Exploring the Possibilities of Emotional Text Analysis with Deep Learning Technologies

Authors:
Publisher website
Google Scholar
Abstract
This manuscript explores the capability of deep learning technology within the software of emotional text analysis (ETA). In particular, we consciousness of using lengthy quick-time period reminiscence (LSTM), convolutional neural networks (CNN), and different deep mastering fashions in identifying emotional content in text. We review the present literature concerning ETA, presenting an in-depth assessment of fundamentals, which includes sentiment evaluation, factor-based total sentiment analysis, emotion recognition, and multi-characteristic sentiment evaluation. We also speak about current datasets available for schooling, check out deep getting-to-know fashions, and provide implementation details. Subsequently, we perceive demanding situations inside the application of deep mastering fashions for ETA, which includes idea flow and shortage of guide annotation. In the end, we offer directions for future research on deeper evaluation of the emotional content material of textual content and its programs.

This publication has 18 references indexed in Scilit: