My understanding of “big data” is massive amounts of data, variety of
data types, and data integrity. With the advent of more and better data
collection methods, it becomes possible to obtain comprehensive data
from multiple dimensions. Data can also be collected and analyzed in
real time. Different AI algorithms are suitable for different types of
data. They extract different data features and may produce different
biases. Combining them through a strategy often allows the different
biases to cancel each other out, yielding better results than a single
algorithm.
My job is to analyze data using statistical methods. One project I
worked on was analyzing hospitalization data for older adults for
Medicare reference. I think artificial intelligence combined with big
data can make disease diagnosis more accurate, reduce misdiagnosis
and missed diagnosis, lower medical costs, and improve medical
quality. The data collected first should be comprehensive and
complete. It includes not only medical records and medical
examination images, but also living environment, eating habits,
working environment, etc. Then a variety of AI algorithms are applied
to analyze and process various types of data such as texts, numbers,
and images. Finally, the results are integrated to more accurately
assess current health status, estimate possible health risks and make
recommendations to reduce health risks.
Personal health information is very sensitive and it is difficult to collect
comprehensive data. I hope there is an opportunity to solve other
problems with a similar approach. The storage capacity and processing
power of a single machine are limited. I think AI and big data will be
more closely integrated. The application space of the combination of
artificial intelligence and big data is almost unlimited. I will try to apply
more the big data technology in my work.
