Update README.md

README.md

@@ -10,7 +10,7 @@ tags:
 - multilabel-classification
 - portuguese
 - administrative-documents
-- intelligent-stacking
+- stacking
 - ensemble-learning
 - bert
 - tfidf
@@ -19,36 +19,34 @@ base_model:
 - neuralmind/bert-base-portuguese-cased
 ---
 
-# Intelligent Stacking: Multilabel Portuguese Administrative Document Classifier
+# CouncilTopics-PT: A multi-label classifier for Portuguese municipal meeting topics.
 
 ## Model Description
 
-**Intelligent Stacking** is an advanced ensemble learning system specialized in multilabel classification of Portuguese administrative documents. The model combines 12 base models with intelligent meta-learning to achieve high performance on municipal document categorization tasks.
+**CouncilTopics-PT is an ensemble learning system specialized in multilabel classification of Portuguese Municipal topics from Meeting Minutes. The model combines 12 base models with meta-learning to achieve usable performance on municipal topics categorization tasks.
 
 **Try out the model**: [Hugging Face Space Demo](https://huggingface.co/spaces/anonymous12321/PT-AdminDocs-Classifier)
 
 ### Key Features
 
-- 🧠 **Intelligent Meta-Learning**: Advanced ensemble combination using stacked generalization
+- 🧠 **Meta-Learning**: Ensemble combination using stacked generalization
 - 📚 **12 Base Models**: 3 feature sets × 4 algorithms for robust predictions
-- 🇵🇹 **Portuguese Optimized**: Fine-tuned for Portuguese administrative language
-- ⚡ **High Performance**: F1-macro score of 0.5486
+- 🇵🇹 **Portuguese Optimized**: Prepared for the Portuguese language
 - 🏢 **22 Categories**: Comprehensive municipal administrative document classification
 - 🎯 **Dynamic Thresholds**: Optimized per-category decision boundaries
 
 ## Model Details
 
-- **Architecture**: Intelligent Stacking with Meta-Learning
+- **Architecture**: Stacking with Meta-Learning
 - **Base Models**: 12 diverse classifiers (LogReg, Random Forest, Gradient Boosting)
 - **Feature Engineering**: TF-IDF + BERTimbau embeddings + Statistical features
-- **Meta-Learner**: Advanced ensemble combination algorithm
-- **Categories**: 22 Portuguese administrative document types
+- **Meta-Learner**: Ensemble combination algorithm
+- **Categories**: 22 Portuguese administrative topic labels
 - **Training Method**: Cross-validation stacking with dynamic threshold optimization
-- **Framework**: Scikit-learn + Transformers
 
 ## How It Works
 
-The Intelligent Stacking system operates in multiple stages:
+The Council topics system operates in multiple stages:
 
 1. **Feature Extraction**: Three complementary feature sets
    - TF-IDF vectorization (word and character n-grams)
@@ -60,7 +58,7 @@ The Intelligent Stacking system operates in multiple stages:
    - Random Forest
    - Gradient Boosting
    
-3. **Meta-Learning**: Intelligent combination of base model predictions using advanced stacking
+3. **Meta-Learning**: Combination of base model predictions using stacking
 
 4. **Dynamic Thresholds**: Per-category optimized decision boundaries for multilabel output
 
@@ -125,7 +123,7 @@ print("Predicted categories:", predicted_labels)
 
 ## Categories
 
-The model classifies documents into 22 Portuguese administrative categories:
+The model classifies topics into 22 Portuguese administrative categories:
 
 | Category | Portuguese Name |
 |----------|-----------------|
@@ -197,7 +195,7 @@ The model was trained on a curated dataset of Portuguese municipal council meeti
 ## Limitations
 
 - **Language Specificity**: Optimized for Portuguese administrative language
-- **Domain Focus**: Best performance on governmental/municipal documents
+- **Domain Focus**: Best performance on municipal documents
 - **Computational Requirements**: Requires significant memory for all model components
 - **Threshold Sensitivity**: Performance depends on carefully tuned per-category thresholds
 - **Class Imbalance**: Some categories may have lower precision due to limited training examples