Runs sentiment and emotion classification using 🤗 transformers models on about 25k tweets mentioning @Dell. Explores the sentiments and emotions to figure out the positives and negatives of the company's product and services.
- The sentiment classifier labels the tweets as either positive, negative and neutral
- The emotion classifier classifies the tweets into 11 emotions: joy, love, optimism, pessimism, trust, surprise, anticipation, sadness, anger, disgust and fear.
Uses a transformer model (cardiffnlp/twitter-roberta-base-sentiment-latest) to classify sentiment and goes a step deeper and classify emotions of the tweets using another transformer model (cardiffnlp/twitter-roberta-base-emotion-multilabel-latest).