Search & Discovery

Library search in ikigize combines fast text filtering with Learning Graph intelligence to surface relevant resources across contexts.

Three-layer retrieval model

Discovery runs as a layered system:

1

Instant Text Filter

Client-side search across title, description, tags, and author — results appear as you type (150ms debounce).

2

Graph-Augmented Search

Your query is sent to the Learning Graph. Topics and Skills matching the query are found in Neo4j, then all resources that DEVELOP those topics are returned (400ms debounce).

3

Global Graph Recommendations

In Global scope, the entity's mapped Topics drive multi-path traversal recommendations — surfacing publicly available resources from across the platform that match this context.

Search and Discovery
Find resources quickly with powerful search and filtering capabilities that work across all your libraries

Search & Filtering

Search across titles, descriptions, and content
Semantic search understanding context
Advanced filters by type, difficulty, and tags

Global Library Benefits

Access to growing global resource collection
Community-curated high-quality content
Cross-organisation knowledge sharing

Intelligent Discovery

Personalised recommendations based on learning patterns
Context-aware suggestions for related resources
Proactive resource discovery notifications

Graph filter panel

Each library context includes a topic-based graph filter. Instead of typing keywords only, users can navigate the Topic taxonomy and filter to resources that explicitly develop selected concepts.

This is especially useful in large libraries, where topic navigation gives more precise results than title matching alone.

Global recommendations

When using global scope, entity topic mappings guide recommendations. The system traverses related topics and ranks public resources from across the platform by relevance.

Next steps