bert-base-uncased-tweet-sentiment-analysis

BERT Tweet Sentiment Analysis

Model Details

Model Description

This model is a BERT-based transformer (bert-base-uncased) fine-tuned for sentiment analysis on tweets. It classifies text into six emotion categories: sadness, joy, love, anger, fear, and surprise. The model was trained with a custom classification head on tweet data to capture short-form, social media style text.

• Developed by: Bibhu Mishra
• Model type: Sequence Classification
• Language(s) (NLP): English
• License: MIT
• Finetuned from model: bert-base-uncased

Model Sources

• Repository: Hugging Face Model Hub

Uses

Direct Use

This model can be used to classify tweets or short text into emotions, useful for social listening, customer sentiment analysis, and marketing insights.

Downstream Use

The model can be integrated into larger NLP pipelines, such as real-time social media monitoring dashboards, chatbots, or recommendation systems that require sentiment-aware behavior.

Out-of-Scope Use

• Not suitable for clinical, medical, or diagnostic use.
• Performance may decrease on very long text, sarcasm, or highly domain-specific language.

Bias, Risks, and Limitations

• Limited to English-language social media text.
• May misclassify sarcastic or context-heavy tweets.
• Emotions are simplified into six categories, which may not capture nuanced sentiment.

Recommendations

Use this model for exploratory sentiment analytics and not for critical decisions. Always review predictions in the context of business requirements.

How to Get Started with the Model

from transformers import AutoTokenizer, AutoModelForSequenceClassification, pipeline

model_name = "mishrabp/bert-base-uncased-tweet-sentiment-analysis"

tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForSequenceClassification.from_pretrained(model_name)

classifier = pipeline("text-classification", model=model, tokenizer=tokenizer)

tweets = [
    "I love spending time with my friends!",
    "I feel so sad about the news today."
]

results = classifier(tweets)
print(results)

Output

[
    {"label": "love", "score": 0.95},
    {"label": "sadness", "score": 0.88}
]

Labels

| ID | Emotion  |
| -- | -------- |
| 0  | sadness  |
| 1  | joy      |
| 2  | love     |
| 3  | anger    |
| 4  | fear     |
| 5  | surprise |

Inference URL

https://mishrabp-twitter-sentiment-analysis.hf.space/