Abstract
Background/Aims
Lately the popular press is rife with tantalizing references to coming advances brought by “Big Data” methods and software. Modern living--mobile and social computing particularly--emits enormous plumes of data. This data, we are told, can be analyzed in real-time to spot trends and yield important insights, to the benefit of business and mankind generally. The general idea of incidentally-produced data that can be exploited to produce valuable insights is one an HMORN audience is eminently comfortable with--it describes quite a lot of our research. But where do we fit in with these new trends? Have we all been “Data Scientists” doing “Big Data” for years now, and the rest of the world is just now catching up to us? Or, are these things really different and new? Are there things we should be appropriating from this “new” field to make our own work stronger?
Methods
The proposed talk will describe and define several commonly-cited ideas and methods--to wit: (a) big data; (b) map/reduce; (c) no SQL; and (d) data science.
Results
The talk will locate these in a larger technological context, list synonyms and closely related technologies and describe situations where expanding into less-familiar tools may well bear fruit for research data projects.
Conclusions
While much of the tools and methods of big data are squarely addressed to problems we don’t frequently encounter in HMORN research, it is good to have a conceptual understanding of them so that when we do hit the limits of conventional methods and resources we have “someplace to go” before giving a project up as infeasible.
Keywords: Big data, Data science