Atlas: An Interactive Graph-Based File System for Enhanced Document Relationship Visualization

Abstract

Traditional hierarchical file systems often obscure relationships between documents stored in different folders, limiting users’ ability to comprehend contextual connections. To address this limitation, this paper presents the development of Atlas, a browser-based application that transforms conventional file systems into interactive graphs. Atlas leverages natural language processing (NLP) techniques to parse documents and generate visual structures linking related files through shared keywords, providing users with a more intuitive way to explore and understand their data. The application was developed using the Scrumban agile methodology, enabling iterative improvements through user feedback obtained from presentations, surveys, and interviews. This approach ensured the design and functionality of Atlas aligned with user needs throughout the product lifecycle. Successfully deployed within an organization, Atlas enhances documentation comprehension and boosts productivity by making relationships between documents readily accessible and visually clear. This innovation highlights the potential for graph-based visualization tools in reimagining file system interactions.