Data Glossary Lab

Principal component analysis (PCA) is a statistical technique of factor analysis and dimensionality reduction that transforms a set of possibly correlated initial features into a smaller set of linearly uncorrelated features called principal components. In this way, PCA preserves as much variance in the dataset as possible, while minimizing the number of features.

The Pareto Principle, commonly known as the 80/20 Rule, is a statistical concept stating that roughly 80% of consequences come from 20% of the causes. While standard reporting might view all data points as having equal weight, Pareto Analysis recognizes an unequal distribution of impact. It answers the question: "Which minority of inputs is driving the majority of my outputs?" By sorting data by frequency or impact, it separates the "Vital Few" (the small percentage of elements that create the most value or damage) from the "Trivial Many" (the large percentage of elements that have minimal impact). In Data Science, this is crucial for feature selection, identifying that a small subset of variables often holds the most predictive power in a model.

Power BI is a unified business intelligence ecosystem, not just a visualization tool. While Excel allows you to organize data, Power BI allows you to experience it. It transforms isolated spreadsheets and disparate databases into a cohesive, immersive narrative. The core difference is connectivity and scale. In a traditional setup, data lives in silos,trapped in SQL servers, CRMs, or local files. In Power BI, these sources are fused into a single Semantic Model. The interface serves as a bridge between raw numbers and human insight, allowing users to aggregate millions of rows from different sources and visualize them instantly. It solves the "version of truth" problem. Instead of emailing five different spreadsheet versions with conflicting numbers, Power BI offers a single, cloud-based destination that updates automatically. It is intelligence through integration.

In Python , a Dictionary is a built-in, mutable data structure used to store collections of data in key-value pairs. Unlike sequences which are indexed by a range of numbers, dictionaries are indexed by unique "keys," allowing for high-performance data retrieval and organized data mapping.

Predictive analytics represents the convergence of historical intelligence and forward-thinking technology.

Prescriptive analytics is where data stops telling you stories and starts giving you orders. It analyzes your data, evaluates possible scenarios, and recommends specific actions to achieve your desired outcomes.