Deep Learning-based Topic-level Examination of Social Media

Ramachandran K.K.; Perez-Mendoza A.; Joshi K.; Parikh S.M.; Saini D.; Gonzales J.L.A. · Proceedings of 5th International Conference on Contemporary Computing and Informatics, IC3I 2022 (2022) · DOI: 10.1109/ic3i56241.2022.10073324

Resumen

Due to its superior processing power in fields like text, picture, and audio processing, deep learning (DL) is becoming more and more popular. The exponential growth and extensive accessibility of digital social media (SM) make the evaluation of this data challenging, if not impossible, using conventional methodologies and technology. It has been determined that DL is a suitable answer to this issue. After the enormous work put into SM analytics, we provide a taxonomy-focused assessment of the DL architectures employed in this research (SMA). However, rather than focusing on a technical explanation, this work aims to explain the SMA-related problems utilizing DL-based remedies. We also discuss the potential and difficulties for DL research in the future, including scalability, heterogeneity, and multimodality.

Instituciones: Universidad Tecnológica del Perú