Pages

Friday, August 30, 2013

Big Data, Big Benefit, Big Challenge

Welcome to Jingmei’s Blog. I am a new graduate student in the Department of Statistics after six years industrial working as a senior integrated circuit designer. My previous working experience focused on designing chips which were used in cell phones of LG, Samsung and Motorola. During working, I collected, organized, analyzed and interpreted simulation (before manufacture) and test (after manufacture) data of circuit parameters. This kind of work generated my strong interest in analyzing and processing data which statistics mainly deals with.


Now, I am a big fan of big data analysis. I think it might be the best era to study the efficient methods for mining valuable information and making correct inferences from data, since big volume data are now quickly accumulating from many sources like enterprise, transaction and social media.  When people send email, post articles in blog, buy things through internet, large amount of data are generated unknowingly. By analyzing these data, we will get the very interesting, useful and valuable results, which might help companies make better decision, provide more customized service and finally get high revenue. Accordingly, with the rapid growth of requirement on dealing with big data, the benefit from big data market also increases with incredible speed as shown in the following figure.

Estimated growth of Big Data market


Of course, the magnitude and complexity of data also bring big challenges. How to deal with messy data from many different sources? How to find the internal relation between diversified data? How to extract the valuable untapped insights from big data? To resolve these problems, we must combine statistical science with computer science to create more efficient tools in dealing with complex data sets. This is also my study and working target. I want to use statistical methodology in constructing models which could improve computing efficiency and help to create more powerful analytic software. 

2 comments:

  1. Manipulating large data is a major challenge as you said. I am working on a project in bioinformatics which requires handling heavy data and I see it is really difficult to process such huge data. It is certainly interesting to know more about big data. Nice write up! All the best.

    ReplyDelete
  2. Wow... I'm impressed with your expertise in data analysis. I wonder what methods and models of data mining have you used specially for economic analysis because as far as I know, the markets (especially stock and bond) are volatile, high volumes of data change every second

    ReplyDelete