Docs

Quick start with the Adjacency Matrix

In this first quickstart, we'll walk through another example using a slightly more advanced format: the Adjacency Matrix.

Step 1 To start, download the Adjacency Matrix example. Then, go to the Dashboard, click on create New Map and select Matrix Format.

The Adjacency Matrix format lists each entity in each sheet. Let's take a closer look at each sheet, starting from the first sheet:

idname
tpg Top Grossing Movies

This first row describes an entity whose id is TPG and whose name is Top Grossing Movies. These two properties are required. In the Matrix format, the type of the entity is given in the sheet name; there are two parts in the name of the sheet group_uber group — the first part before the dash tells Rhumbl that it is a group, and the second part after the dash is the name of the group.

Do not change the name of the Uber Group sheet.

The second sheet is called group_origin and it looks like:

idnamegroup id
domes Domestic tpg
intl International tpg

The first row says that there's an entity (of type origin), whose name is Domestic, and it sits in the tpg group. Similarly, the second row says there's an entity called International which also sits in the tpg group.

Click to the tab node_movie:

idgroup idshort name rotten tomato score
Inception scifi Inception 97

This first row says that there is an entity of type movie called Inception. It belongs to the group scifi — note that the group id must exactly match the id of the group where you first defined it (in the group_genre tab).

We've now gone through how to define entities in this format. We'll look at how we define relationships in the next section.

Step 2 Next, upload your data as prompted. You will see the following parsed result:

Note that we have three types of relationships. The first two relationships are defined in the two tabs, rel_d_has prequel of and rel_d_acts in. In the Matrix format, edge relationships (relationships that show up as edges) are defined in sheets that start with rel_. Let's look at the rel_d_has prequel of tab:

Source ID \ Target IDInceptionStar Wars: Episode VI - Return of the Jedi Star Wars: The Force Awakens
Inception
Star Wars: Episode VI - Return of the Jedi 1
Star Wars: The Force Awakens

In this matrix form, we see a 1 in the 2nd row, 3rd column. This indicates that Episode VI has a prequel of The Force Awakens. The directionality is important here:

  1. The presence of the d in the tab name says that it is a directed relationship: rel_d_.
  2. The direction of the relationship always point from the entity in the row to the entity in the column.

What about the third relationship, HAS_PARENT_OF? Those are parent-child relationships created from the group_id columns in each tab describing an entity. For example, when we wrote:

idgroup idshort name rotten tomato score
Inception scifi Inception 97

We created a parent relationship of type HAS_PARENT_OF from the entity Inception pointing to the entity scifi.

There it is! You have now learned how to create a map using the matrix format. Jump ahead to the chapter below on how to customize your map.