Emotion Recognition on Twitter: Comparative Study and Training a Unison Model

Abstract

Emotion Recognition on Twitter: Comparative Study and Training a Unison Model java project report Despite recent successes of deep learning in many fields of natural language processing, previous studies of emotion recognition on Twitter mainly focused on the use of lexicons and simple classifiers on bag-of-words models. The central question of our study is whether we can improve their performance using deep learning. To this end, we exploit hashtags to create three large emotionlabeled data sets corresponding to different classifications of emotions. We then compare the performance of several wordand character-based recurrent and convolutional neural networks with the performance on bag-of-words and latent semantic indexing models.

System Configuration:

H/W System Configuration:-

System          : Pentium I3 Processor.
Hard Disk       : 500 GB.
Monitor          : Standard LED Monitor
Input Devices : Keyboard
Ram               : 4 GB

S/W System Configuration:-

Operating system              : Windows 7/8/10.
Available Coding Language : Java and Phonegap
Database                          : MYSQL

Conclusion

Emotion Recognition on Twitter: Comparative Study and Training a Unison Model java project report We proposed an enriched graph-based feature extraction mechanism to extract emotion-rich representations. The patterns are enriched with word embeddings and are used to train several effective emotion recognition models.

Our patterns capture implicit emotional expressions which improves emotion recognition results and helps with interpretability. We demonstrate a basic application of the proposed affective lexicon on a gender dataset. We hope to improve the pattern weighing mechanism so as to improve the performance on emotion recognition tasks and minimize trade-off between pattern coverage and performance.