What is data science?

June 23, 2020

For a few years now, the term Data science has been present in many media, but what exactly is behind it?

The first time we talked about data science was in 1992, during a conference on statistics in France. At that time, data analysis was summarized as statistics. We must remember that this field was reserved for specialists. The tools to carry out this type of study were expensive, complex to implement and reserved for specialists...statisticians.

The main source for statistical studies is data. Since then, the amount of data surrounding us has exploded. Internet, smartphones, social networks, digitization, digitization of public functions of States, scanning of old documents, connected objects... All these elements are not necessarily only qualitative or quantitative values, we can speak here of images, videos, sounds, emotions, feelings... All this information that surrounds us can be stored to better understand us and anticipate our possible reactions. All these vectors allow us to recover data that can be processed and analyzed.

Data storage is one of the keys to the success of data science. In recent years, new ways of storing data have appeared, so that they take up less space and are available more quickly and easily. We are currently hearing about NoSQL databases, Graph databases, JSON storage... The architecture of data storage is evolving to make it more available.

A new way of storing data has appeared for some time, storage on clouds, such as AWS (Amazon web services), Microsoft Azure, Google cloud ... This way of storing data makes them more easily accessible and avoids the problems of network, information transfer and security. These tools do not only allow to store data, but also to process and analyze these data.
With Moore's law, we know that, since the beginning of the 70's, the computing power doubles every 2 years on average. The computer tools we have are therefore increasingly powerful, which allows us to store even more information and to analyze it ever more quickly. At the same time, the power of the Internet networks has been improved thanks to the multiplication of fiber optic technologies that allow us to exchange information more and more rapidly throughout the world.

All the elements are gathered, the data is brought back and stored, more complex algorithms are more easily used such as random forest or artificial neural networks thanks to the increase in computing power. We have moved from data mining to machine learning. Data is the raw material of data science. Once this data is available and properly organized, it is possible to aggregate it, make it visual and analyze it to draw conclusions. The role of these algorithms is to help us make decisions based on facts (transformed into data) and on the past. We are talking about artificial intelligence here.

The biggest difference between data mining and machine learning is the fact that the algorithms are self-learning, i.e. once a first mathematical model has been created on the initial data, it will evolve and improve to become more efficient as new data arrives. This was not the case with data mining models.

Whether in the world of services, industry or finance, data is everywhere, it is increasingly stored in clouds and the tools to process it are more and more accessible. Data science is no longer reserved for specialized sectors, it can now be used everywhere!

  • UL6S
    40 rue des Arts
    94170 Le Perreux-sur-Marne

  • +33 (0)6 07 23 00 12

  • contact@ul6s.com

Veuillez renseigner votre adresse email pour recevoir les détails de notre événement.

En renseignant mes coordonnées, j'autorise l'UL6S à me communiquer occasionnellement des informations complémentaires.

Presque terminé ! Remplissez le formulaire pour recevoir les informations supplémentaires

Nous vous remercions de votre intérêt à l'événement du 27 janvier 2022. Nous vous communiquerons plus de détails dès que possible.

Veuillez renseigner votre adresse email pour recevoir le référentiel de compétences

En renseignant mes coordonnées, j'autorise l'UL6S à me communiquer occasionnellement des informations complémentaires.

Presque terminé ! Remplissez le formulaire pour accéder au téléchargement

No Spam

08 décembre 2020 de 11h00 à 12h30 :
« quelles compétences pour mener à bien les projets d’Excellences Opérationnelles de demain ? »

« quelles compétences pour mener à bien les projets d’Excellences Opérationnelles de demain ? »

Votre demande d'inscription a bien été prise en compte.

 

Veuillez confirmer votre inscription en cliquant sur le lien dans l'email que nous venons de vous envoyé. 

 

Le lien de connexion vous sera communiqué à partir du vendredi 4 décembre 2020.

« quelles compétences pour mener à bien les projets d’Excellences Opérationnelles de demain ? »

Scroll to Top