Update README.md

README.md

@@ -19,25 +19,24 @@ should probably proofread and complete it, then remove this comment. -->
 
 # neuralmind/bert-base-portuguese-cased
 
-This model is a fine-tuned version of [neuralmind/bert-base-portuguese-cased](https://huggingface.co/neuralmind/bert-base-portuguese-cased) on the None dataset.
-It achieves the following results on the evaluation set:
-- Loss: 0.0460
-- Accuracy: 0.8367
-- F1: 0.7871
-- Recall: 0.8194
-- Precision: 0.7687
-
 ## Model description
 
-More information needed
+This model is a fine-tuned version of the pre-trained neuralmind/bert-base-portuguese-cased model. It was specifically adapted to classify Brazilian legislative proposals (PLs) as either favorable or unfavorable to women’s rights, based on the content of their summaries (ementa) and full text (inteiro teor). The model aim is to uderstand the nuances of gender impacts in legal contexts.
+The model leverages the BERT architecture, which is designed for natural language understanding tasks. The use in this specialized task allows the model to identify patterns and terminology indicative of how a project of law aligns with women’s rights.
 
 ## Intended uses & limitations
 
-More information needed
+Primary Use: To classify Brazilian legal proposals (PLs) as either favorable or unfavorable to women’s rights.
+Target Audience: This model is intended for use by legal professionals, gender equality advocates, and researchers analyzing legislative texts, as well as automated systems that categorize legal documents based on gender equity considerations.
+Application Areas:
+Legislative analysis
+Gender equality advocacy
+Political and legal research
+Automated legal document classification
 
 ## Training and evaluation data
 
-More information needed
+The model was fine-tuned using a custom dataset of legislative proposals (PLs) from Brazil, specifically focused on women's right topics, though details on the exact composition of the dataset are unavailable. The dataset contains both ementas and full texts of PLs.
 
 ## Training procedure
 
@@ -82,3 +81,7 @@ The following hyperparameters were used during training:
 - Pytorch 2.5.1+cu121
 - Datasets 3.2.0
 - Tokenizers 0.21.0
+
+### Ethical considerations
+This model is designed to classify legislative texts, which may have significant social and political implications. As such, careful consideration should be given to how the model's outputs are interpreted and used, especially in sensitive contexts.
+The dataset used for training the model must be periodically reviewed and updated to ensure it reflects current legislative language and evolving gender equality standards.