Div Explorer

In traditional Exploratory Data Analysis (EDA), analysts often face "information overload." Tools frequently go over hundreds of redundant patterns, forcing analysts to manually sift through overlapping results.

To solve this, I develop and document code for “actionable labels,” filtering data based on ease of interpretation, high coverage, and anomalous behavior. DivExplorer uses these labels to guide which patterns to surface.

This experience taught me that effective computer science is not just about processing data, but about balancing automation while executing clear deliverables.