Gangeshwar Krishnamurthy

Focuses on detecting propaganda techniques in news articles by analyzing emotional salience features at the sentence level.

Proposes a variational autoencoder-based multimodal fusion approach that maintains the fidelity of multimodal representation with respect to unimodal representations.

Proposes a multi-task learning framework to detect the stance of online posts using auxiliary tasks to augment models.

Proposes a multimodal neural model for deception detection, combining features from video, audio, text, and micro-expression analysis.

Explores a deep learning approach for aspect-based sentiment analysis using inter-aspect dependencies.

Investigates the co-occurrence of semantic roles in verb representations, suggesting hidden restrictions on role combinations.