Page 78 - Building Big Data Applications

P. 78

Building big data applications

Human language is the new UI layer, bots are like new applications, and digital
assistants are meta apps. Intelligence is infused into all of your interactions
Satya Nadella

Data storyboard

Information ﬂoats around us nonstop. It is a fascinating layer of details that can be
harnessed, integrated, analyzed, repeated for play purposes, implement metrics, and
move to next steps. This layer of details contains data which is structured, semi-
structured, and unstructured, all of which can be often connected to one user or multiple
users of one group, or even multiple groups. The fascination to study and interpret
different patterns of behavior and hidden insights that can lead to wonderful business
outcomes has always been a dream for any enterprise or individual in the world. We
constantly see many examples of this today with the virtual connected world we live in,
but the same patterns have existed since the stone age of human existence, and we have
read in history how many times our minds have innovated ideas for trade and have
created situations of advantage to one individual or group, over the others. The bottom-
line today is the availability of all tipping points in one unison to make this possible;
however, it requires a methodology, the ability to box and isolate all the raw noise and
interpret the needed signals, the ability to scale up and scale out on demand, the pro-
cessing of all kinds of data in own format and integration into one useable format, and
the requirements of neural networks and machine learning to implement backend
processing and automation of tasks. Let us take a deeper look into all these areas with
examples.
Information Layersda multi layered series of data points that are interconnected
loosely and can be harnessed to create a lineage and join the different touchpoints of
granularity efﬁciently and produce the outcomes. However, the issue here is the fact that
we do not harness all the raw data today and hence there is a generation of data con-
sumers who do not know the intricacies of this approach and the lower level methods of
integration. An example of the layers is discussed below.
As seen in Fig. 3.1, there are interactions that I call as “market interactions” and these
are all events that occur, in parallel at multiple locations and can be tethered to deliver
outcomes which are delivered as analytics. The foundation of our lives has always been
driven by the usage of analytics based on the time of its delivery. Whether it is history or
space exploration, we are constantly guided by the events and outcomes. There are

73 74 75 76 77 78 79 80 81 82 83