Mixed HASOC 2020

Mixed HASOC 2020 research focuses on developing and improving methods for analyzing datasets containing mixed data types or scenarios, such as combining real and synthetic data for training, or handling both continuous and categorical variables in generative models. Current efforts leverage deep learning architectures, including variations of UNet and EfficientNet, along with multilingual BERT models, to address challenges in areas like image segmentation, grammatical error correction, and offensive language detection in code-mixed text. These advancements have implications for various fields, improving efficiency in tasks ranging from waste management and airport pavement inspection to drug discovery and mental training applications.