Synergy Data Science and Synergy Analytics Logo

HOW IS DATA SCIENCE DEFINED?

     There is no shortage of definitions typically based on business needs, industry practice, "science paradigm", or legal and regulatory requirements. Data science typically includes: data source tracking, data extraction, data taming, data blending, data validation, data mining, visualization, modeling, coding, sampling, forecast accuracy measurement, back-testing, statistical operations, machine learning, artificial intelligence (AI), analytics, and BI (Business Intelligence).


     In addition, data science includes such specific topical areas as: deep learning, adaptive learning, marketing modeling, psychometrics, forecasting, simulation, econometrics, finance analytics, and many others.


You can see things get rather complicated in this fast-growing field. Often, each side does not listen to the other sometimes with disastrous consequences for the optimal solution and the truth. For this reason, Synergy Data Science focuses on masterfully combining "new school" data science with "old school" data science by using our unique Synergistic Method™ .

NEW SCHOOL DATA SCIENCE

Data science is a relatively new term that did not gain widespread traction until 2015 in the Fortune 500 business community outside Silicon Valley. Many “new school” data science models are from paradigms (scientific world views) such as engineering or computer science. Those two paradigms often focus on problems concerning inanimate objects such as computer networks or engineering components.  The focus of many techniques therefore was on deterministic or non-stochastic problems solved using mathematical or logic-based algorithms. (Business translation: modeling of things and not people, often no need for statistical hypotheses or the scientific method, no need to ask why, goal of accurate prediction not causality.) Although developed based on things like robots and machines, Artificial Intelligence (AI) and deep learning can often produce biological, genetic, and environmental predictions of human behavior (or that of other living organisms) more accurately than techniques from traditional scientific fields.

OLD SCHOOL DATA SCIENCE

Numerous “old school” paradigms are based more on human behavior. These areas bring business judgment, the scientific method, statistics, econometric models, finance, psychology, education, and marketing modeling to applied data science. The focus of many old school techniques therefore is on probabilistic problems also called stochastic processes instead of pure mathematics and logic. (Business translation: modeling of people and not things, ability to test hypotheses and perform what if analyses, a need to ask why, with likely goals of accurate prediction, getting financial impacts of causative factors, and explanation of what is most important.) Although developed based on people and financial transactions, regression models and simple statistical techniques can often produce easy-to-understand models of machine and IoT (Internet of Things) behavior that are simpler to run, interpret, and create than complicated black box algorithms.

SYNERGISTIC DATA SCIENCE

Our unique advantage at Synergy Data Science is the “Synergistic Method™ we pioneered to leverage old and new school approaches targeted to a client’s needs. We obtain stronger, more robust and accurate solutions with our diverse mixture of “new school” and “old school” data scientists and approaches. Note that often using BOTH new and old school techniques together is the key to optimal results for clients.





EXTERNAL LINKS

(DEFINITION OF DATA SCIENCE OR OTHER TOPICS AS NOTED)

Berkeley Science Review

Cal Poly

National Center for Educational Statistics

National Network of Libraries of Medicine

Data Scientist is the "Sexiest Job"

Pro's and Con's of Data Science from Data Flair



SYNERGY DATA SCIENCE LINKS

Return to Founder's Page Index

Return to Main SYNERGY DATA SCIENCE Index

Please follow this link for more about data science.



Deterministic   Stochastic