Intelligent search with deep learning clusters

This paper proposes an Intelligent Search Assistant (ISA) that emulates a brain learning structure based on the Random Neural Network with Deep Learning clusters. Our ISA rearranges the Web results retrieved from several Web Search Engines after a Web user issues a search request. Our algorithm measures and evaluates Web result relevance by assigning each Deep Learning cluster with a specific Web Search Engine; in addition, our ISA selects the best performing learning cluster to teach other clusters. We measure and compare the Web result relevance between different Web search engines against our proposed model; we validate our ISA as a biological inspired method that improves relevance and learning speed. On average; the learning clusters outperform other Web search engines and we confirm that cluster performance can be improved by learning from best learning clusters.