Which type of analysis is used to discover correlations within a dataset?

Correlation analysis is specifically designed to identify and quantify relationships between variables within a dataset. It focuses on measuring how closely two or more variables move in relation to each other, determining whether a change in one variable is associated with a change in another. The results are usually expressed in terms of a correlation coefficient, which indicates the strength and direction of the relationship.

In contrast, regression analysis goes a step further by not only examining the relationship between variables but also modeling the relationship to predict one variable based on another. While both correlation and regression analyses are useful, correlation analysis is tailored for the primary purpose of discovering associations without the intent to predict outcomes.

Sentiment analysis, primarily used in natural language processing, evaluates and interprets feelings or opinions expressed in text. Predictive analysis uses statistical algorithms and machine learning techniques to forecast future outcomes based on historical data. Both of these do not focus specifically on discovering correlations.

Therefore, correlation analysis is the most suitable choice when the objective is to uncover relationships within a dataset.