Sentiment Checker

What sentiment analysis does

Sentiment analysis estimates the emotional tone of text. It is widely used to summarise customer reviews, programme feedback, social media posts and open-ended survey answers, so that teams can see at a glance whether opinions lean positive or negative and which comments need attention. It is a quick first look; for a full study, see our sentiment analysis service.

How this tool works

The checker uses a dictionary of positive and negative words, each with a strength from 1 to 3. It scores every sentence by adding the values of the words it contains. It also looks back three words for negations such as not or never, which weaken or reverse a score, and for intensifiers such as very or extremely, which strengthen it. Sentences with a score above 0.5 are labelled positive, below −0.5 negative, and the rest neutral.

Limitations

• It is designed for English text.
• It cannot detect sarcasm, irony or cultural references.
• Words that are negative in everyday life may be neutral in your topic. For example, a health report discusses disease without sounding negative.
• Short comments give less reliable results than long ones.

For published research or high-stakes decisions, validate the scores against a manual coding sample, or ask our team to run a machine learning based analysis.

Practical uses

Use the tool to triage feedback, to compare the tone of two sets of comments, or to find the most negative sentences to read first. Combine it with the word frequency counter to see which words drive the negative scores.

Related analysis services

Our sentiment analysis service goes beyond a word list, with text classification, social media analytics and Python and R text analysis. For survey comments, see survey data analysis. See all services or tell us about your project.