# SKATER

Spatial C(K)luster Analysis by Tree Edge Removal (SKATER) is an optimized algorithm to prune the minimum spanning tree into several clusters that their values of selected variables are as similar as possible while retaining the contiguity structure.  For more information, please read <https://geodacenter.github.io/workbook/9c_spatial3/lab9c.html#skater>

## skater()

```sql
function skater(
    WeightResult w,
    Number k, 
    Array vals,
    Number min_bound, 
    Array bound_vals,
    String scale_method,
    String distance_type)

```

### Arguments

| Name              | Type          | Description                                                                                        |
| ----------------- | ------------- | -------------------------------------------------------------------------------------------------- |
| `weights`         | WeightsResult | The weights object `WeightsResult`                                                                 |
| `k`               | Number        | The number of clusters                                                                             |
| `values`          | Array         | The list of numeric vectors of selected variable                                                   |
| `min_bound`       | Number        | The minimum value that the sum value of bounding variable int each cluster should be greater than  |
| `bound_vals`      | Array         | The numeric vector of selected bounding variable                                                   |
| `scale_method`    | String        | The scaling method: {'raw', 'standardize', 'demean', 'mad', 'range\_standardize', 'range\_adjust'} |
| `distance_method` | String        | The distance method: {"euclidean", "manhattan"}                                                    |

### Return

| Type             | Description                                                                            |
| ---------------- | -------------------------------------------------------------------------------------- |
| ClusteringResult | The Clustering object: {'total\_ss', 'within\_ss', 'between\_ss', 'ratio', 'clusters'} |

**Examples**

{% tabs %}
{% tab title="Node.js" %}

```javascript
const jsgeoda = require('jsgeoda');
const fs = require('fs');

// load data
const data = fs.readFileSync('./data/natregimes.geojson').buffer;

// create jsgeoda instance
const geoda = await jsgeoda.New();

// load geojson in jsgeoda
const nat = geoda.read_geojson(data);

// create a queen contiguity weights
const w = geoda.queen_weights(nat);

// get values
const hr60 = geoda.get_col(nat, "HR60");
const ue60 = geoda.get_col(nat, "UE60");

// set minimum bound
const po60 = geoda.get_col(nat, "PO60");

// apply skater
const skater = geoda.skater(w, 10, [hr60, ue60], 17845200, po60);


```

{% endtab %}

{% tab title="React" %}

```javascript
import React, { Component } from "react";
import ReactDOM from "react-dom";
import DeckGL from "@deck.gl/react";
import { GeoJsonLayer } from "@deck.gl/layers";
import { StaticMap } from "react-map-gl";
import colorbrewer from "colorbrewer";
import jsgeoda from "jsgeoda";

// Set your mapbox access token here
const MAPBOX_TOKEN =
  "pk.eyJ1IjoibGl4dW45MTAiLCJhIjoiY2locXMxcWFqMDAwenQ0bTFhaTZmbnRwaiJ9.VRNeNnyb96Eo-CorkJmIqg";

// The geojson data
const DATA_URL = `https://webgeoda.github.io/data/natregimes.geojson`;

class App extends Component {
  constructor() {
    super();
    this.state = {
      mapId: "",
      layer: null,
      viewPort: {
        longitude: -100.4,
        latitude: 38.74,
        zoom: 2.5,
        maxZoom: 20
      }
    };
  }

  // load spatial data when mount this component
  loadSpatialData(geoda) {
    fetch(DATA_URL)
      .then((res) => res.arrayBuffer())
      .then((data) => {
        // load geojson in jsgeoda, an unique id (string) will be returned for further usage
        const nat = geoda.read_geojson(data);
        const w = geoda.queen_weights(nat);
        const hr60 = geoda.get_col(nat, "HR60");
        const ue60 = geoda.get_col(nat, "UE60");
        const po60 = geoda.get_col(nat, "PO60");
        const redcap = geoda.skater(w, 10, [hr60, ue60], 17845200, po60);
        //const redcap = geoda.redcap(w, 10, [hr60, ue60], "fullorder-wardlinkage", 17845200, po60);
        //const redcap = geoda.azp_tabu(w, 20, [hr60, ue60], 10, 10, 1, [], [po60],[17845200]);
        //const redcap = geoda.azp_sa(w, 20, [hr60, ue60], 0.85, 1, 1, [], [po60],[17845200]);
        //const redcap = geoda.maxp_greedy(w, [hr60, ue60],  1, [po60],[17845200]);
        const colors = colorbrewer["Paired"][10].map((c) =>
          c
            .toLowerCase()
            .match(/[0-9a-f]{2}/g)
            .map((x) => parseInt(x, 16))
        );

        // Viewport settings
        const view_port = geoda.get_viewport(
          nat,
          window.innerHeight,
          window.innerWidth
        );

        // Create GeoJsonLayer
        const layer = new GeoJsonLayer({
          id: "GeoJsonLayer",
          data: DATA_URL,
          filled: true,
          getFillColor: (f) => this.getFillColor(f, redcap.clusters, colors),
          stroked: true,
          pickable: true
        });

        // Trigger to draw map
        this.setState({
          mapId: nat,
          layer: layer,
          viewPort: view_port
        });
      });
  }

  componentDidMount() {
    // jsgeoda.New() function will create an instance from WASM
    jsgeoda.New().then((geoda) => {
      this.loadSpatialData(geoda);
    });
  }

  // Determine which color for which geometry
  getFillColor(f, clusters, colors) {
    const i = f.properties.POLY_ID - 1;
    const c = clusters[i] - 1;
    return colors[c];
  }

  render() {
    return (
      <div>
        <DeckGL
          initialViewState={this.state.viewPort}
          layers={[this.state.layer]}
          controller={true}
          getTooltip={({ object }) =>
            object && `${object.properties.NAME}: ${object.properties.HR60}`
          }
        >
          <StaticMap mapboxApiAccessToken={MAPBOX_TOKEN} />
        </DeckGL>
      </div>
    );
  }
}

ReactDOM.render(<App />, document.getElementById("root"));

```

{% endtab %}
{% endtabs %}

**Try it yourself in the playground (jsgeoda + deck.gl):**

{% embed url="<https://codesandbox.io/s/7spatialclustering-uvz12>" %}


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://xunli.gitbook.io/jsgeoda/spatial-clustering/skater.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.