The Big Data eCourse on the Laudius platform offers more than just theoretical knowledge. In addition to learning what Big Data is and exploring the ethical considerations surrounding it, the course also explains the four most commonly used statistical analysis methods in Big Data. These four types of analysis are widely applied not only in Big Data projects but also in many other forms of research.
The Big Data eCourse uses the book ‘Succes met Big Data’ as its foundation, but the learning material goes well beyond the brief explanations provided in the book. Each of the four major analysis methods is supported with additional resources, including detailed video lessons and PDF guides that clarify how the methods work in practice. As an extra learning opportunity, the course also includes case studies that allow students to apply and practice these analysis techniques in realistic scenarios.
In the eCourse, Excel is combined with the Real Statistics add‑on to analyze the data used in the case studies. Real Statistics is a free add‑on that can be installed in Excel using the provided installation guide. It makes performing statistical analyses much easier compared to building formulas manually in Excel. Students are, of course, free to use SPSS, R, Python, or any other preferred tools, as long as they have access to them and indicate this when submitting their homework.
There is a full blog on my website dedicated to the decision tree analysis method, including a practical example. In short, a decision tree is an analysis technique used to map out choices that follow one another. These choices can be simple or part of a more complex decision-making process within research. The blog, titled “Using a decision tree to determine where to go”, explains how this method works and illustrates it with a real example.
With the cluster analysis method, the student groups the available data into clusters. These clusters are formed based on similar attributes shared by the data points. Depending on the type of clustering used, the resulting groups may contain the same number of data points, or they may vary in size, reflecting natural patterns in the dataset.
In the book, the linear regression analysis method is introduced as the foundation of regression techniques. As an additional resource, the Big Data eCourse includes a video that explains multiple regression in more detail. Most research, especially Big Data projects, relies on multiple regression, because more than one variable (or characteristic) typically influences the outcome being studied.
Nearest Neighbor is, in simple terms, a combination of clustering and linear regression techniques. It is widely used to recognize patterns and images, making it an essential method in modern data analysis. Today, this technique plays a major role in systems that rely on public‑space observation cameras, which track and classify what they capture. Because of this, Nearest Neighbor is closely connected to privacy concerns. Important questions include:
Which images may legally be used
How the analysis is configured to respect the privacy of people moving through the public spaces being monitored
This makes it crucial to maintain a balance between privacy and safety, ensuring that data analysis supports public security without violating individual rights.
There are, of course, many additional analysis methods that can be used to examine Big Data. However, most of them are variations or extensions of the four core methods already described. In practice, these foundational techniques are applied first to interpret the data and to determine which advanced analysis, such as predictive analytics or other specialized approaches, should be used next.