Where the Mood figure comes from

The positive, neutral and negative split across every discussion item that was scored, published as a split and as a single figure saying how that balance compares with its own usual level.

How it is put together

The discussion that may be read
Public discussion from the platforms whose terms permit reading them, which today is one. What each source is, what it permits, and whether it is running at all is published rather than described here.
The source list
If this produces nothingA platform that cannot be read contributes nothing and is never estimated from the one that can. The figure names the platform it came from, so a single-platform reading is visible as such.
Which items are about a subject
Each item is matched against the same list of subjects the attention figure uses, and an item that matches nothing is counted and set aside.
How sure a match has to be, and what becomes of an item that falls short, is published with the tone method. How the match itself is made is not written down here, for the attention figure or for this one.
If this produces nothingAn item nothing recognises contributes to no subject at all. It is never filed under the nearest match, because a wrong attribution is worse than a missing one.
Reading each item
Items are cleaned, near-duplicates and spam-shaped items are set aside, the language is identified, and what is left is scored as positive, neutral or negative by a model whose identity is stored with the figure.
In full: Reading one item
If this produces nothingAn item in a language the classifier cannot read is excluded and counted, never scored as neutral. Neutral is a reading, and not being able to read something is not one.
This figure
Items are pooled within a platform and the platforms are then combined equally, so a busier platform cannot decide the result by volume alone. What is published is the positive, neutral and negative split, together with a single figure saying how far that balance sits from what it has usually been.
In full: Sentiment — the balance of the discussion that could be read
If this produces nothingA period in which nothing could be scored produces no figure, never a neutral one. The split is the reading; the single figure is a summary of it.

Sentiment — the balance of the discussion that could be read

Items are pooled within each platform and the platforms are then combined equally. Two comments on one platform are two of the same thing and add up; a comment on one platform and a post on another are not, and letting the busier platform's item count decide the overall figure is exactly what this arrangement avoids.

The overall balance of the discussion is the quantity, and the published figure is how far that balance sits from what it has usually been over the past two weeks. The middle of the scale means *the discussion leans about as it normally does*, not *the discussion is balanced* — a conversation that is steadily grim reads in the middle, and the split beside it is what says so.

The balance itself is published as well, so nothing about the split is recoverable only through the figure. That is what the plain rescaling this replaced was for; what it could not do was move. A mean over thousands of items has almost no room to vary, and the published figure spent a whole week inside 1.6 points of its hundred — which is arithmetic rather than a quiet world.

Items whose language the classifier cannot read are excluded rather than counted as neutral, and the figure describes the discussion that could be scored.

What this does not say

**This is one online community's discussion, not a public.** One discussion platform is available to this product today; the other is not reachable under its operator's current terms. The people who post there are not representative of anyone, and the figure names the platform it came from.
It measures the tone of what was written, not what anybody believes, and not how many people hold a view. A loud minority and a quiet majority are indistinguishable from here.
The classifier is a model with its own errors, and its identity is stored with every figure so that a change of model is visible rather than looking like a change in the discussion.
The single figure is a summary of the split *and* a comparison with the past. Where the actual balance matters — and it usually does — the split is what to read.
Because it is a comparison, a persistently negative discussion reads as ordinary. This figure describes whether the tone has shifted, never how dark it is in absolute terms.

Numbers this uses 2

Constants used by Sentiment — the balance of the discussion that could be read, with their units and how each was arrived at
Value	What it does	How it was arrived at
50 index points `sentiment.neutral`	Where a discussion that is neither positive nor negative overall lands. Above it the balance is positive, below it negative.	Fixed by this product's own specification.
50 index points `sentiment.polarityScale`	How far the ends of the underlying scale reach from the middle. It makes the published figure a plain rescaling of the split beside it, so the two can never disagree.	Fixed by this product's own specification.

What this product does not claim

Stated positions about what this figure is, not a list of things not built yet.

This describes the topics and questions this product tracks, not the world

Every figure here is an average over a chosen set: a few hundred topics with names somebody wrote down, and the forecast questions that passed an admission policy. Nothing in the arithmetic knows what it is missing.

The whole position

Every threshold on this page is stated, not measured

All the numbers here are stated defaults: reasoned choices with an argument behind each and no observed distribution to place them against. That is stronger than it sounds — it means none of them was fitted to anything, not even second-hand.