Emotions

Where sentiment measures the direction of opinion, the Emotions tab measures its character: which feelings the conversation runs on. The distinction matters most when sentiment is negative, since anger, fear, and disappointment at the same score describe different problems and tend to call for different responses. Posts are labeled with the emotions they express, and a post can express several at once, so counts here measure how widespread each feeling is rather than splitting posts into exclusive groups; sums can therefore exceed the number of posts. Classification draws on eight categories, covered in depth in the emotion classification guide.

The page adapts to the number of queries in the query bar. With a single query it shows the layout described in the next section. When two or more queries are active, a toggle appears above the results with three views: Compare sets the queries against each other, Aggregate combines them into one conversation, and Select focuses on one of them at a time. For the query bar itself, including syntax, time ranges, and display names, see Getting Started.

Single query (Select)

The layout for one query, and what the Select view renders per query.

Stats bar

Top Emotion is the most frequently detected emotion, with its share of all emotions detected. The share tells you whether that emotion defines the conversation or merely edges out the rest: a 40% share marks a conversation with one dominant feeling, while a 15% share marks an even spread where the leader barely leads. The Emotions volume chart below shows the full field.

Runner-Up Emotion is the second most detected emotion, with its share. A close runner-up means two feelings compete for the conversation's character, and the pairing itself is descriptive: Joy with Anticipation reads as excitement around something upcoming, while Anger with Disgust tends to mark moral outrage rather than mere frustration.

Emotion Diversity counts how many emotions hold a meaningful share of detections, at least 10%: Concentrated when one does, Moderate for two or three, Diverse for four or more. Concentrated conversations are usually reacting to one thing in one way, so a small sample of posts explains them. Diverse conversations contain several audiences feeling differently, and an average reading flattens them; the feed's emotion badges are the quickest way to see the separate camps.

Highest Concentration is the completed interval where a single emotion reached its highest share, with the date, the share, and the emotion. It is the conversation's most emotionally uniform moment, which usually marks the day one story set the tone. Narrow the time range to it and the feed shows the story; the over-time chart below shows whether it was a one-day flash or the start of a phase.

Emotions volume

One bar per emotion shows how many posts express it, and pointing at a bar shows the exact count. The muted track behind the bars represents all matching posts, so each bar reads as that emotion's reach within the conversation. Watch how large Neutral runs: a dominant Neutral bar means most of the conversation carries no detectable feeling, and the emotional analysis describes only the remainder.

Emotions distribution

The pie chart shows each emotion's share of all detections, the conversation's emotional mix at a glance. It is also the quickest place to weigh the positive cluster (Love, Joy, Anticipation) against the negative one (Anger, Disgust, Sadness, Fear) as blocks, an emotional balance that complements the sentiment score rather than repeating it, since it is computed from expressed feelings rather than stances.

Emotions over time

Stacked bars track the emotional composition across the range, in two views:

• Volume stacks absolute counts per interval, tying feeling to activity: when the conversation surges, the color of the growth shows which emotion arrived with it, which is often the fastest characterization of what the triggering event meant to people.
• Distribution normalizes each interval to 100%, isolating mood shift from size. A negative emotion's band steadily widening inside a stable-sized conversation is an early warning that may precede a sentiment decline, and is worth cross-checking against the Sentiment tab over the same days.

Aggregate

Aggregate renders the emotional profile of the combined conversation, the right reading when the queries are phrasings of one subject and the question is its overall mood; combining also steadies the shares when individual variants match few posts.

Compare

The Compare view contrasts the emotional character of the conversations, each in its query bar color.

Stats bar

Top Emotion names the strongest emotional concentration anywhere in the comparison: the emotion with the largest share within any single query, with that share and the query it belongs to. A standout concentration, such as one subject running 40% Anger while nothing else comes close, is usually the first thing in the comparison worth investigating.

Runner-Up Emotion is the second-ranked emotion by the same measure, with its query. Whether it sits in the same query as the top cell or a different one tells you if one conversation carries the comparison's emotional extremes or if each subject has its own.

Highest Variation names the emotion whose share differs most between queries, with the gap. This is the feeling on which the conversations most disagree, and therefore the most informative segment of the Emotions chart below; it is also the natural choice for the over-time selector when asking when the difference emerged.

Most Diverse names the query with the most emotions above the 10% threshold, with the count. The most emotionally fragmented conversation is the one to be most careful with: its averages summarize several different audiences, and switching to its Select view to read the feed shows what those camps are.

Emotions

One bar per query, stacked to 100% and segmented by emotion category. Normalizing the bars makes emotional composition comparable even when volumes differ widely. Look for the segment that differs most between bars: two subjects with similar sentiment scores can run on entirely different feelings, such as one fueled by Anticipation and the other by Anger, and that difference often says more about the two conversations than the scores do. The Highest Variation cell names the segment to look at first; to read the posts behind a standout segment, switch to that query in the Select view.

Emotions over time

This chart inverts the single-query version: instead of all emotions for one conversation, it tracks one emotion across all of them. A selector chooses the emotion, starting on the most prevalent one in the results, and the queries stack within each interval. Volume shows how much of that feeling each conversation produces over time; Distribution shows each query's share of it per interval, making it visible when a feeling migrates from one subject to the other. Use it to put a timeline behind the stats bar: pick the Highest Variation emotion and see whether the gap is constant or appeared around a specific event, then narrow the time range there and read the feed.