Introduction: Intimate partner violence is a major public health problem and a violation of women's human rights. It is important to explore the speech of the victim at the time of the complaint. Objective: To analyze the text classification with word cloudas a tool to understand the pattern of patient functioning, complementing the qualitative analysis. Method: It is a text classification study with a word cloud analysis technique, complementing the qualitative analysis. The sample is fifty-six women victims of self-reported domestic violence, who came from a public service in South of Brazil. Results: A word cloud was developed from the speech of the 56 participants. Discussion: The women's discourse in the interviews had focused on understanding what had really happened in their relationship, reporting their abusive situations. Studies using this technique benefit from including heterogeneous patients considering their idiosyncrasies to develop a complex non-linear pattern relating predictors to the clinical outcome.