Your sentiment is based on the number of positive, neutral and negative reviews that you receive.
This score is determined by an algorithm that analyses keywords in each review that indicate whether or not it has a positive, neutral or negative sentiment.
If a traveller gave a review (see image below for an example) and they used words such as “excellent” and “lovely” the review would be seen as having a positive sentiment. If, however, the guest used words such as “poor” and “slow” in their review it will be seen as having a negative sentiment.
Sometimes guests have good and bad points to share. When this happens, the Sentiment Analysis algorithm will try to find an average sentiment between the positive and negative keywords used.
Positive rating and negative sentiment
“Lovely stay but service can be slow.”
It is entirely possible that you receive a review where the rating is high, but the sentiment is seen as negative. Remember, Sentiment Analysis does NOT reflect your rating, just the sentiment of the review based on keywords used.
The double negative
“When I'm in London, I will never not stay here!”
There may be times when a good review is picked up as having a negative sentiment. Because of the sheer complexity of language and the fact that sentiment is a score produced by an algorithm (which can’t think for itself) there can be a margin of error present in any Sentiment Analysis score.
It is important to remember that despite the slight margin of error due to an algorithm's imperfect understanding of complex semantics, when looking at reviews in volume, Sentiment Analysis is seen as a useful way to track overall guest sentiment on different review sites and Online Travel Agencies across the web.
If you ever come across a case where the sentiment of the review has been incorrectly categorised, please let our support team know by dropping an email to email@example.com.