Where does this data come from?
For every article on this site we pass the content through IBM's "Tone Analyzer" which returns likely emotional response to the text across five key emotional dimensions: Anger, Disgust, Fear, Joy and Sadness. For each emotional category a score between 0 and 1 is provided. A score of one would be a sure emotional response and score over .5 indicates a likely emotional response. A higher score does not necessarily indicate a greater intensity of emotion, but rather a greater likelihood that that particular emotion will be evoked.
The table below is from IBM's description of the Tone Analysis tool.
|joy||Joy or happiness has shades of enjoyment, satisfaction and pleasure. There is a sense of well-being, inner peace, love, safety and contentment.||Less than 0.5 - less likely to be perceived as joyful.||More than 0.75 - Highly likely to be perceived as joyful.|
|fear||A response to impending danger. It is a survival mechanism that is a reaction to some negative stimulus. It may be a mild caution or an extreme phobia.||Less than 0.5 - less likely to be perceived as scared.||More than 0.75 - Highly likely to be perceived as scared.|
|sadness||Indicates a feeling of loss and disadvantage. When a person can be observed to be quiet, less energetic and withdrawn, it may be inferred that sadness exists.||Less than 0.5 - less likely to be perceived as sad.||More than 0.75 - Highly likely to be perceived as sad.|
|disgust||An emotional response of revulsion to something considered offensive or unpleasant. It is a sensation that refers to something revolting.||Less than 0.5 - less likely to be perceived as disgusted.||More than 0.75 - Highly likely to be perceived as disgusted.|
|anger||Evoked due to injustice, conflict, humiliation, negligence or betrayal. If anger is active, the individual attacks the target, verbally or physically. If anger is passive, the person silently sulks and feels tension and hostility.||Less than 0.5 - less likely to be perceived as angry.||More than 0.75 - Highly likely to be perceived as angry.|
Why these emotions?
The emotional categories used are generally accepted by many in the psychological community. In fact, the Disney movie "Inside Out" which explored the emotional world of a young girl, used these categories as its character base. This in no way means these categories are beyond scrutiny but they are not necessary controversial either.
While the Tone Analysis tool provided by IBM is the most accurate human language analysis tool available by a substantial margin, it should in no way be considered "accurate" in our normal usage of the word. For this reason, individual article scores should be considered less likely to be accurate than aggregate data, such as charts and trends.
Why So Angry?
I am sure one of the first things you noticed is how consistently "anger" is the dominant emotion. I am not certain yet why this is the case. It could be the Tone Analyzer is accurate and people are seriously pissed off about politics or that media companies want "angry" in their content. However, it could be that the generally accepted op-ed writing style is simply angrier than everyday communication style. Either way, I hope that as we move forward we will be better able to answer this question.
How to you choose the articles?
Due to the expense of consuming the Tone Analysis and Alchemy API ( which provides keyword sentiments ) I am unable to include as much content as I would like. I have hand chosen several data sources that I feel provide a reasonable sample of opinion across the political spectrum. This sample is not scientific and no effort has been made to establish a scientific sample. It is my hope that through a small amount of success we can fund further expansions of content.
What happens when the API updates?
For the same reason I cannot afford more content, I will not be able to retroactively reanalyze every article in the event that IBM updates their Tone Analyzer. This maybe possible in the future.