Boneyard Tools

AI Sentiment Analyzer

Find out whether a piece of text reads as positive or negative. Paste a product review, customer comment, or feedback message and a DistilBERT AI model classifies the sentiment, then breaks it down sentence by sentence so you can see exactly where the mood shifts. The model runs entirely in your browser, so nothing is uploaded. It downloads once on first use, then is cached.

How to analyze sentiment

  1. Paste the review, comment, or feedback you want to check into the box.
  2. Click Analyze and wait a moment while the model loads on first use.
  3. Read the overall positive or negative verdict, then the per-sentence breakdown.

Examples

Mixed product review

The battery life is amazing and lasts all day. Unfortunately the camera is disappointing in low light.
Overall: Mixed. Sentence 1: Positive (99%). Sentence 2: Negative (98%).

Frequently asked questions

Is my text uploaded anywhere?

No. Nothing is uploaded. The DistilBERT model runs entirely in your browser via WebAssembly, so your text is analyzed on your device and never sent to a server. Only the model itself is downloaded, once, then cached.

Which AI model does this use?

It uses DistilBERT fine-tuned on SST-2 (the Stanford Sentiment Treebank), a fast and widely used sentiment model. It runs locally in your browser through transformers.js and ONNX, with no API calls.

Why does the first run take a moment?

The AI model downloads the first time you use the tool, then is cached by your browser for instant reuse. After that, analysis is nearly instant for short text.

Does it detect emotions like anger or joy?

No. This model is binary: it tells you whether text is positive or negative, with a confidence score. It does not classify fine-grained emotions such as anger, joy, or sadness. For nuanced text, read the per-sentence breakdown rather than the single overall label.

Can it handle long text?

Yes. Longer text is split into individual sentences and each one is scored separately, then combined into an overall verdict. This also lets you see which specific sentences are positive and which are negative.

Related tools