M
Michael W. Macy
Researcher at Cornell University
Publications - 144
Citations - 17562
Michael W. Macy is an academic researcher from Cornell University. The author has contributed to research in topics: Cultural diversity & Social media. The author has an hindex of 44, co-authored 138 publications receiving 15440 citations. Previous affiliations of Michael W. Macy include Brookings Institution & Johns Hopkins University.
Papers
More filters
Posted Content
Automated Hate Speech Detection and the Problem of Offensive Language
TL;DR: This article used a crowd-sourced hate speech lexicon to collect tweets containing hate speech keywords and trained a multi-class classifier to distinguish hate speech from other offensive language, finding that racist and homophobic tweets are more likely to be classified as hate speech but that sexist tweets are generally classified as offensive.
Journal ArticleDOI
Network Diversity and Economic Development
TL;DR: It is found that the diversity of individuals’ relationships is strongly correlated with the economic development of communities and the relation between the structure of social networks and access to socioeconomic opportunity is investigated.
Life in the network: The coming age of computational social science: Science
David Lazer,Alex Pentland,A. Adamic,Sinan Aral,Albert-László Barabási,Devon Brewer,Nicholas A. Christakis,Noshir Contractor,James H. Fowler,Myron P. Gutmann,T. Hebara,Gary King,Michael W. Macy,Deb Roy,M. Van Alstyne +14 more
Journal ArticleDOI
In Search of Excellence: Fads, Success Stories, and Adaptive Emulation1
David Strang,Michael W. Macy +1 more
Abstract: The faddishness of the business community is often noted and lamented but not well understood by standard models of innovation and diffusion. We combine arguments about organizational cognition and institutional mimicry to develop a model of adaptive emulation, where firms respond to perceived failure by imitating their most successful peers. Computational experiments show that this process generates empirically plausible cascades of adoption, even if innovations are entirely worthless. Faddish cycles are most robust across alternative treatments of managerial decision making where innovations have modest positive effects on outcomes. These results have broad implications for the faddishness of a business community increasingly marked by media‐driven accounts of success, and for the properties of organizational practices that are hot one day and cold the next.
Journal ArticleDOI
Learning dynamics in social dilemmas.
Michael W. Macy,Andreas Flache +1 more
TL;DR: In this paper, a learning-theoretic alternative to analytical game theory is proposed, based on a random walk from a non-cooperative equilibrium into a self-reinforcing cooperative equilibrium.