How is Sentiment Assigned in Meltwater?
Whether you’ve only spent a few minutes in Meltwater or it’s the most significant part of your day, chances are you’ve seen Sentiment - in Explore, Monitor, Analyze you name it.
But how is Sentiment assigned in Meltwater?
Meltwater uses a mixture of Sentiment Analysis and Natural Language Processing (also known as NLP - a branch of artificial intelligence) to identify and assign sentiment. Both of these are used in tandem to understand public opinion and the overall emotional tone of the text related to your brand, products, or competitors. Based on the analysis of a hit or clip in Meltwater, it will automatically be categorized into one of these three buckets - Positive, Negative, or Neutral.
Positive: This article exclaims love for different running shoes based on the type of run.
Negative: This hit tells a story of someone expressing a negative emotion towards a running shoe brand due to the quality of the shoe.
Neutral: This article simply states a fact that contains no specific implied emotion.
NLP and Sentiment Analysis are by no means perfect. For instance, slang and sarcasm can alter how you perceive sentiment versus someone else. Two people might even look at the same message and have differing opinions or sentiment.
This is one of many instances in which you might want to override the given sentiment of an article. Follow these steps to do so:
- Open a content stream in Explore or Monitor
- Click on the current Sentiment in the bottom right-hand corner of the hit
- Make your selection of Positive, Neutral, Negative, or Not Rated.
Note: If the document has duplicate articles, a pop-up window asking Apply to All Documents? will appear. You will be able to apply the sentiment to all duplicates or to just one selected result.
- Once you have assigned a new sentiment rating, the sentiment icon will no longer be hollow.
Learn more about Assigning Sentiment.