Predicting discoveries or automated hypotheses from the literature

Spangler S, Wilkins AD, Bachman BJ, Nagarajan M, Dayaram T, Haas P, Regenbogen S, Pickering CR, Comer A, Myers JN, Stanoi I (2014) Automated hypothesis generation based on mining scientific literature. InProceedings of the 20th ACM SIGKDD international conference on Knowledge discovery and data mining 2014 Aug 24 (pp. 1877-1886). ACM.  URL: http://scholar.harvard.edu/files/alacoste/files/p1877-spangler.pdf

This early study tackles a basic problem that is challenging progress in every field of human intellectual activity: we have become much better at generation of information than at its integrative analysis. This leads to deep inefficiencies in translating research into progress for humanity. No scientist can keep up with the unrelenting flow of new studies and results, even within specialized fields.

The method is trained by chronologically ordering the literature, and using the past to try to predict the future.

Leave a Reply