At school , college or the university , Mathematics as a subject has been taught to sharpen the intellect. Solving the trickiest problems and modelling complex assumptions.All this using a paper and pen. But when it comes to the real world , graduates of Mathematics are shunned and confined to the teaching profession. Multiple reasons being
- lack of training on tools and software
- unfamiliarity of frequently used engineering methods
- ignorance of real world usage and applications
- others not knowing how to utilize their skills
To solve all the above issues a strategy was thought upon to make mathematics more practical.
Datascience : Mathematics in practice
Whether its Ramanujan's Number theory , Euclidean geometry or Pythagoras theorem - all deal with data and formulas to manipulate the data. When it comes to the modern age of computers it becomes possible to implement the formulas as algorithms to manipulate the data. Hence there's no reason maths graduates should be confined to pen and paper. Rather explore their skills and make themselves more practical utilizing the big machines and solve the complex of problems without much stress or strain. What remains is to gain some familiarity of where mathematics is being used and how to improve upon that !
Usage of Mathematics
Whether its Physics , Chemistry or Biology - Psychology , Sociology or Criminology - Astronomy , Meteorology or Oceonography - Genetics , Diagnostics or Forensics and all the other subjects taught at the school or university. Mathematics has been an integral part of the subjects knowingly or unknowingly at various levels of complexity. Overlooking the multidisciplinary aspect of the subjects was the sin. Bridging this gap and building a multidisciplinary platform for mathematics becomes the fundamental reason for DataScience. Statistics , probability, optimization and logic playing an important role.
Datascience for scientific discovery
Although experimentation is an important aspect of scientific discovery, the data analysis performed on experimental data forms the crux of research papers. The modelling ,analysis ,reporting, decisions and conclusions from the data decides the quality of the research. And to perform such unquestionable research there needs to be continuous innovation of the models and methods. Mixing and matching multiple data sources can enhance the output and more processing power provides the speed to put research on the fast track. Thus amazingly datascience plays a pivotal role in modern research.
Datascience in automation
Robotics , Cyber Physical systems or Electromechanical have been in existence over 50 years. The fundamental drawback in the systems being their limited memory and processing power limiting their functionality. With applications hard coded into the limited memory over an RTOS. Reimagining their functionality as data-driven systems responding to stimuli makes their development far more streamlined. With ability to embed more complex application logic their performance becomes extremely flexible doing away with human intervention. That's real automation.
Event driven systems , Machine Learning and Artificial Intelligence too have been in existence with limited success. Partly due to limited resource availability , affordability and minimal practical use. But when we reimagine their functionality as data-driven their use becomes extremely flexible due to the algorithm run software defined approach.
A lot has changed with a change in architecture. Applying the data driven approach wonderful new technologies have developed. Be it the drones in the sky or automated cars on the roads. Precision robots or optimized factories. Mathematical Biology or Analytical Chemistry. All utilizing their domain specific data and unprecedented computing power , otherwise impossible earlier.
No comments:
Post a Comment