# 🍒 1.1. What is GarganText? GarganText is an innovative software technology for collaborative knowledge exploration and mapping. GarganText is a digital and collaborative work environment that operates on decentralized interfaces where each organization, institute, laboratory, user... is sovereign of the data it produces, defining its accessibility, sharing and reuse by the community of users. GarganText combines tools for automatic language processing, data mining, complex graph analysis and dynamic, interactive visualization of multi-level textual data: - At the micro level, document text can be highlighted and annotated, - At meso level, tables can be used to build dictionaries with controlled vocabulary (towards ontologies), - At the macro level, static and dynamic graphs (phylomemias) provide a global, shared view of the knowledge under construction. With just a few clicks and a few minutes, users can map a given body of knowledge (state of the art) from sets of unstructured texts, whatever their language. Documents can be sourced from the collaborative space or by connecting to all open-access data repositories via APIs. Each user contributes to enriching the artificial intelligence by annotating and classifying concepts, terms and expressions. GarganText is a free software (AGPL, CECILL) developed by CNRS at the Institut des Systèmes Complexes de Paris Île-de-France (ISC-PIF) and its partners. ## How to cite GarganText? For the moment : Alexandre Delanoë & David Chavalarias, Gargantext, CNRS/ISC-PIF. iscpif.fr/projects/gargantext/ ## Is GarganText free? Yes, GarganText is free software: * can be used by anyone, * open source, * can be modified by anyone. The software is "free" as long as you have the skills to run it. ## Architecture ### Why have your own instance? - To protect your data - To enrich your artificial intelligence - To enhance your knowledge ### What is a learning instance? - The instance builds lists from a sample corpus constructed on the instance. - The quality of the data on the instance produces a learning quality.