Brazilian Court Documents Clustered by Similarity Together Using Natural Language Processing Approaches with Transformers

Recent advances in Artificial intelligence (AI) have leveraged promising results in solving complex problems in the area of Natural Language Processing (NLP), being an important tool to help in the expeditious resolution of judicial proceedings in the legal area. In this context, this work targets the problem of detecting the degree of similarity between judicial documents that can be achieved in the inference group, by applying six NLP techniques based on transformers, namely BERT, GPT-2 and RoBERTa pre-trained in the Brazilian Portuguese language and the same specialized using 210,000 legal proceedings. Documents were pre-processed and had their content transformed into a vector representation using these NLP techniques. Unsupervised learning was used to cluster the lawsuits, calculating the quality of the model based on the cosine of the distance between the elements of the group to its centroid. We noticed that models based on transformers present better performance when compared to previous research, highlighting the RoBERTa model specialized in the Brazilian Portuguese language, making it possible to advance in the current state of the art in the area of NLP applied to the legal sector.