[MUSIC] Hello and welcome. The emergence of data is transforming society and the global economy in a remarkable manner. The abundance of data creates a need for strategies involving innovative technology to effectively extract, manage, and analyze the data and use it to our advantage. Data is not being considered a critical asset and a vital source of economic growth across various industries and organizations. And hence, there is a need for the new H-Manager to be data literate. In this course my aim is to introduce you to the nuances of generating insights, using data, and equip you with skills that will help shape your career in the future. With that aim in my mind, let me introduce you to Doctor David Pitt. Doctor Pitt is a professor and head of the department of actuarial studies and business analytics at Macquarie University. We'll be discussing with him the importance of generating insights using data today. Welcome to the course introduction, Doctor Pitt. Let me start this discussion by asking you about your views on the data industry in genera, and why do you think generating insights from data is so important these days? >> Thanks very much mate. Generating insights is important because businesses thrive when the correct decisions are made. Good decisions are often made when appropriate analysis of data has been performed. For example, suppose you are considering launching a new social media advertising campaign to inform people about ways to look after their health and well-being. On the surface, this sounds like a good idea, educating the public about the importance of living a healthy lifestyle. It will hopefully lead to a healthier society and less strain on health services, such as hospitals and doctors surgeries. How do we know though if our advertising campaign will be effective? Perhaps we should collect data on the use of social media platforms by individuals. Do these users look at advertising material and click-through it to find out exactly what is being advertised? Are the users of social media, those who may be likely to need advice about their health, or are they likely to be from parts of the community that are already healthy? The answers to these questions can only be obtained by collecting data on users of social media and on their involvement in advertising material shown on social media. With so many users of social media these days, we may not be able to analyze all users, but we may instead collect data from a representative sample. Thanks David. So can you maybe throw some light as to how we can use this data? >> Yeah, sure. Data from social media platforms can be used to create graphs of the age profile, or the gender split, or the number of hours spent on social media by users. Graphical data analysis, or it is as it is sometimes known these days, visualization, can be helpful in giving us a picture that summarizes the millions of data records we may have available. We can also use the data to run prediction algorithms and assess which users are most likely to click on advertising material. And whether those users are the people most likely to benefit from advertising that aims to improve public health knowledge. >> So, David, in your answer you used words like graphical data analysis and visualization. What do these terms mean and why are they important? Sure, we all know that data can be converted into information using statistical analysis. Data is excellent in providing very specific information. Data by itself, however, is not easily interpretable. Looking at a spreadsheet of data, it's hard to determine trends, or relationships between variables. Data visualization comes to the rescue here, and allows data to be presented both graphically and interactively, such that it becomes easier to understand and to draw inference. Technology has only recently made graphical representation of very large amounts of data accessible to individual users on their desktops and has led to an explosion of data visualization. The reason why graphics are so important is because they make data interpretable. Now, Smith, I have a question for you. With all the digital transformation that we're seeing in industry at the moment, what do you see as the role for the new age manager in harnessing this transformation? >> Sure, yeah, I agree that many industries are going through digital transformation. As a result of that, data collection has become easier thanks to lots of sensors and devices. At each step of a business process we have lots of data that can be analyzed. Businesses are investing in technologies that would make information extraction and insights generation process easier. As a result, the skills of manager and decision makers are also changing, and there is a huge demand for data literacy nowadays. >> Well, thanks Smith, but we've talked quite a little bit about the industry and related aspects. I think the students would want to know now what's in the course. So what can we expect from each of the six weeks of the course? >> Sure, sure, this six-week course is a mix of theory and practical lessons. The focus will be on doing as many hands-on tasks as possible. So let's see what we can expect from each of the six weeks. In the first week, I'll give you an overview of the generating insights subject. You will learn how data is now being treated as an asset in companies. We'll familiarize ourselves with the concept of the data value chain. And finally, I'll introduce you to Tableau which is a platform will be using throughout the course for generating insights. In second week, we'll focus on data and its structure, what are variable contained in it, and the types, and how Tableau treats data in the form of measures and dimensions. Then we'll move on to the statistical concepts of measures of central tendency and measures of dispersion. In third week we'll continue to learn some more statistical concepts by covering normal distributions and histogram in detail. Will also studied the empirical rule to see if our distribution follows a normal curve or not. And finally, finish the week by getting an understanding on measures of relationships between variables mainly, covariance and correlation. In the fourth week, we'll shift our focus from the world of statistics to the world of visualizations. We'll probe with the help of Anscombe's quartet that you need visualizations, along with some of these statistics, to understand complete data story. We'll get into the details of cleaning the data. More often than not, on a data insight’s project, you'll spend a significant amount of time cleaning the raw data. And hence, it's important that you learn this technique. Once our data is clean, we'll start to create graphs and charts. In the fifth week, we'll continue to learn a few more graphs and charts and understand their purpose. We'll move onto a social media case study and find out insights from those data. And we'll finish the week by learning about an important visualization tool called Dashboards. We'll see their types and how they're used in a company. And finally, in the sixth week I'll introduce you to the concept of predictive insights using regression analysis. We will also discuss demand forecasting. And we'll end the week by learning about smoothing methods that can help in both forecasting, as well as in reducing the amount of unwanted noise in the data. So thank you David for your time today. >> Thanks, mate, no problem. So, good luck. [MUSIC]
