Data Glossary Lab

A categorical variable is a variable that can have one of a limited number of possible values (categories) without any intrinsic ordering involved.

The Bias-Variance Tradeoff is the balance between two types of errors that prevent an algorithm from generalizing beyond its training set.

Bayes’ Theorem is a mathematical equation used to calculate conditional probability, determining the likelihood of an event based on prior knowledge of conditions related to that event.

Apache Spark is an open-source, multifunctional parallel processing framework designed for analyzing and modeling Big Data. Unlike traditional processing tools that handle data on a single machine, Spark enables data and computations to be spread over clusters with multiple nodes. It is the industry standard for high-speed data processing because it primarily operates in-memory, allowing it to process massive datasets up to 100 times faster than older disk-based systems like MapReduce. Spark represents the "Heavy Lifting" capability of an organization, providing the infrastructure necessary to execute complex machine learning and real-time analytics at a global scale.

In the field of Data Science, Bias refers to the distance between the average prediction of a model and the true value we are trying to predict. High Bias indicates that the model is overly simplistic, failing to capture the underlying trends of the data—a phenomenon known as Underfitting. Beyond the mathematical dimension, the term also encompasses Algorithmic Bias, where a model reproduces or amplifies prejudices inherent in the training data, leading to unfair or skewed decisions against specific groups. If Bayes’ Theorem is about updating our beliefs, Bias is about the "blind spots" that prevent the model from seeing the full picture.

Business Analytics (BA) is the practice of using historical and current data to discover operational insights, anticipate market trends, and make data-driven business decisions. Unlike simple reporting, which only describes what happened, BA focuses on why it happened and what is likely to happen next. It is the bridge between raw data and executive action, transforming numbers into a roadmap for growth. 

The Binomial Distribution is a discrete probability distribution that models the number of "successes" in a fixed number of independent trials. It is the mathematical foundation for scenarios where there are only two possible outcomes—often simplified as Success vs. Failure, Yes vs. No, or Default vs. Payment. For a distribution to be considered Binomial, it must meet four specific criteria: the number of trials (n) is fixed, each trial is independent, there are only two possible outcomes, and the probability of success (p) remains constant throughout the process. It allows a Data Scientist to move from guessing to calculating exactly how likely a specific volume of results is within a given sample.

Logistic regression is a regression algorithm that uses a logistic function on the input features to predict the class probability or directly the class label for the target variable. In the second case, the output represents a set of categories instead of continuous values, meaning that the logistic regression acts here as a classification technique. A typical data science use case for logistic regression is predicting the likelihood of customer churn.

Normalization is the process of rescaling the data so that all the attributes have the same scale. Normalization is necessary for making a meaningful comparison between the attributes and also is required for some machine learning algorithms.