While traditionally data synthesis often refers to descriptive or interpretative narrative and tabulation in studies like meta-analyses, in the big data context, data synthesis refer to the process of creating synthetic data. In the big data context, the digital technology provides unprecedented tremendous data information. The rich data across various fields can jointly offer extensive information about individual persons or organization for finance, economics, health, other research, evaluation, policy making, etc. However, fortunately our laws necessarily protect our privacy and data confidentiality; this necessary data protection becomes increasing important in our big data world where thefts and various levels of data breach could become much easier. The synthetic data has the same or highly similar attributes of the real data for many analytic purposes but masks the original data for more privacy and confidentiality. Synthetic data was first proposed by...
- Abowd, J.M., & Lane, J.I. (2004). New Approaches to Confidentiality Protection: Synthetic Data, Remote Access and Research Data Centers. In Domingo-Ferrer, J. & Torra, V. (Eds.), Privacy in Statistical Databases: CASC Project International Workshop, PSD 2004, Barcelona, Spain, June 9-11, 2004, Proceedings. (pp. 282-289). Berlin: Springer.Google Scholar
- Abowd, J. M., & Woodcock, S. D. (2004). Multiply-imputing confidential characteristics and file links in longitudinal linked data. In Domingo-Ferrer, J. & Torra, V. (Eds.), Privacy in Statistical Databases: CASC Project International Workshop, PSD 2004, Barcelona, Spain, June 9-11, 2004, Proceedings. (pp. 290–297). Berlin: Springer. Google Scholar
- Rubin, D. B. (1993). Discussion: Statistical disclosure limitation. Journal of Official Statistics, 9, 462–468.Google Scholar
- Winkler, W. E. (2007). Examples of easy-to-implement, widely used methods of masking for which analytic properties are not justified. Tech. Rep., U.S. Census Bureau Research Report Series, No. 2007–21.Google Scholar