Attitude Analysis Model

Model Description

This model is designed for Attitude Analysis of text, with a focus on three key sub-systems from the Appraisal Framework:

1. Affect: Analyzes the emotional response in the text, such as feelings of happiness, sadness, or fear.
2. Judgment: Analyzes moral or ethical evaluations, such as praise or condemnation.
3. Appreciation: Evaluates the aesthetic or value-based judgments, such as whether something is considered beautiful or perfect.

The model is fine-tuned on news articles to analyze how media outlets express different attitudes towards events, specifically focusing on political and social contexts.

Model Training

• Training Dataset: The model is trained on a dataset of news articles covering political events, especially focusing on Trump's election victory.
• Base Model: The model is based on the RoBERTa architecture, which has been fine-tuned for sentiment and attitude analysis.

Intended Use

This model can be used for:

• Affect Analysis: Classifying text as positive or negative in terms of emotions.
• Judgment Analysis: Determining ethical evaluations (e.g., praise or criticism).
• Appreciation Analysis: Evaluating the value or aesthetic judgment in the text.

Example Usage

from transformers import pipeline

# Load the pre-trained model
classifier = pipeline("text-classification", model="your-username/attitude-model")

# Example text
texts = [
    "Trump's victory marks a historic moment in American politics.",
    "Many people are worried about the future under Trump's leadership."
]

# Get predictions
results = classifier(texts)

for result in results:
    print(result)