Eurekometrics Analyzing the Nature of Discovery 英文参考文献.docVIP

Perspective Eurekometrics:AnalyzingtheNatureofDiscovery SamuelArbesman1,2*,NicholasA.Christakis1,2,3,4 1Department of Health Care Policy, Harvard Medical School, Boston, Massachusetts, United States of America, 2Institute for Quantitative Social Science, Harvard University,Cambridge,Massachusetts, United States ofAmerica,3DepartmentofSociology, Harvard University, Cambridge, Massachusetts, United States ofAmerica, 4DepartmentofMedicine,HarvardMedicalSchool,Boston,Massachusetts,UnitedStatesofAmerica Untilrecently,thequantitativestudyof sciencehasfocusedonstudyingpatternsin publications [1,2], such as citation counts to discern impact, and in coauthorship networks to discern collaboration. How- ever,twomajortrendsareconvergingthat offer the field of scientometrics a novel opportunity to understand scientific dis- coveryandalsotoinfluencehowscienceis done. The first is the advent of vast computational resources and storage ca- pacityavailabletoscientists[3,4],andthe second is automated science [5,6]. These innovations offer the potential for a new type of scientometrics: quantitatively ex- amining scientific discoveries themselves. This study of discoveries, rather than simplyofscientificpublications,offersthe opportunity to understand science at a deeperlevel.Wetermthisdiscovery-based approachtoscientometricsaseurekometrics. Eurekometrics aims to supplement the traditionalbibliometricapproachofscien- tometrics by examining the properties of whetherornot,onaverage,thecodingof those codons with less entropy can be foundusingmoretypesofexperiments[7]. In other words, a simple eurekometric approach could examine whether or not thosecodonswithlessinformationcanbe moreeasilyunderstood. related to its popularity in the literature [13]. Withtheincreaseofautomateddiscov- ery and large-scale data collection, eur- ekometric research has the potential to explode. First, automated science will necessarily have the property of creating large amounts of discovery data. Illu