270 lines
6.6 KiB
JavaScript
270 lines
6.6 KiB
JavaScript
|
|
/**
|
|
* @requires ./PriorityQueue.js
|
|
*/
|
|
|
|
if (typeof module !== 'undefined' && module.exports) {
|
|
var PriorityQueue = require('./PriorityQueue.js');
|
|
}
|
|
|
|
/**
|
|
* OPTICS - Ordering points to identify the clustering structure
|
|
*
|
|
* @author Lukasz Krawczyk <contact@lukaszkrawczyk.eu>
|
|
* @copyright MIT
|
|
*/
|
|
|
|
/**
|
|
* OPTICS class constructor
|
|
* @constructor
|
|
*
|
|
* @param {Array} dataset
|
|
* @param {number} epsilon
|
|
* @param {number} minPts
|
|
* @param {function} distanceFunction
|
|
* @returns {OPTICS}
|
|
*/
|
|
function OPTICS(dataset, epsilon, minPts, distanceFunction) {
|
|
/** @type {number} */
|
|
this.epsilon = 1;
|
|
/** @type {number} */
|
|
this.minPts = 1;
|
|
/** @type {function} */
|
|
this.distance = this._euclideanDistance;
|
|
|
|
// temporary variables used during computation
|
|
|
|
/** @type {Array} */
|
|
this._reachability = [];
|
|
/** @type {Array} */
|
|
this._processed = [];
|
|
/** @type {number} */
|
|
this._coreDistance = 0;
|
|
/** @type {Array} */
|
|
this._orderedList = [];
|
|
|
|
this._init(dataset, epsilon, minPts, distanceFunction);
|
|
}
|
|
|
|
/******************************************************************************/
|
|
// pulic functions
|
|
|
|
/**
|
|
* Start clustering
|
|
*
|
|
* @param {Array} dataset
|
|
* @returns {undefined}
|
|
* @access public
|
|
*/
|
|
OPTICS.prototype.run = function(dataset, epsilon, minPts, distanceFunction) {
|
|
this._init(dataset, epsilon, minPts, distanceFunction);
|
|
|
|
for (var pointId = 0, l = this.dataset.length; pointId < l; pointId++) {
|
|
if (this._processed[pointId] !== 1) {
|
|
this._processed[pointId] = 1;
|
|
this.clusters.push([pointId]);
|
|
var clusterId = this.clusters.length - 1;
|
|
|
|
this._orderedList.push(pointId);
|
|
var priorityQueue = new PriorityQueue(null, null, 'asc');
|
|
var neighbors = this._regionQuery(pointId);
|
|
|
|
// using priority queue assign elements to new cluster
|
|
if (this._distanceToCore(pointId) !== undefined) {
|
|
this._updateQueue(pointId, neighbors, priorityQueue);
|
|
this._expandCluster(clusterId, priorityQueue);
|
|
}
|
|
}
|
|
}
|
|
|
|
return this.clusters;
|
|
};
|
|
|
|
/**
|
|
* Generate reachability plot for all points
|
|
*
|
|
* @returns {array}
|
|
* @access public
|
|
*/
|
|
OPTICS.prototype.getReachabilityPlot = function() {
|
|
var reachabilityPlot = [];
|
|
|
|
for (var i = 0, l = this._orderedList.length; i < l; i++) {
|
|
var pointId = this._orderedList[i];
|
|
var distance = this._reachability[pointId];
|
|
|
|
reachabilityPlot.push([pointId, distance]);
|
|
}
|
|
|
|
return reachabilityPlot;
|
|
};
|
|
|
|
/******************************************************************************/
|
|
// protected functions
|
|
|
|
/**
|
|
* Set object properties
|
|
*
|
|
* @param {Array} dataset
|
|
* @param {number} epsilon
|
|
* @param {number} minPts
|
|
* @param {function} distance
|
|
* @returns {undefined}
|
|
* @access protected
|
|
*/
|
|
OPTICS.prototype._init = function(dataset, epsilon, minPts, distance) {
|
|
|
|
if (dataset) {
|
|
|
|
if (!(dataset instanceof Array)) {
|
|
throw Error('Dataset must be of type array, ' +
|
|
typeof dataset + ' given');
|
|
}
|
|
|
|
this.dataset = dataset;
|
|
this.clusters = [];
|
|
this._reachability = new Array(this.dataset.length);
|
|
this._processed = new Array(this.dataset.length);
|
|
this._coreDistance = 0;
|
|
this._orderedList = [];
|
|
}
|
|
|
|
if (epsilon) {
|
|
this.epsilon = epsilon;
|
|
}
|
|
|
|
if (minPts) {
|
|
this.minPts = minPts;
|
|
}
|
|
|
|
if (distance) {
|
|
this.distance = distance;
|
|
}
|
|
};
|
|
|
|
/**
|
|
* Update information in queue
|
|
*
|
|
* @param {number} pointId
|
|
* @param {Array} neighbors
|
|
* @param {PriorityQueue} queue
|
|
* @returns {undefined}
|
|
* @access protected
|
|
*/
|
|
OPTICS.prototype._updateQueue = function(pointId, neighbors, queue) {
|
|
var self = this;
|
|
|
|
this._coreDistance = this._distanceToCore(pointId);
|
|
neighbors.forEach(function(pointId2) {
|
|
if (self._processed[pointId2] === undefined) {
|
|
var dist = self.distance(self.dataset[pointId], self.dataset[pointId2]);
|
|
var newReachableDistance = Math.max(self._coreDistance, dist);
|
|
|
|
if (self._reachability[pointId2] === undefined) {
|
|
self._reachability[pointId2] = newReachableDistance;
|
|
queue.insert(pointId2, newReachableDistance);
|
|
} else {
|
|
if (newReachableDistance < self._reachability[pointId2]) {
|
|
self._reachability[pointId2] = newReachableDistance;
|
|
queue.remove(pointId2);
|
|
queue.insert(pointId2, newReachableDistance);
|
|
}
|
|
}
|
|
}
|
|
});
|
|
};
|
|
|
|
/**
|
|
* Expand cluster
|
|
*
|
|
* @param {number} clusterId
|
|
* @param {PriorityQueue} queue
|
|
* @returns {undefined}
|
|
* @access protected
|
|
*/
|
|
OPTICS.prototype._expandCluster = function(clusterId, queue) {
|
|
var queueElements = queue.getElements();
|
|
|
|
for (var p = 0, l = queueElements.length; p < l; p++) {
|
|
var pointId = queueElements[p];
|
|
if (this._processed[pointId] === undefined) {
|
|
var neighbors = this._regionQuery(pointId);
|
|
this._processed[pointId] = 1;
|
|
|
|
this.clusters[clusterId].push(pointId);
|
|
this._orderedList.push(pointId);
|
|
|
|
if (this._distanceToCore(pointId) !== undefined) {
|
|
this._updateQueue(pointId, neighbors, queue);
|
|
this._expandCluster(clusterId, queue);
|
|
}
|
|
}
|
|
}
|
|
};
|
|
|
|
/**
|
|
* Calculating distance to cluster core
|
|
*
|
|
* @param {number} pointId
|
|
* @returns {number}
|
|
* @access protected
|
|
*/
|
|
OPTICS.prototype._distanceToCore = function(pointId) {
|
|
var l = this.epsilon;
|
|
for (var coreDistCand = 0; coreDistCand < l; coreDistCand++) {
|
|
var neighbors = this._regionQuery(pointId, coreDistCand);
|
|
if (neighbors.length >= this.minPts) {
|
|
return coreDistCand;
|
|
}
|
|
}
|
|
|
|
return;
|
|
};
|
|
|
|
/**
|
|
* Find all neighbors around given point
|
|
*
|
|
* @param {number} pointId
|
|
* @param {number} epsilon
|
|
* @returns {Array}
|
|
* @access protected
|
|
*/
|
|
OPTICS.prototype._regionQuery = function(pointId, epsilon) {
|
|
epsilon = epsilon || this.epsilon;
|
|
var neighbors = [];
|
|
|
|
for (var id = 0, l = this.dataset.length; id < l; id++) {
|
|
if (this.distance(this.dataset[pointId], this.dataset[id]) < epsilon) {
|
|
neighbors.push(id);
|
|
}
|
|
}
|
|
|
|
return neighbors;
|
|
};
|
|
|
|
/******************************************************************************/
|
|
// helpers
|
|
|
|
/**
|
|
* Calculate euclidean distance in multidimensional space
|
|
*
|
|
* @param {Array} p
|
|
* @param {Array} q
|
|
* @returns {number}
|
|
* @access protected
|
|
*/
|
|
OPTICS.prototype._euclideanDistance = function(p, q) {
|
|
var sum = 0;
|
|
var i = Math.min(p.length, q.length);
|
|
|
|
while (i--) {
|
|
sum += (p[i] - q[i]) * (p[i] - q[i]);
|
|
}
|
|
|
|
return Math.sqrt(sum);
|
|
};
|
|
|
|
if (typeof module !== 'undefined' && module.exports) {
|
|
module.exports = OPTICS;
|
|
}
|