Database Querying TRIZ Cases
- Digital Twin Search Using Graph Embeddings for Sparse Data
A trained graph neural network combines query semantics with twin relationships to improve relevance for sparse user data.
- Key-Ordered Hash Tables for Database Query Cache Efficiency
Key-ordered record processing improves hash-table cache hits, reducing cache misses and memory access overhead during database query execution.
- Personalized Query Auto-Completion Through Purchase-History Reranking
Purchase-history training and search-session context rerank query candidates to make e-commerce auto-completion more relevant to user behavior.