Statistical machine learning techniques like Supervised Learning can be accurately implemented using Hadoop to accurately predict and forecast a consumer behavior. For example, Nate Silver (The New York Times) used such techniques to predict election results very accurately. Oracle Big Data has already shown ways to convert the user clicks (on an ecommerce site) in to valuable data. Similar techniques on Hadoop can be applied to any customer data that can result in precious insights towards customer intelligence.
Understanding and Optimizing Business Processes
With Hadoop many new ventures have got tremendous advantage in optimizing their business operations. Evernote uses Hadoop to improve the conversion and empower their customers to remember everything .Human Resource departmental activities like talent acquisition require churning of large amounts of data but this can be very easily implemented using Hadoop. With Hadoop and Big Data, the supply chains can be optimized for minimum environmental footprint. The other problems of Operations management like dynamic delivery route optimizations, which require GPS and other dynamic data of large number of delivery vehicles, can be done at very cheap costs to the company as Hadoop runs on commodity hardware and is relatively easy to implement the framework. In the field of financial trading, Hadoop & Big Data algorithms are already being used to perform High Frequency Trading which could require heavy computations to be done in less than a millisecond time.
Facilitates Healthcare Research
DNA analysis is one of the hardest computations to do as the data generated during such analysis is tremendously large. Hadoop plays a key role in analysing and decoding these DNA Strings in a
span of few minutes and facilitates the discovery of newer or better cures. Such analysis could also result in predicting disease patterns more accurately .Hadoop and Big Data analysis of sensor data are used to accurately monitor the babies in intensive care units at Neonatal Intensive Care .Statistical analysis of such data can be useful to predict diseases and save valuable lives.
Hadoop and Sports Analytics
Today's sports generate different kinds of data including video footages, social media discussions, blogs, analyses from specialists etc. The real time data generated from the sensors or video cameras set up on a cricket field or a football ground can be analyzed in split seconds and predictive algorithms can be run with bull's eye accuracies. Sports brand managers can utilize Hadoop to track a player not only on the field but also off the field and asses the player’s performance effect on team's reputation through a Hadoop based social media data analysis. Sensors fixed in the clothes and apparel of the sportsmen will generate real time data which can be analyzed using algorithms to know the speed or agility of a player in real time. Such efforts were already implemented with considerable success in Major League Soccer and in Formula One . Here, Hadoop can play a significant role as it is scalable in real time and the data is secure by default.
I have dealt with a few major areas where Hadoop has already had a strong impact. Of course there are a lot more applications of Hadoop today and many more will emerge along with their respective processing needs. I am confident that Hadoop, as a framework, is rugged enough to handle many such needs of business in this era of Big Data.