Unlocking Bias Detection: Leveraging Transformer-Based Models for Content Analysis