it is not a priority, it is first collected to serve as, or illustrate, a reasoning or an idea. The major challenge of the last two decades has refocused on another field: that of understanding and analysis. Offering a clear report facilitates the management and steering of an activity.
Data processing opens the door to strategic recommendation. Now that the value of data is known, its value can further improve, provided it is operational and ready to use.
The individual, the consumer and the voter are regularly solicited, sometimes against their will, often in ignorance of the risks involved. In the face of the abuses that have been noted, the law has had to be involved.
Before this stage of maturity, “just in case” storage is reassuring without really knowing why. This intuition leads companies to sometimes accumulate incomplete or unorganized data. Ignorance of its purpose has resulted in the accumulation of data in huge volumes, with the only limit being the technical constraints of the moment, the cost per byte.
So far, it has been impossible to use all this data due to lack of time and resources. Therefore, it is not possible to educate the overall population. Using a sample is imperative, with representativeness being a key concept.
Technological progress has made it possible to overcome this constraint of representativeness. The entire dataset is now vulnerable to attack even though the size is discouraging. These masses of data, veritable “assembled pieces”, have whetted the appetite of the greedy known as Big Data, machine learning, artificial intelligence and other “datavores”, reinventing a 2.0 statistic: data science.
At the same time, the proliferation of equipment (tools, applications) has democratized the use of data, making connection and leading to automatic and endemic branches. The variety of media (computers, tablets, smartphones), the explosion of networks and access providers facilitate the exchange, sharing and processing of information. Mobile consumption has become instantaneous.
This continuous feeding creates a new addiction, a demand of a different nature. In order to respond before it becomes obsolete, it is necessary to move quickly, even if it means returning the information to its original role as simple data. In this sense, pre-selections are made, and the data is delivered, sorted and “ready to read”, not necessarily ready to use. The preferences and tastes of the “target” are identified; its supposedly known expectations are identified in relation to its history. Trespassing?
However, the absence of filters or segmentation leads to fears of saturation. The data is ubiquitous despite its limited lifespan. It is refreshed and constantly updated. The required reactivity sometimes weighs on its quality. The corollary concession is the “average” information, which is difficult to exploit. It is against this risk of drowning or submersion that analysts, true lifeguards, fight. Their responsibilities are cleaning, reducing and synthesizing data; detailing or refocusing it if necessary; and using it in the delegated question. At the heart of the process, this human presence guarantees the surpassing of a simple mechanical treatment and therefore the creation of value. The mission of the latter consists, after all the stages of data processing, of restoring the true dimension of the data while making it profitable.
Here, Professor Monino's book brings the main tools that are indispensable to his mission, which, in the course of time, will become more and more indispensable. Unfortunately, not all companies have the necessary resources (time, budget, analyst) to extract or exploit this wealth. However, the use of specialists such as E2S-Conseils allows us to make the most of this information to help companies develop and flourish.
I hope that you enjoy reading and learning from this book.
Philippe RIBIERE
Chief Executive Officer
E2S-Conseils
Acknowledgements
I wish to dedicate this book to my sister who unfortunately was unable to read my work on data.
Also, all the members of the “Réseau de Recherche sur l'Innovation”, RRI, and in particular, Dimitri Uzunidis, who encouraged me to publish this work and I thank him warmly for his encouragement.
This book is the culmination of many years of research dedicated to data processing, to statistics and to econometrics in the TRIS laboratory (Traitement et Recherche de l'Information et de la Statistique). It is the fruit of various work carried out within the framework of R&D (Research & Development) for several start-ups in the Languedoc-Roussillon of France and large private and public groups.
Thank you to all those who have supported me during difficult times and have helped transform an individual intellectual adventure into a collective one. In particular I offer my thanks to Philippe RIBIERE, President of E2S-Conseils with whom I have written several R&D books over the years.
Thank you also to the team of researchers at the “Montpellier Recherche en Economie” laboratory who demonstrated their trust in my research and granted me my emeritus status.
I would also like to thank my children: Christine, Laurent, Caroline and Daniel, as well as their partners: Laurent, Alexandra, Guillaume and Louise who have all supported and surrounded me during the difficult times I had to face.
To my grandchildren: Lily-Rose, Jean-Baptiste, Alice and Gabriel.
To Anne-Marie, my wife.
Jean-Louis MONINO
Introduction
The world has become digital, and technological advances have increased the number of different ways for accessing, processing, and distributing data. New technologies are now reaching a certain maturity.
Today, data comes from all sides: geolocation sensors, smartphones, social networks where we share files, videos, photos, etc., Internet shopping transactions by customers, banking transactions through credit cards and so on. In France, out of 65 million people, 83% are Internet users; 42% are registered on Facebook, or 28 million members. More than 72 million phones are activated, and the French spend on average more than 4 hours a day surfing the Internet. French mobile users spend 58 minutes or more there; 68% of the population is registered on social networks. French people spend more than 1 hour and 30 minutes a day on social networks. The development of these masses of data and their access represents what is called “Big Data”. These immaterial data arrive continuously; their processing poses problems, particularly in knowledge extraction. Thus, new methods of automatic information extraction are implemented: for example, “data mining” or “text mining”. They underlie profound changes that affect the economy, marketing, research and even politics. The amount of data will increase sharply with the arrival on the market of connected objects that will gradually come into use. Elements of our daily life are already connected: the car, the television and some household appliances. They are or will be equipped with a chip to collect and transmit data to their users via a computer, tablet or smartphone. The most important thing is that these items will also be able to trade with each other! This will allow us to remotely manipulate the equipment in our home, in our vehicles by connecting to them from our home and outside our home, using smartphones or any other equipment. This is called “the Internet of Things”.
This phenomenon is now of interest to operational decision-makers (marketing managers, financial managers, etc.) when it comes to analyzing
