Opportunities and Challenges of Deep Learning Models for Short Text Data Analysis

The rapid growth of the internet usage on mobile devices has increased the opportunities to communicate with other people and publicly share news and opinions. As a result, a huge amount of data is generated every day, and it becomes a great resource for investigating different research problems such as sentiment analysis, entity finding, and prediction problems. Recent advanced machine learning methods use deep neural network models to solve the problems that require huge datasets to train and converge the models. However, people mostly use short text data for such online communication purposes. Analyzing the sentiment of the short text is challenging because of its natural characters, such as sparseness, immediacy, and misspelling. Therefore, this study focuses on the opportunities and challenges of deep learning models for short text data analysis. The paper discussed research works that applied deep learning (deep neural network) models to analyze short text data and explained unsolved problems in this area.

Keywords:Short text data; natural language processing; NLP; deep neural network models; social media data; Twitter data