End-to-end vertical web search pseudo relevance feedback queries recommendation software

Users' web information needs are increasingly exploratory, seeking to navigate unfamiliar domains and discover knowledge. However, existing search engines struggle with ambiguous queries, leading to irrelevant results. To address this, we propose an architecture that autonomously extracts domain knowledge from initial queries. This system transforms queries into a semantic model, suggesting relevant queries to aid exploration. Evaluated against Google, our architecture achieved 89 % accuracy in automated query recommendation, with 85.4 % system usability and an "A" ranking. This approach enhances exploratory search by facilitating clearer, more effective query formulation, thus improving information retrieval for users.