradiction:

Stability of the object's compositionVSLoss of time

Solution Approach 1:

The patent enables the system to automatically generate consistent faceted structures without manual document tagging. By computing term frequencies and extracting facets programmatically from the document corpus, the system serves itself, eliminating the time-consuming manual curation process while maintaining structural consistency.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent replaces the manual mechanical process of document tagging with an automated computational system. The algorithm automatically analyzes document terms, computes frequencies, and generates consistent facet hierarchies, substituting human effort with machine processing to reduce curation time while maintaining structure quality.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

4Quantity of substance

If large numbers of facets are generated, then comprehensive coverage is achieved, but it becomes difficult to navigate and overwhelming to users

Engineering Contradiction:

Improvefacet coverageVSAvoiduser navigation

Core Design Contradiction:

Quantity of substanceVSEase of operation

Solution Approach 1:

The patent extracts only the most significant facets by selecting top-N terms based on query-specific term frequencies. This extraction filters out less relevant facets, providing comprehensive coverage of important topics while preventing user overwhelm through selective presentation.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent dynamically changes the number of facets presented based on query relevance metrics. By adjusting the facet count parameter according to term frequency analysis, the system provides comprehensive coverage when needed while limiting the number of facets to manageable levels for user navigation.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS11275796B2Dynamic faceted search on a document corpus

Publication Date: 2022.03.15 INTERNATIONAL BUSINESS MACHINE CORPORATION

AI summary

A query-focused faceted structure generation method, system, and computer program product for generating a query-focused faceted structure from a taxonomy for searching a document collection, including ingesting a document corpus, generating a vector space representation of a query and instances from a taxonomy of the document corpus, and producing a dynamic structure of a relevant facet categories and facet values using a two-vector space representation from the generated vector space representation.