Data Comprehension
Vidya Subramanian · 2025
This chapter focuses on Exploratory Data Analysis (EDA). In Data Comprehension, there are three critical steps: Data Analysis, Data Profiling and performing EDA. Data Profiling helps us detect issues with data quality. A combination of different methods are used to discover and understand the data: univariate analysis; bivariate analysis; and multivariate analysis. A Stratified pattern sorts the data into distinct groups. A Blocked pattern is a situation in which subsets of observations belong together. In a Nested pattern, we collect data from multiple individuals in a group. A Run chart plots the process data in the order of collection. Data visualization techniques are used to explore unstructured data. Text analysis techniques can be applied to unstructured data to identify themes, patterns, and sentiments. Network analysis techniques explore relationships between entities in unstructured data. Dimensionality reduction techniques reduce the dimensionality of unstructured data, making it easier to visualize and analyze.