MetadataShow full item record
PublisherThe University of Arizona.
RightsCopyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction, presentation (such as public display or performance) of protected items is prohibited except with permission of the author.
EmbargoDissertation not available (per author's request)
AbstractBig data-driven talent analytics is an emerging field, which has attracted great interest from both industry and academia. Compared to the traditional survey-based approaches which are slow and costly, big data-driven talent analytics can provide advanced techniques to help get a better understanding of talent management and build better business decision systems. However, big data-driven talent analytics usually suffers from poor data quality since these data are generated without unified standards. To overcome this problem and exploit the value of large-scale career, education, and job data, this dissertation aims to design advanced and sophisticated deep learning approaches to learn data characteristics and address challenging problems in talent analytics. This dissertation presents three essays to study problems in big data-driven talent analytics. The first essay develops a novel deep learning model to learn job and employee embeddings. The second essay proposes a new deep survival analysis model to predict employee turnover behaviors. Finally, the third essay designs a holistic artifact with multiple functions to measure and bridge the skill gap between college students and jobs. This study aims to address burning problems in talent analytics, which not only have practical implications in the industry, but also could contribute to computational design science in information systems research.
Degree ProgramGraduate College
Management Information Systems