Text Classification into Emotional States Using Deep Learning based BERT Technique

Classifying text into states of emotions is the most challenging task in processing natural languages, and it is even harder due to the absence of facial expressions in the sentences. Therefore, it is needed to determine what kind of emotions the texts or words communicate. Categorization of texts into sentimental text is extremely helpful and could be utilised across a variety of fields that include information mining, psychology, human-computer interaction, and many more. This research paper uses the BERT transformer from the hugging face library to classify and generate the results. The dataset comes from reviews collected from the google play store and is used to train the model; the model can be used to conclude which apps are better among the selected apps. Apart from this, the model may help developers to read the massive number of reviews written for a particular app and get new features or bugs written in the reviews.