Data science: a growing field
Appeared in the 2000s, data science has become key issue of the future due to the emergence of a continuous stream of data. Find out what data science is, and the benefits it could bring to businesses on a day-to-day basis.
The emergence of data science
A specific context: Big data and the explosion of data
Data has been around for a long time. But before, it was present in small quantities and quite limited in some areas...
Today, we can only acknowledge that data has exploded: we are faced with a growing mass of data. This is the famous "Big Data". Big Data is when three elements coexist, the "3Vs": volume, variety, velocity. From there, we began to face an explosion of data but also a need to find a way to "rationalize" this data: this is where data science appeared.
If the science of data has emerged, it is also because we now have the means to process this data. Clearly, the emergence of data science has also been fostered by the development of techniques. Algorithms were slow and laborious... but from now on, they have become much more accessible. Machines’ improvement has become palpable: computers have a better processing capacity, their computing power is multiplied... We are way behind the super-cumbersome computers of the 1980s!
Today, a much higher power than that of older computers fits in your pocket on a tiny smartphone... From now on, data is everywhere and easily accessible to all!
And it starts with your smartphone: don't you already have a lot of apps that exploit your data? Sports app to calculate calories lost, number of steps taken, your heart rate or your music streaming app that recommends titles based on your musical tastes...
All these technologies are already using raw data and processing it quickly and simply at the level of all users. In other words, data and its analysis have really spread.
In summary, three factors have therefore contributed to the emergence of data science:
- Data processing opportunities have become widespread;
- Computer processing capacity has improved;
- The data has become more and more numerous.
A definition of data science
A solution had to be found that would allow this data to be fully exploited and valued for businesses. This is where the full value of data science really came to life.
In France as in America, the discipline is said to have been forged in the 1990s before being theorized and spread in the 2000s. But it was especially in the 2010s that the first data scientists began to appear when the data began to become massive and usable.
In the midst of a boom, the profession is now in high demand. The main role of the data scientist is to recover exploitable data and then create a statistical model that the company can industrialize.
Data science is at the crossroads of several disciplines: it uses methods derived from mathematics and classical statistics (correlations, etc.). It also builds on the latest technologies such as machine learning and artificial intelligence that allow us to go further and faster in data analysis. Data scientists also have advanced computer skills (programming, R language and Python most of the time). Now, data science can rely on these new computer tools and technologies to exploit data that a human could not exploit alone or that a machine did not have the capacity to process before.
To sum up in one definition what data science is: data science is therefore the science that recovers and processes data to extract information that can be used by companies.
What is data science for? Infinite possibilities...
If the field of data science is expanding, it is because data science can be used in many areas, if not all or almost all!
Above all, data science aims to extract value from the magma of data, which is useless as a raw mass. What is the benefit of data science?
The discovery of information
Data science aims to improve the level of knowledge of companies. By updating elements that were previously difficult to identify, it allows you to discover new systems.
For example, one of the favorite areas of data science is marketing. Thanks to the data already present in your company, you probably already have a customer knowledge and therefore feel that you have a good level of information about your customers. But data science can really deepen this customer knowledge with more relevant models. With a higher level of customer knowledge, you can classify and categorize them according to defined criteria. You can also improve recommendation algorithms to make them generate a personalized experience.
Another example in the banking sector is data science, which identifies rare events such as anomalies or breakdowns. This is useful for detecting fraud.
Decision support and predictive analysis
Data science helps companies make more informed decisions. This is the goal of any data science project: to increase the level of information the company can project itself into the future more easily.
With this new information, the company can draw all the consequences from data and make "predictions". Beware, data science doesn’t offer the solution to all problems! Predicting the evolution of a particular thing is only possible if the data science project is oriented towards a clear and precise goal.
A famous example: the Netflix series, House of Cards. Its production was decided after analysing the behaviors of users on the free streaming platform: what they watch, what they are looking for... Netflix was inspired by their tastes to make the house of cards series that is now very successful.
An example in marketing: predicting sales of a new campaign. Another example is health with disease prediction. The weather is also concerned on ‘basic’ forecast algorithms but also on a larger scale: data science could allow better prevention of pollution peaks or natural disasters.
Automation
Take the example of the autonomous car working thanks to all its sensors, artificial intelligence... and data! The development of automation possibilities should allow for a redistribution of human/machine tasks. Humans will be able to focus on high-value-added tasks that can only be accomplished by them, while the machine performs low-value-added tasks. The advantage is time savings’ that companies could gain on certain time-consuming tasks. Another example of automation is machine and simultaneous translation.
As data continues to increase, the field of data science is gaining ground. Businesses have a strong interest in exploiting the wealth contained in their data. And that is why this emerging discipline will undoubtedly be crucial for years to come.
La Ryax Team.