Expanding the Capabilities of Data Analytics
The field of Analytics, or more sophisticatedly known these days as ‘Data Science’ has intrigued many. Perhaps, because of the highly complex nature of the concepts, techniques and technology involved, people refrain from getting close enough to understand the same from the experts.
In part one of this article, let me try and decode the discipline. In the next, I would dwell upon how the world around us is leveraging the power of Analytics for all organisational strategies and informed decision-making.
Data Science is a field that’s a combination of various disciplines that uses scientific methods, intelligent processes & algorithms, and the relevant technology suite to analyse and extract insights from Structured as well as Unstructured data. It’s about intelligently exploring the data available, analysing the same thoroughly, and extracting meaningful & actionable information or insights out of the same. Quite often it culminates in the development of a Predictive Model.
A Predictive Model generally learns from the behaviour of historical data in a very scientific way, and then tries to predict the same for future events. Consider these three factors important in this regard:
- The more data that we have, more accurate is the prediction.
- The earlier we can predict an event, the better - as we get enough time to act.
- The passing of time ensures the availability of more data.
By putting these 3 statements together, we have a kind of circular reference - something we Data Science professionals try to solve by arriving at an optimal balance between the time of prediction and the data available at that time. The scientific processes mentioned above themselves involve a combination of various tools, techniques & innovations, and hence the outcomes may vary from person to person.
The ideal skill-set that makes a well-rounded Data Science professional includes raw intelligence, analytical mindset, logical & numerical ability, thorough Knowledge of Mathematics/Statistics, Computer Science & Programming, Data Mining techniques, Predictive techniques, mastery in using Analytical & Visualisation tools & Languages, understanding of Data Systems & Architectures, exposure to various avenues of Data Science (Advanced Statistical Analytics, AI, ML, Deep Learning, Big Data/Hadoop etc.), domain knowledge, ability to convert business objectives or problems statements to analytical objectives, problem solving approach, innovative thinking, and the ability to think through till the end before conceptualising and developing the best of Solution(s) with an optimal combination of all of the above. Possessing the above to the best extent for each role within an analytics team leads to a lean but highly efficient in-house unit.
It also requires streamlining of Deployment, Tracking & improvisation to extract maximum value for the Organisation. As we see, it includes a combination of natural abilities as well as attained & developed knowledge. Engineers, Mathematicians, Statisticians etc. generally come closer in terms of fitment as well as interest levels.
The beauty of the discipline lies in the fact that one must manage all the nuances and complexities involved internally, but present the outcome in the simplest of ways to the stakeholders - so that the solutions are embraced quickly for reaping maximum benefits.
Particularly agree with the last paragraph and I have seen you doing exactly that in past very effectively. The requirements for a rounded professional laid out in your write-up seems to be too ideal though in real world (there would be very few in market who could tick all the boxes). Good write-up.