Book review: BIG DATA DEMYSTIFIED
How to use big data, data science and AI to make better business decisions and gain competitive advantage
Genres:
- Information management
Review posted on:
14.10.2021
The number of pages:
240 pages
Book rating:
3/5
Year the book was published:
First edition published 2018
Who should read this book:
- People who have no knowledge of Big Data and want to know the basics.
- CEOs or managers in positions to make the decision to implement Big Data and Analytics in their company/organization.
- People in IT or Analytics who already have experience will get little or no value from this book.
Why did I pick up this book and what did I expect to get out of it:
One of the reasons I picked up this book was that I want to know more about Big Data not just from the perspective of the person who is targeted by Big Data (the customer/user), but also from the perspective of those who use Big Data to create better experiences for their customers/users. To see the “big picture” or as the saying goes ” The grand scheme of things”. The fact is that Data, not just Big Data is increasingly affecting our lives. Based on Data more and more companies are creating products and services and ways how we interact with them, and one of the issues I have with this is that we as individuals don’t know much how this is happening.
When reading the covers of the book the author promised deeper insight and understanding of different actions a company should take based on data. My expectation was that the book would be packed with examples and case studies that would serve as a “guide”, or at least provide guidelines on how to, where, and why to collect which data. More specifically which data you should collect with which devices and by which method? And then what are the best practices to use and interpret data collected this way to make better decisions?
My thoughts about the book:
If you ask me if you should pick up this Big Data Demystified I would say it depends on how much you know about Big Data, and are you already doing something with data analytics. If you are just starting out, then I think that this book might deliver some value from the perspective of technical demands and the evolution of Big Data. But it will deliver little value in helping you plan which data and analysis you should do to help you out with your business.
If you are in a management or HR position and are charged with developing your analytics team then Big Data Demystified will offer you some insight on what type of skills your team should have as the author covers the structure of an analytics team with a description of skills for each role. That was actually really interesting to see as many people can’t tell the difference between different roles in data analytics. But if you are already well versed in the “data world” then the book will offer you little value. I see this book as one of those “introduction to ….. book”.
All in all the book sticks to its “red thread”. What I mean by that is that the author is not mindlessly jumping from one topic to another. The information in the book is well connected and in a logical sequence and I can say I got an overview of the Big Data ecosystem. But that’s it. In my opinion, the author fails to deliver in-depth case studies that I feel would make the book perfect. Don’t get me wrong the book has simple case studies, that offer some basic insights that may make you interested in Big Data, but where is the in-depth case study, that will make you go “wow, I didn’t know that could be done. “
If you picked up this book please let me know what you think about it in the comment section.
A short summary of the book:
The author starts the book with a short history recap of the first issues involving big data. He talks about storage and hardware capabilities, then you learn a little bit about software so that you have an overview of the big data ecosystem. Then the author continues how customer data is important in today’s world and he suggests a couple of steps you should have in mind when preparing your data strategies, how your data team should look like etc…
The author also goes into artificial intelligence and machine learning and what it takes to develop them. He also explains the types of databases that are out there, and which analytic tools and programming languages you need to start with data analytics. Many people just hoard data which is not always good from different perspectives. Unstructured data that you can’t structure or use in any other way is not worth having and also if you’re operating in the European Union you have to keep in mind the constraints of GDPR or you will be in violation of the EU law. The author emphasizes that you are responsible for abiding by the laws where you operate and that not abiding by them you can make more harm to your business than good.
My notes from the book:
- A machine learning technique that is increasingly used today is Artificial Neural Network (also called deep learning).
- Despite early enthusiasm, the AI we have today is still "narrow AI" useful for the specific application for which it was designed and trained. Deep learning has brought only marginal improvements.
- AI is very dependent on large quantities of diverse, labeled data. AI trained with insufficient data will make more mistakes.
- Your customers are continuously expressing preferences as teh type search terms and subsequently, select or ignore the results.
- You'll need a broad understanding of programming languages and analytic tools, not only the most recently trending technologies, but also a strong understanding of traditional methods.
- The accuracy of your CLV (customer lifetime value) calculation increases with your ability to sub-segment customers and your ability to compute the corresponding churn rates. That comes with the use of such as webpage navigation, email open rates, content downloads, and activity on social media.
- Make a list of all the data sources you would need to link to get the full view of your customer's interactions.
- Not all ideas will work in analytics due to insufficient or poor-quality data, noise in the data, or the process that we are examining does not lend itself to standard models.
- The formulation of a problem is often more essential than its solution.
- As you move into more advanced analytics you will need to choose analytic models. Models are sets of formulas that approximately describe events and interactions around us. We apply models using algorithms, which are sequences of actions that we instruct a computer to follow like a recipe.
-
The strength of your analysis depends on:
1. Discovering what data is most meaningful.
2. Selecting an analytical tool appropriate to the task.
3. Having enough data to make the analysis work. -
For each interaction point you have with your customers make an inventory of:
1. What data you can possibly collect.
2. What are the potential uses of that data.
3. What privacy and governance factors you should consider. -
Gartner’s framework for classifying application areas of analytics:
1. Descriptive analytics:
- It is the process of collecting, cleaning, and presenting data to get immediate insights.
- To do descriptive analytics well you need specially designed databases/data warehouses, a dedicated BI tool for constructing and delivering regular reports and dashboards, and a system that allows your business users to do self-service analytics.
2. Diagnostic analytics:
- The diagnostic efforts consist largely of bringing together potentially relevant source data and teasing out insights either by creating graphs that visually illuminate non-obvious trends or creating new fields from existing data.
- You will need creativity and visualization skills to discover and communicate insights through well-designed tables and graphs. 3. Predictive analytics:
- Helps you understand the likelihood of future events, such as providing revenue forecasts or the likelihood of credit default.
- RFM is an example of a traditional customer segmentation approach, based on purchase history.
4. Prescriptive analytics:
- It tells you what should be done. Have in mind things like optional pricing, product recommendations, minimizing churn, fraud detection, and inventory management…
