In text classification, many recent works deal with the interpretation of neural networks by the production of prediction explanations. In this context, EBBE-Text offers an interactive visualization of the decision boundary, positionings of texts with respect to it (and thus the certainty of a network in its predictions), paths leading from a text to the decision boundary, information concerning the proximity between texts, trought different localities in the text representation space. These information make it possible to intuit how the classification neural network functions and thus helps in its interpretability. Our method creates data on the decision boundary and then uses simplistic fuzzy sets to create a graph before linearly aligning the created data on the decision boundary. Finally an iterative process places the input data around the linear arrangements of the boundary data.