Building a pipeline for stellar network characterization using Kedro

Abstract

This paper presents the steps required to implement a data processing pipeline using the Kedro framework for preprocessing astronomical data to characterize stellar groups and determine whether the target dataset constitutes a small-world network. The tool’s characteristics and the criteria analyzed for its adoption are detailed, along with the steps for querying and downloading data, defining preprocessing stages, their implementation, execution, and final results. The conclusions also analyze the value that these tools contribute to the repeatability and reliability of both the data preprocessing stages and the subsequent data analytics performed on said data, aimed at obtaining reliable, repeatable, and high-quality data analysis flows.