Cartolabe: A Web-Based Scalable Visualization of Large Document Collections
This system addresses the challenge of making large-scale text data visualization accessible to a wide audience, including non-experts, though it is incremental as it builds on existing NLP and projection methods.
The authors tackled the problem of visualizing and exploring large textual corpora by developing CARTOLABE, a web-based multi-scale system that uses NLP and UMAP projection to enable interactive exploration, handling datasets like Wikipedia with 4.5M entries and achieving broad accessibility for non-experts.
We describe CARTOLABE, a web-based multi-scale system for visualizing and exploring large textual corpora based on topics, introducing a novel mechanism for the progressive visualization of filtering queries. Initially designed to represent and navigate through scientific publications in different disciplines, CARTOLABE has evolved to become a generic framework and accommodate various corpora, ranging from Wikipedia (4.5M entries) to the French National Debate (4.3M entries). CARTOLABE is made of two modules: the first relies on Natural Language Processing methods, converting a corpus and its entities (documents, authors, concepts) into high-dimensional vectors, computing their projection on the 2D plane, and extracting meaningful labels for regions of the plane. The second module is a web-based visualization, displaying tiles computed from the multidimensional projection of the corpus using the U MAP projection method. This visualization module aims at enabling users with no expertise in visualization and data analysis to get an overview of their corpus, and to interact with it: exploring, querying, filtering, panning and zooming on regions of semantic interest. Three use cases are discussed to illustrate CARTOLABE's versatility and ability to bring large scale textual corpus visualization and exploration to a wide audience.