Step 4 — Filtering with Facets¶
Narrowing the set¶
The facet sidebar is how you sculpt the set into something more specific. Each filter group is a dimension you can use to include or exclude entities. Let us try a few combinations using the six researchers.
Adding a first filter¶
Click SE in the Nationality facet group.
The set narrows immediately to two researchers: Julia Lindström and Anders Bergström — the two Swedes. The count next to Professor updates to 2; PhDStudent dims out entirely (greyed out), because there are no Swedish PhD students in this dataset. The context header updates to: "SE Researchers".
Notice that Moire did not make you wait for the filter to load — the set updated as soon as you clicked. Also notice the greyed-out PhDStudent value: this is Moire being honest with you. It is saying "if you click PhDStudent right now, with Nationality = SE already active, you will get zero results." The greyed-out value is not disabled forever — remove the nationality filter and it becomes available again.
Adding a second filter¶
With Nationality = SE still active, click Professor in the Type facet.
The set narrows further to exactly two Swedish Professors: Julia Lindström and Anders Bergström. Both filters are now active simultaneously — AND logic across facet groups. The context header now reads: "SE Professors". (Moire promotes the type label — Professors — over the more generic Researcher when a type filter is active.)
The empty state¶
Let us push the filters until they produce nothing. With SE and Professor active, click female in the Gender facet.
Moire shows an empty state: "0 SE female Professors." But it does not leave you stranded. The empty state page tells you exactly which filter is causing the problem and offers specific, one-click suggestions:
- "Remove 'female' → would show 2 results"
- "Remove 'SE' → would show 1 result (Julia Lindström)"
This is Moire's honesty principle in action. It never silently produces an empty result — it explains why the combination is impossible and shows you the fastest path back to something useful.
Clearing filters¶
Click Clear all filters. All three active filters are removed simultaneously, and the full set of six researchers reappears. The context header returns to: "Researchers".
You can also remove filters one at a time: click the × button next to any active filter chip to remove just that one, leaving the others in place.
How filters combine¶
The logic behind facet filtering is intuitive once you see it in action:
- Within a facet group (e.g. ticking two nationalities), Moire uses OR — show entities that match any of these values. Tick both SE and NO, and you get the two Swedish researchers plus the one Norwegian researcher: three in total.
- Across facet groups (e.g. a nationality filter and a type filter active simultaneously), Moire uses AND — show entities that match all active groups. SE AND Professor gives you only the entities that are both Swedish and Professors.
This combination of OR-within and AND-across is the standard behaviour for faceted search, and it mirrors the way you naturally think about filtering: "Show me Norwegian or Swedish researchers (within nationality) who are Professors (separate condition)."
The live counts¶
The count next to each facet value updates constantly as you add and remove filters. It always shows you how many entities in the current set have that value — not how many exist in the whole graph. This is important: the counts reflect the reality of your current exploration, not an abstract total. If you are looking at SE Professors and the count next to Knowledge Graphs is 1, that means one of the two Swedish Professors works in Knowledge Graphs — a meaningful, contextual number.
What to do¶
Clear all filters to return to all six researchers. Then continue to Step 5 to see what happens when you move the entire set through a relationship.