phoenix28/bloom-taxonomy-classifier

Bloom's Taxonomy Classifier

This is a fine-tuned BERT model for classifying educational learning outcomes according to Bloom's Taxonomy.

Model Description

This model is based on bert-base-uncased and has been fine-tuned to classify learning outcomes into six categories of Bloom's Taxonomy:

• Remember
• Understand
• Apply
• Analyze
• Evaluate
• Create

Intended Use

This model can be used to automatically classify learning outcomes or educational objectives into Bloom's Taxonomy levels, helping educators and curriculum designers understand the cognitive complexity of learning goals.

Usage

from transformers import AutoTokenizer, AutoModelForSequenceClassification
import torch

# Load model and tokenizer
model_name = "phoenix28/bloom-taxonomy-classifier"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForSequenceClassification.from_pretrained(model_name)

# Example text
text = "Describe the key concepts of machine learning."

# Tokenize and predict
inputs = tokenizer(text, truncation=True, padding='max_length', max_length=128, return_tensors='pt')

model.eval()
with torch.no_grad():
    outputs = model(**inputs)
    logits = outputs.logits

# Apply sigmoid for multi-label classification
sigmoid = torch.nn.Sigmoid()
probs = sigmoid(logits).squeeze().numpy()

# Map to categories
categories = ['Remember', 'Understand', 'Apply', 'Analyze', 'Evaluate', 'Create']
threshold = 0.5

predictions = {}
for i, category in enumerate(categories):
    predictions[category] = {
        'probability': float(probs[i]),
        'predicted': bool(probs[i] > threshold)
    }

print(predictions)

Training Data

This model was trained on a dataset of learning outcomes labeled with Bloom's Taxonomy categories.

Model Performance

The model achieves strong performance across all Bloom's Taxonomy levels with:

• Micro-averaged F1 Score
• Per-class precision and recall metrics

See the training logs for detailed evaluation metrics.

License

Apache 2.0