Chapter 1: The Enduring Relevance of Big Data

Big data is here to stay, but it's essential to grasp when and why certain data formats yield greater value. Many of our supply chain students pursue a minor in Business Analytics, where we delve into a variant of wide and small data known as Association Analysis, particularly in our Business Data Mining course.

In our discussions, we refer to a significant trend: a transition from relying solely on big data to incorporating both small and wide data. For example, wide data encompasses information sourced from various places about individual entities, such as products or customers. Consider a scenario where you want to gather insights on customers. Traditionally, one might create a customer database with a fixed number of columns for demographic and contact details. However, that approach is often insufficient today. Businesses now aim to capture additional details, such as the frequency of customer service calls or feedback from product reviews. This evolution highlights several key characteristics of wide data:

1. Rigid data tables with a set number of columns are less effective, as customers may have varying numbers of service interactions and product reviews.
2. Data originates from multiple sources, including demographic information, service logs, and online product reviews.
3. Not all data is structured; for instance, product reviews often contain unstructured information.

Section 1.1: The Essence of Small Data

While small data may seem like the antithesis of big data, it offers a more nuanced understanding of individual items. For example, if you want to analyze why a specific product is underperforming in sales, you might typically consider various factors such as seasonality, geographic location, and competition. However, a deeper examination of when and why a product falls off a potential customer's shortlist can reveal significant insights for manufacturers, allowing them to influence buying decisions before the purchase. If suppliers are aware of competing products in a customer's shortlist, they can implement strategies such as competitive pricing, instant discounts, or unique product features to sway the decision.

The effectiveness of small data is often attributed to the unique attributes of specific data points, which may not conform to broader trends observed in big data. Consequently, conclusions drawn from big data analyses may not apply universally in these cases.

Subsection 1.1.1: Expanding Our Analytics Curriculum

Our Analytics major encompasses a wider array of topics. The curriculum addresses unstructured data, real-time streaming data, AI-driven insights extraction, and cloud-based analytics. These elements facilitate the examination of not just wide and small data but also big data. Streaming data, which continuously evolves, is another area of focus; relying solely on static data can quickly become obsolete. We introduce simple methodologies for working with streaming data in our minor, with an in-depth exploration in the major.

Chapter 2: Insights from Video Resources

To further enrich your understanding of these concepts, check out the following videos:

The first video titled "Common Mistakes in Big Data Models" discusses prevalent errors in big data strategies and offers tips for improvement.

The second video, "Sunburst Charts in Excel - Everything You Need to Know," provides a comprehensive overview of using sunburst charts for data visualization in Excel.