Skip to content
Free public beta until December 4, 2026: every course is open, no card needed. After that, the first chapter of each course stays free and full access is a one-time $39 ($19 for up to 100 founding members chosen for helpful feedback). Your progress is kept. How the beta works

Recipe

Compute a vertex's eccentricity

Use when: you're given a connected graph and one specific vertex v, and asked for ecc(v)

Goal: the single largest distance from v to any other vertex in the graph

Taught in Eccentricity

  1. 1

    Make sure the graph, or at least v's own piece of it, is fully connected before computing anything.

    Example

    The two-triangles graph (edges AB, AC, BC, CD, CE, DE) is a single connected piece, so every distance from A is finite.

    • the graph is connected — every distance from v is guaranteed finite, continue to Step 2
    • the graph is disconnected — restrict the whole computation to v's own component before continuing

    Why: an infinite distance to some other component can't be the "greatest finite distance" eccentricity is meant to report, so reachability has to be settled first.

  2. 2

    For each other vertex u, find d(v,u) using Section 4's own distance procedure: a direct edge gives 1, otherwise search increasing path lengths.

    Example

    From A on the two-triangles graph: d(A,B)=1 (direct edge), d(A,C)=1 (direct edge), d(A,D)=2 (via A,C,D), d(A,E)=2 (via A,C,E).

    Why: eccentricity is defined as a maximum over every other vertex, so every one of those distances has to be found before the largest can be picked out.

  3. 3

    Compare every distance found in Step 2 and keep the largest one; a tie for largest is fine, only the value itself matters.

    Example

    The four distances from A are 1, 1, 2, 2; the largest is 2, so \operatorname{ecc}(A) = 2.

    ⚡ Shortcut

    On a path graph P_n, skip listing every distance individually. An end vertex always has eccentricity n-1, the full length of the path. A vertex at position i (counting from either end) has eccentricity \max(i-1, n-i), and the middle vertex minimizes that at \lceil (n-1)/2 \rceil. For P_5: an end vertex has eccentricity 4, and the middle vertex C has eccentricity \lceil 4/2 \rceil = 2, matching a full distance list without ever writing one out.

    Why: eccentricity is exactly \max_u d(v,u) by definition, not a count of vertices or edges touching v.

Common mistakes

  • Stopping after finding one large distance instead of checking every other vertex for an even larger one.
  • Computing eccentricity on a disconnected graph without first restricting to the one vertex's own component.
  • Confusing eccentricity (a single number, the worst case for one vertex) with the full list of distances it was computed from.
  • Applying the path-graph shortcut to a graph that only looks like a path in a drawing but has an extra edge creating a shortcut of its own.