Data Analytics 1 PDF

Do you hear me? So guys, please confirm: do you see my screen? Yes, we can. Okay. Great. So let\'s get started. First of all, let me introduce myself. Hello everyone. Again, my name is Carmelo. I\'ve been a faculty member at RBS since 2021. I teach data science courses for both executive and inte...

Do you hear me? So guys, please confirm: do you see my screen? Yes, we can. Okay. Great. So let\'s get started. First of all, let me introduce myself. Hello everyone. Again, my name is Carmelo. I\'ve been a faculty member at RBS since 2021. I teach data science courses for both executive and international master\'s programs. My focus is generally on data mining, machine and deep learning, and distributed architecture. Beyond academia, I\'m also a principal solution architect and the technical manager at Olidata, which is an Italian IT company. I try to bring real-world industry experience to the classroom. I\'m passionate about bridging the gap between theory and practice and ensuring my students are well-prepared for the challenges and opportunities of the data-driven world. In my spare time, I enjoy spending time with my two daughters. I look forward to getting to know you all. That\'s all. Welcome to the course. Welcome to the course. Okay, so before we start, I want to share with you the slide deck for today\'s lecture. Let me upload it. I want to share it with you using WeTransfer, okay? Let me share with you. Okay, here I\'ve pasted the URL. Please confirm you are able to download the slide deck. Okay, great. Okay, so if I\'m not wrong, this is a blended online course where we have students from Finance, Supply Chain, and Pharma master\'s programs, is that correct? Okay, great. Great. Okay guys, so let\'s get started. So, do you have some experience in the data science world, in data generally speaking, or something like that? Oh, okay. Yes, Rosa, great. Okay, great, great. Okay. So our lecture today and during the second part is generally covering theory but also some practice. Yeah, but my aim is to give you an overview of the data science world and the basics behind the data-driven concept, and why it is important for a company to be a data-driven company. And during part two, we will look further into AI, and I will introduce to you some main concepts. We\'ll start to understand all the problems that are related to the AI world and all the challenges behind it. So, with that, we can start with the lecture. This is the agenda for today. We will start with an introduction to data science and a deep dive into data-driven companies, data culture, strategy, and data governance. During the first part, we will talk about data science organization and methodologies, some concepts about privacy, security, and ethics. So, during part two, we will dive into privacy, security, and ethics. We will talk about generational bias, copyright, and many other topics. As I said before, we\'ll cover the main concepts about advanced analytics organization and effective communication. At the end of this lecture, I\'ll give you a simple homework, okay? A really simple homework that we will discuss at the beginning of part two of this course. Then we will see the real business case related to the homework that I\'ll give you later, okay? We will do an introduction to data science. So the concept is, the question is, what is data science? It\'s\... it\'s\... Tell me. Okay. Sorry, I heard a message from the chat. Okay, you\'re all clear. Okay, so what is data science? Here we have a Venn diagram. In the middle, we have data science. As you can see, data science is the intersection between three main sets: in this case, we have computer science (information technology), math and statistics, and domain business. So, talking with you, the domain business is related to, for example, pharma, banking, and so forth. Firstly, looking at this Venn diagram and analyzing the different intersections, we have the first intersection between information technology and math and statistics, as you can see here. This intersection represents machine learning, which is a critical subfield of artificial intelligence. What is the main difference of machine learning related to canonical traditional programming? Unlike traditional programming, which follows explicit instructions, machine learning relies on algorithms that can learn and adapt from data patterns. This means that rather than telling a program exactly what to do, we train it with data. So the model---machine learning or deep learning---discovers solutions in an independent way. Machine learning powers predictive analytics, recommendation engines, and even complex systems like autonomous driving, for example. So through it, we are able to uncover---and in the next lecture, we will see better---this concept of patterns in data and make predictions that would be impossible to define by coding rules. Okay? So this is, as I say, the main concept behind machine learning, and in this case, the intersection between computer science (or information technology) and mathematics and statistics. The second intersection is at the crossroads of computer science and domain business knowledge. Here, we find software development. Developers work to understand specific industry needs---whether in finance, healthcare, e-commerce, banking, pharma, and so forth---to understand and craft solutions through software. In this space, the focus is on creating robust applications that address specific problems within these industries, blending coding expertise with an understanding of business requirements. So it\'s work that a computer science professional follows together with a domain business or a subject matter expert. For example, a financial application might require algorithms for calculating investments, or an application could integrate tools to analyze patient data to improve dialysis treatments. So, in this way, we see the difference between software development and machine learning. Technically speaking, in both cases, we always have software. The difference is that, as I said before, in machine learning, we have an independent way where algorithms can perform and follow business objectives. Software development is always predefined, okay? So this is the main difference. Finally, we have the third intersection that involves mathematics and statistics and domain business knowledge. Here we have, for example, in the finance industry, data modeling is used to forecast economic trends and assess investment risks. Or in medicine, AI and machine learning assist in discovering potential drug interactions, predicting patient outcomes, and so forth. So in this case, we have three different intersections beside three different concepts, but the concept in the middle---the intersection between all these sets---is data science, because the Venn diagram is based on mathematical sets. Okay? So we have data science. So now the question is, what is data science? I would say we are discovering information using business knowledge, approaching the problem using computer science technology, but using machine learning---systems able to perform tasks in an independent way---in a natural study of data to extract meaningful insights for business. This is the definition of data science, not from a technical standpoint, okay? So in this case, the definition of data science is the study of data to extract meaningful insights for business. For sure, the main concept, the core concept, is data. So, studying data---this is the main concept. And starting from the data concept, we have a new quote. Let me say, it\'s also a commonplace: \"Data is the new oil.\" What does it mean? Why is data the new oil? As you can see here in this picture, we have some platforms to illustrate, but the brands are Google, Meta---so the picture is a little bit old---we have Tesla, Microsoft, Uber, and so forth. Why is data the new oil? Because I think that there are several similarities between data and oil. It\'s a valuable resource. Just as oil is a valuable resource that fuels industries, data is a valuable resource that fuels decision-making and innovation across every sector. What do you think? Another similarity is something about limited availability. Well, not exactly scarce, but valuable data often requires effort and resources to collect, clean, understand, and analyze. So data is not always ready to use; it\'s not something off the shelf like petrol or other resources. In order to use it, like oil, data can be used to drive economic growth. And if you think about it---think about Twitter, all the social media---the economy is based on data, on sharing data, sharing information. And while we have similarities, we also have some limitations of this analogy. For example, first of all, accessibility. Because oil can be concentrated in specific geographical locations like the Middle East or in North America---so just specific geographical locations---while data can be more widely distributed. Okay? Another limitation of the analogy is, for example, sustainability. Because unlike a limited resource like oil, data can be replicated, can be shared. For that, we can consider data a potentially renewable resource. What do you think? The overall meaning---let me say, just to close about this concept---is that this phrase emphasizes three main points: data is a powerful asset in today\'s world; data has the potential to drive innovation, growth, and competitive advantage; and just like oil, data needs to be managed and analyzed effectively to unlock its full potential. What do you think? So, in your opinion, what is your thinking about this phrase, about this concept? So, \"Data is the new oil.\" Do you agree with me? Do you have something to add? Let me switch to the chat and collaborate just to see. What do you think? Yes. Agree, great. So do you have something to add? It\'s about the concept. Eric says, \"Very interesting; actually, governments like El Salvador are investing in data mining for their currencies. Yeah, it\'s for Bitcoin, but in the end, it\'s deep.\" Yes, for sure. Yes, yes, for sure. I think that all governments are investing in data mining and, generally speaking, in digital transformation to become data-driven organizations. The same applies to governments, because---and we will see later on---there are a lot of advantages to becoming data-driven. Let me see the other chat. \"To some level, I agree---not so much on sustainability. Data needs to be stored and manipulated in data centers, and it consumes a lot of resources.\" Yes. But I think that it\'s not just\... So the problem is that today\'s economy is based on oil. A lot of data\... Because in order to produce energy, we need oil, okay? So, I think\... So, Jane says, \"Any decision-making process needs data?\" Yes, for sure. And we will see later on that our power between governments and companies is something like oil. Yeah, you are right. Completely agree---centralized and decentralized big dinosaurs. \"Yes, AI will only achieve this?\" Yes, I agree with you. Yes, I think that it\'s something that we need in this capitalist world. Yeah, it\'s a relatively romantic quote. Interesting. I think data becomes very interesting when you start comparing trends in diverse datasets with each other. Yes, agree. With Wolf on the point of it not being very accessible. Yes. But let me say, a lot of big companies like Microsoft and Google are working on building sustainable data centers under the sea in the north of Europe, reducing the need for cooling energy and so forth. So there are a lot of policies and, I\'d say, a lot of regulations about this. So, Andrea Gallo says, \"I think that while other energy sources can be used, this is even more challenging in the modern world.\" Yes, Andrea, I agree with you. True, but it\'s only a very small part and there\'s a lot of greenwashing. Yeah. Well, today---so this is the correct vision of today\'s landscape---you are right. But I hope, I think that in the next years, there will be a revolution. Okay. Oh, about the ethical goals, the ability, and we will talk about ethical sustainability in part two, so the next lecture that we will have next Monday. So Andrea says, \"There can be too much useless data, but never too much.\" Okay, I agree with Andrea. I think even more crucial is how our data is used in the future. Okay, I share your view. Thank you all for that. This is the question: how is the data we\'re talking about used? We will talk about this. I can also think about how big retailers, Walmart for instance, are using machine learning to improve their forecasts, resulting in more self-buying. Totally. So today, the concept of forecasting is, I would say, considered old. Today we talk about predictive and prescriptive analytics, okay? So machine learning is everywhere today. I can say honestly that machine learning and deep learning are everywhere---in all industries, yeah, for sure, for sure. Okay guys, I\'m so happy that you are sharing your opinions and thinking about this point, okay? Now, we are going to introduce the Big Data concept. So everyone talks about Big Data. You agree, Big Data is everywhere---we can consider it a common topic. What Big Data means in reality is not so immediate. Okay, so there is a formal definition of Big Data. Today, we talk about five Vs to define Big Data, but you can think that five years ago, the definition of Big Data was just related to the first three Vs. Now, I\'m going to show you this history. Obviously, volume. So, the first V is volume, which refers to the huge quantity of data generated worldwide. Think about social media posts, IoT sensors, wearable devices, all the weather data, and so on. And with the right tools, organizations can leverage this vast volume to gain insights. Later on, we will see this concept better. And this is the first V in the Big Data definition. The second one is velocity. Velocity stands for the speed at which data is created, captured, and processed. So, think about how fast data arrives in real-time systems like stock market exchanges or smart cities, where immediate insights are essential. And this is the second V of Big Data definition. The third one is represented by variety. Variety represents the diverse forms the data takes, including structured formats like databases and unstructured formats like emails, videos, social media content, and so forth. So this variety allows us to analyze different dimensions of information but requires flexible processing methods. So, to better define this concept, the variety for Big Data stands for the possibility to store and analyze different formats of data from various sources. In business intelligence and data warehousing, it is still alive this word. A few years ago, all the companies used business intelligence infrastructures and technologies to build different forecasts for their organizations---for selling and so forth. Okay, so the problem is that these technologies are not able to handle different formats of data because they will only process structured data. So, unstructured data is not possible to analyze. For example, email contents, videos, pictures, and things like that. So for that reason, in terms of variety, volume, and velocity, the Big Data world is, let\'s say, an improvement over data warehousing and business intelligence work. Okay. So, could you explain what different formats of data are? Yes, for sure, for sure. Um, sorry, but when you chat, I need to change the window. An example of structured data---yes, you can write in Excel. Structured data, for example. Yes, that is a correct mention. So, Puja, is it correct? Sorry for my pronunciation. Excuse me for my pronunciation. So, in the world of data, we have two main categories: the first one is represented by structured data, and the second by unstructured data. Structured data is everything that you can handle with software in a strict way. For example, Excel can represent structured data. Spreadsheets can serve as a starting point for structured data because you have an organization where you have different columns and rows---so the values, okay? And you can apply some formulas to these values. For example, if you have the list of your monthly expenses, you can add formulas to sum them or to calculate the mean value, okay? That\'s something like that. So, in this case, using structured data, it\'s possible to apply and analyze your data better because all the content is in the same form. Another point for structured data is that it is something where you have an engine---an engine that can ensure the quality of data, the integrity of data, the security, and the accessibility of the data. So, for example, SQL Server, Microsoft SQL Server, Oracle Database, IBM DB2 Database, and so forth are all engines---in particular, relational database management systems---where you have structured data. So, talking about structured data, we are discussing well-known and well-defined formats of data where it\'s more simple to analyze. Okay. So, the other part is related to unstructured data. It could be, for example, emails, a PDF, or a Word document. Now, imagine having the list of your monthly expenses in a Word file. Are you able to sum all the expenses? Is it possible? So you need to use, I don\'t know, your phone, your smartphone, to perform the sum, and then you can type the sum into the Word document. So it\'s not possible to do that directly, okay? Why? Because a Word document, like an email, like videos, like pictures, and things like that, are all formats of raw data. So you cannot do everything easily without any limits. Okay, Andrea says, \"Are quantitative data equal to structured data, qualitative unstructured?\" Is this the question? I never think about structured data and unstructured data in terms of quantitative or qualitative data. Now I think that both are quantitative and qualitative because we have huge amounts of data. We can have huge amounts of data in relational database systems---think, for example, of all the customer credit information, so all the credit cards and all the customer information handled by an RDBMS in a bank, for example. And in the same way, we have a lot of information in\... No, I don\'t think that there are some differences in terms of qualitative and quantitative. So, got it. Thank you for the explanation. Yes, okay, great, great. Okay, but I have good---no, it\'s not good news---a bad news because we have\... A bad news because we have\... A bad news because we have another kind of format that is called semi-structured format. So, what is semi-structured format? So, semi-structured is something between structured and unstructured data. So what do we have? Where the data doesn\'t fit into a strictly rigid schema, okay, but still has an organization. An example of semi-structured data format could be XML files, JSON files, YAML files, and so forth. So we have at least a flat file---we have a text file---but the content of this file has an organization, a level of organization, okay? So in the case of JSON, we have a key-value organization; in the case of XML, we have tags, and so forth. Okay, so this is the definition of semi-structured data. The second bad news is that we have, daily worldwide, we produce only 20% of data in a structured way, and all the rest is represented by unstructured data and semi-structured data, falling into the unstructured data definition. Okay. So, and this is everything about the definition of Big Data. So, just to close together these five Vs---so volume, variety, vel\... Oops, sorry, I forgot to explain value and veracity. You are right, sorry, sorry. So, lessons before about veracity. So, veracity represents the data accuracy, so the level of accuracy of the data. And so, because it\'s important in order to have reliable insights, to have data at a good level of quality. So, it\'s important. So for that reason, during the years, the international and scientific community added to the definition of data the veracity and at least the fifth V, value, that is the ultimate goal of Big Data by driving meaningful insights. Because when we define data science, we say that the game is to extract meaningful insights so organizations can drive smarter decisions, improve customer experience, I don\'t know, gain competitive advantages. So, all this is not possible without value. So without value, Big Data is simply just a collection of information without purposes, okay? So together, these five Vs---velocity, value, variety, volume, and veracity---form the foundation of Big Data, guiding us to understand its complexity and the incredible potential it holds for innovation and strategic decision-making. Okay? So this is the power of Big Data and Big Data work, Big Data technology, so data science. Let me say all the data science work. And starting from this concept, you can imagine that we have, let\'s say, a wide world of information, a wide world of methodologies, technologies, so whatever you want to add. Okay. Standing by Gartner. Do you know Gartner? So Gartner\... Okay, let me\... Oops, I lost the presentation. Give me just one second. Okay. So, Gartner is an American technological research and consulting firm that uses data visualization and analysis tools to help information technology professionals all around to make decisions regarding technology investments for their organizations. Okay. So, this is Gartner. And in the report of 2023, Gartner defined the top trends in data and analytics. The first is thinking like business. So, focused on adopting a proactive approach to delivering value from platforms to ecosystems. So, as a manager, you must have a holistic view of both the technological and the organizational context. One is don\'t forget the humans. So, evolved to be too data-centric or technology-addicted, okay, and lost touch with the employees or the customers. So, don\'t forget the humans. These are the three top trends in data and analytics related by Gartner, okay? So in every report in 2023, you can find the report online; it is possible to download it from the internet. This is another way to represent this report, okay? So leaders leading to the top trends act more like an internal consulting company with its own product lines than a business support function that represents a cost center to the rest of the business, okay? This is the first point. That the leader must plan their DNA strategy in the context of the ecosystem. So we are talking again about thinking about an ecosystem and not a platform. And the last one, you plan on engaging with the supporters across the organization to understand the best approach to drive data and analytics adoption aligned with human psychology and value. And with the last point, we are connected to something related to the ethical part and other concepts that we will see together in the next lecture. Okay. So, generally speaking, sorry, this is the first part and introduction of our lecture. So, if you have questions, I can try to answer, okay? Questions. Is it clear? Everything is clear? Is it okay? Am I going too fast? Am I going too slowly? Let me say, you can also give me your feedback so I can tune the rest of the lecture. And for the next one, is it okay for you until now? Everything is clear? Guys, are you with me? Okay. Okay, great, great. Okay, perfect. Perfect music. So, great. So if we use\... Okay, we can have a first break, okay? Just 10 minutes, so I can rest my voice, and we will see soon in 10 minutes. Is it okay for you? Okay. Okay, so we are about to talk about data-driven companies, data culture, strategy, and data governance. We have introduced the main concepts about data-driven companies, governance, and so forth. I want to show you this spreadsheet. What do you think about this spreadsheet? We have some data---some economic data, I believe. We have references to different goods; we have milk. So, what do we have here? It\'s not clean data; it\'s unclear data. Or more simply, we have some data here. Okay, yes, it\'s saying nothing yet. Because we don\'t have the context. Data without context cannot become information. So we need data in order to have information. In the first case, it\'s just raw information, but not valuable information. Okay, so this is the important difference between data and information. Just to give you another thought: imagine a pile of scattered puzzle pieces. Each piece on its own is just a fragment, a raw element without much meaning. These pieces represent data. Okay, but they are the building blocks. Picture a completed puzzle---the scattered pieces are now arranged in a meaningful pattern, forming a recognizable image. So this major image is information; it\'s our information. Without data, there can be no information, but data alone is not enough. It needs to be processed, analyzed, and interpreted to become meaningful information---a transformation from the raw to the refined, from the chaotic to the coherent. This is the main difference between data and information at a high level. So, in this case, our context starting from the raw data is given by a pie chart. In this case, the context is represented, and we are able to reach the information level. Okay. Just an aside: in the next lecture, we will see this concept better. Now, I want to extend this concept and give you some other elements by talking about the DIKW pyramid. We have a pyramid where, on the bottom, we have data, and on the top, we have wisdom, passing through information and knowledge. So, D stands for data, I stands for information, K stands for knowledge, and W stands for wisdom. Imagine that our data is simply the color red---we have just this data point: red, our raw information, our raw data. Okay. So, red alone is not enough to give us information, knowledge, or reason. Adding the context---in this case, the information is that it\'s a traffic light. So now our thought process advances by computing the data and then the information. So, we are driving our car. We see a red color that refers to a traffic light. In this case, thanks to the information that we are able to interpret, we reach knowledge. The meaning, in this case, is that we have a red traffic light in our direction. So it\'s up to us---the wisdom, in this case, is to stop the car. Do you agree? And this is how we, as humans, approach the DIKW pyramid. We all start from data, from raw data, and we can reach the wisdom level by going through information and knowledge. Referring back to the example of the puzzle, data is the foundation of the pyramid. It\'s raw, unorganized facts and figures that lack context or meaning---like the scattered puzzle pieces. Going a level up, we have information. Data has been processed, organized, and given context. It\'s like arranging the pieces into a recognizable picture. After that, knowledge is information that\'s been interpreted and applied to a specific situation or problem. It\'s like understanding the picture and knowing how to use it. Lastly, wisdom is the highest level of the pyramid. Wisdom is the ability to apply knowledge to make judgments. This is how our mind works. So, if you think about it, we are data-driven humans. But we are talking about companies. So how do companies become data-driven companies? We can use both approaches; we can change from a bottom-up approach to a top-down approach. In this case, we have data---raw and processed facts. Here, the aim is to reach insight, meaningful insight. We are talking about meaningful output to obtain meaningful insight. Starting from data, here we have an example. Imagine being a manager in a company and saying, \"Do you know, last month the marketing team received 2,000 leads?\" What does it mean? Nothing---we don\'t have the context; we don\'t know if it\'s good news or bad news. We just know that last month, the marketing team received 2,000 leads. Adding the context---transforming data into information---last month, the leads grew by 20 percent over the previous month. So now we have information, and let me say we have good news. Why? Because leads grew by 20 percent over the previous month. In this case, we have data, we have information, and we have something useful for our insight---information leading to knowledge. Why is that? In this case, we have the insight: the growth in leads came from the website\'s live chat that was recently deployed. Now, as a manager, with the data-driven approach and thanks to our insight, we can make data-driven decisions. We can build a data-driven strategy. For example, another organizational unit of your company---perhaps the sales department---wants to introduce live chat in the portals. Starting from this insight from marketing, you can make a new decision based on data. So, for that reason, we can ask: what is a data-driven company? A data-driven company is a company that relies on data and information for decision-making, from daily business operations to long-term strategic decisions. Don\'t forget---you want to spread an internal culture that recognizes the value of data and information to face and overcome business challenges. This is the definition of a data-driven company. And you, as managers---I hope you become great managers---it\'s important to know and understand what a data-driven company is. Yes, for sure, you can ask me, why become a data-driven company? Simple: to increase productivity, increase efficiency, increase revenue, gain a competitive advantage, improve customer experience, and enhance customer retention. To support that, 81 percent of managers agree that data is at the heart of all decision-making today. We must rely on data. Why? Because today we live in a world that is, obviously---let me say---a tough world. Do you like it or not? It\'s a little bit complicated; there is a lot of volatility and uncertainty. To be more precise, we live in a world that we can define as VUCA. This is a concept---VUCA stands for Volatile, Uncertain, Complex, and Ambiguous. This concept originated with the U.S. Army War College during the Cold War in the late 1980s. It was used to describe the complex geopolitical landscape following the Cold War. However, VUCA has become a widely used term across various sectors and industries, including business, education, government, and so forth. Deep diving into the VUCA world, let me explain each letter: Volatile: Because things change daily and often unexpectedly. Events can unfold quickly with significant consequences. Uncertain: The future is difficult to predict. Using traditional methods or models of forecasting may not be reliable. Complex: Issues are multifaceted and often interconnected. It can be challenging to understand the root causes of problems. Ambiguous: Information can be unclear or open to multiple interpretations. This ambiguity makes it difficult to interpret situations and make decisions. We live in a VUCA world, and this concept is widely recognized today. For that reason, managers must rely on data. But we have an evolution of the VUCA world, which is the BANI world---a new concept used to describe the characteristics of our current global environment. This concept was introduced by the anthropologist and futurist Jamais Cascio. If you want, there\'s a book called Antifragile by Nassim Nicholas Taleb, which explains related ideas. Let me define the BANI world---it\'s an evolution of the VUCA world where: Brittle: Systems and structures---our social, economic, and technological structures---are increasingly fragile and susceptible to disruption. Anxious: The uncertainty and rapid change can lead to widespread anxiety on both individual and social levels. Non-linear: Cause-and-effect relationships are becoming less predictable, so events can unfold in unexpected ways, making traditional planning methods less reliable. Incomprehensible: The complexity of information can make it too difficult to understand the world around us. For that reason, the BANI framework is seen as an evolution of the VUCA concept. It highlights the general characteristics of a rapidly changing world, emphasizing the fragility and difficulty in understanding these changes. Now, in your opinion, do you have an example? The BANI world is not just a concept or a framework but is something real. What do you think? Do you think BANI is just a concept or is it something real? Do you have an example of where the BANI world can be observed as a real concept? Yes, war, for sure. It\'s a framework because it helps us better understand different concepts; that\'s why it\'s called a framework. Great. For example, COVID-19 is proof that we live in a BANI world. The difference is that events like El Niño or other problems are concentrated in a specific geographic location. COVID-19, like World War I and World War II, is a global situation affecting everyone. And the proof that we live in a BANI world is also evident in the financial crises. If you refer to the depression or the 2008 financial crisis, for example, it\'s another proof that we live in a BANI world. So, what is the concept? We are 8 billion people, which makes it more difficult to understand and predict events compared to a world with just around 1 billion people. This adds to the difficulty of the real world. From a supply chain perspective, even after almost three years since COVID-19 began, supplementary things such as chips and some robotics are still very unstable. Yes, because it\'s a consequence of a BANI world. It\'s not possible to predict the future because systems are non-linear; the future is incomprehensible. COVID-19 represents proof that we live in a BANI world, and it can probably be applied to everything in the real world---like in Spain yesterday with the heavy rains. This morning, I was talking with my wife about the problem in Spain yesterday. But I think that social media are part of this world as well. They contribute to anxiety and represent brittleness. So I think the social media world is part of this BANI world. From VUCA to BANI, the speed and impact of events and actions increase exponentially. Because from VUCA to BANI, there are almost 40 years of chronology, and for these reasons, I think that the BANI world is more related to the different technological evolutions we have had in the last 40 years. Now, it\'s not important whether we live in a VUCA or BANI world. What\'s important is that strategic traditional decisions need to be based on data rather than intuition, opinions, or experience. This is the main concept behind the VUCA or BANI world.