text2graphAPI: A library to transform text documents into different graph representations

Andric Valdez-Valenzuela, Helena Gómez-Adorno

Abstract: This paper introduces a new Python API called text2graphAPI. It is an easy-to-use library for transforming text documents into different graph representations, such as Word-Cooccurrence, Heterogeneous, and Integrated Syntactic Graphs. In addition, it contains a text pre-processing module that supports input text in different languages: English, Spanish, and French. These generated graph structures can be used to solve tasks in various areas, such as Authorship Analysis, Information Retrieval, and Topic Classification, to name a few.