SBIR Phase II: Serendipitous Search System Using Lateral Analogy to Match Potential Solutions to Unmet Needs:Feasibility Study Based on Screening Approved Drugs for Repurposing

点击翻译按钮获取中文摘要

The broader impact/commercial potential of this project is to accelerate the paceof research and development to enable more rapid deployment of technologiesinto commercial / industrial contexts. In many fields, information is expanding atsuch an exponential rate that finding relevant results to technical knowledgesearches is increasingly difficult. Further, content is expanding so fast that mostfields are rapidly forming sub-disciplines, leading to the ?silo-ing? of differentknowledge sub-domains, a clear challenge to both academia and industry. Weneed ever better ways to organize and present information to users. There aredisadvantages of the current search engines, mostly relating to excessivesimilarity in search results. Further, while these engines present informationrelating to a known search target, they are less effective at presentingunexpected results for information that a user has never heard of but that wouldbe useful. What is therefore needed is an exploration system giving searchers astrong serendipitous element with a maximum likelihood of results from diverse,unexpected, and potentially provocative sources. This will break down silos byproviding a rapid, relevant means for knowledge-transfer between differentdisciplines, fostering interdisciplinary innovation. This system has been designedto provide a means for systematic, automated discovery.This Small Business Innovation Research (SBIR) Phase 2 project is focused onoptimizing and scaling a serendipitous document search system for repurposingtechnologies by analogy into lateral fields. Both by sub-parsing discrete contentinto ontologically separable entities, such as capability, characteristic, andcomposition, and by comparatively assessing certain of these attributes betweensuch entities, the attribute relatedness of these entities can be used to drive theirself-assembly into related attribute networks. This approach provides asignificant value proposition for drug repurposing, which is the current focus ofthis project. To scale the pair-wise comparison and network assembly of millionsof documents, a map-reduce based text-processing framework will be developedso that massively parallel computations can be carried out in a time- and costefficientmanner. A distributed search engine technology will be deployed toenable rapid querying of the emerging document relationship network. A series ofmachine learning algorithms will then be used to determine potentially hiddenstructural architectural features within the document relationship network.Machine learning will elucidate the nature of the relationships in drug networksthrough analyses of inter-node relationships and sub-graph motifs (termed?innovation motifs?). Documents including U.S. patents and scientific papers willbe processed in the system.