/// <summary>
/// Creates a new <see cref="CachedDissimilarityMetric{TInstance}" /> according to the given base dissimilarity metric
/// and the known set of instances.
/// </summary>
/// <param name="dissimilarityMeasure">The metric to be used to cache the dissimilarities between all instances.</param>
/// <param name="allInstances">The set of instances for which to calculate the pair-wise dissimilarities.</param>
public CachedDissimilarityMetric(
IDissimilarityMetric <TInstance> dissimilarityMeasure, ISet <TInstance> allInstances)
{
this._dissimilarityMeasure = dissimilarityMeasure;
// registers instances' indexes
var instances = allInstances.ToArray();
for (var i = 0; i < instances.Length; i++)
{
this._elementIdxs.Add(instances[i], i);
}
// initializes cache with distances between all instances
this._dissimilarities = new double[instances.Length][];
for (var i = 0; i < allInstances.Count; i++)
{
for (var j = i; j < allInstances.Count; j++)
{
if (i == 0)
{
this._dissimilarities[j] = new double[instances.Length];
}
this.Store(i, j, instances);
}
}
}
/// <summary>
/// Returns the medoid of a given <see cref="Cluster{TInstance}" />, i.e., a representative object whose dissimilarity
/// to all the instances in the cluster is minimal.
/// </summary>
/// <remarks>
/// "Medoids are representative objects of a data set or a cluster with a data set whose average dissimilarity to all
/// the objects in the cluster is minimal. Medoids are similar in concept to means or centroids, but medoids are
/// always restricted to be members of the data set. Medoids are most commonly used on data when a mean or centroid
/// cannot be defined, such as graphs. They are also used in contexts where the centroid is not representative of the
/// dataset like in images and 3-D trajectories and gene expression (where while the data is sparse the medoid need
/// not be). These are also of interest while wanting to find a representative using some distance other than squared
/// euclidean distance (for instance in movie-ratings)."
/// <see href="https://en.wikipedia.org/wiki/Medoid" />
/// </remarks>
/// <param name="cluster">The cluster whose medoid we want to retrieve.</param>
/// <param name="metric">
/// The dissimilarity metric used to compare elements in the cluster, i.e., to calculate the distance between them.
/// </param>
/// <returns>The medoid of the given cluster. If the cluster has two elements, it returns the first element of the cluster.</returns>
/// <typeparam name="TInstance">The type of instance considered.</typeparam>
public static TInstance GetMedoid <TInstance>(
this Cluster <TInstance> cluster, IDissimilarityMetric <TInstance> metric)
where TInstance : IComparable <TInstance>
{
// if count is 1 or 2 return the first elem
if (cluster.Count < 3)
{
return(cluster.First());
}
// stores sum of all distances to each point
var sumDists = new double[cluster.Count];
var clusterList = cluster.ToList();
for (var i = 0; i < clusterList.Count; i++)
{
for (var j = i + 1; j < clusterList.Count; j++)
{
var dist = metric.Calculate(clusterList[i], clusterList[j]);
sumDists[i] += dist;
sumDists[j] += dist;
}
}
// selects elem index minimizing the mean distance (the medoid)
var minSumDist = double.MaxValue;
var minSumIdx = -1;
for (var i = 0; i < clusterList.Count; i++)
{
if (sumDists[i] < minSumDist)
{
minSumDist = sumDists[i];
minSumIdx = i;
}
}
return(clusterList[minSumIdx]);
}
private void LoadDataSet()
{
// loads data-points
var parser = new CsvParser();
this._dataPoints = parser.Load(Path.GetFullPath(this.openFileDialog.FileName));
// clears series
this._chartDataPoints.Clear();
this.ChartPoints.Clear();
// adds points to series
var maxX = double.MinValue;
var minX = double.MaxValue;
foreach (var dataPoint in this._dataPoints)
{
var chartDataPoint = new ChartDataPoint(dataPoint.Value[0], dataPoint.Value[1])
{
Label = dataPoint.ID
};
this.ChartPoints.Add(chartDataPoint);
this._chartDataPoints.Add(dataPoint, chartDataPoint);
maxX = Math.Max(maxX, dataPoint.Value[0]);
minX = Math.Min(minX, dataPoint.Value[0]);
}
// resets
this._numClusters = int.MinValue;
this._clusteringResult = null;
this._dissimilarityMetric = new CachedDissimilarityMetric <DataPoint>(new DataPoint(), this._dataPoints);
// adjusts track-bar according to num clusters
this.numClustersTrackBar.SmallChange = (uint)(this.numClustersTrackBar.Maximum / this._dataPoints.Count);
this.numClustersTrackBar.LargeChange = this.numClustersTrackBar.SmallChange * 5;
this.datasetChart.ChartAreas[0].AxisX.Maximum = Math.Ceiling(maxX);
this.datasetChart.ChartAreas[0].AxisX.Minimum = Math.Floor(minX);
}
/// <summary>
/// Creates a new <see cref="ModifiedGammaStatistic{TInstance}" /> with given dissimilarity metric.
/// </summary>
/// <param name="dissimilarityMetric">The metric used to calculate dissimilarity between cluster elements.</param>
/// <param name="centroidFunc">
/// A function to get an element representing the centroid of a <see cref="Cluster{TInstance}" />.
/// </param>
public ModifiedGammaStatistic(
IDissimilarityMetric <TInstance> dissimilarityMetric, CentroidFunction <TInstance> centroidFunc)
{
this._centroidFunc = centroidFunc;
this.DissimilarityMetric = dissimilarityMetric;
}
/// <summary>
/// Creates a new <see cref="RSquared{TInstance}" /> with given dissimilarity metric.
/// </summary>
/// <param name="dissimilarityMetric">The metric used to calculate dissimilarity between cluster elements.</param>
/// <param name="centroidFunc">
/// A function to get an element representing the centroid of a <see cref="Cluster{TInstance}" />.
/// </param>
public RSquared(IDissimilarityMetric <TInstance> dissimilarityMetric,
CentroidFunction <TInstance> centroidFunc)
{
this._centroidFunc = centroidFunc;
this.DissimilarityMetric = dissimilarityMetric;
}
/// <summary>
/// Creates a new <see cref="DaviesBouldinIndex{TInstance}" /> with given dissimilarity metric.
/// </summary>
/// <param name="dissimilarityMetric">The metric used to calculate dissimilarity between cluster elements.</param>
/// <param name="centroidFunc">
/// A function to get an element representing the centroid of a <see cref="Cluster{TInstance}" />.
/// </param>
public DaviesBouldinIndex(
IDissimilarityMetric <TInstance> dissimilarityMetric, CentroidFunction <TInstance> centroidFunc)
{
this._centroidFunc = centroidFunc;
this.DissimilarityMetric = dissimilarityMetric;
}
/// <summary>
/// Creates a new <see cref="WithinBetweenRatio{TInstance}" /> with given dissimilarity metric.
/// </summary>
/// <param name="dissimilarityMetric">The metric used to calculate dissimilarity between cluster elements.</param>
/// <param name="centroidFunc">
/// A function to get an element representing the centroid of a <see cref="Cluster{TInstance}" />.
/// </param>
public WithinBetweenRatio(IDissimilarityMetric <TInstance> dissimilarityMetric,
CentroidFunction <TInstance> centroidFunc)
{
this._centroidFunc = centroidFunc;
this.DissimilarityMetric = dissimilarityMetric;
}
/// <summary>
/// Creates a new <see cref="RootMeanSquareStdDev{TInstance}" /> with given dissimilarity metric.
/// </summary>
/// <param name="dissimilarityMetric">The metric used to calculate dissimilarity between cluster elements.</param>
/// <param name="centroidFunc">
/// A function to get an element representing the centroid of a <see cref="Cluster{TInstance}" />.
/// </param>
public RootMeanSquareStdDev(
IDissimilarityMetric <TInstance> dissimilarityMetric, CentroidFunction <TInstance> centroidFunc)
{
this._centroidFunc = centroidFunc;
this.DissimilarityMetric = dissimilarityMetric;
}
/// <summary>
/// Creates a new <see cref="CentroidLinkage{TInstance}" /> with given dissimilarity metric and centroid
/// function.
/// </summary>
/// <param name="metric">The metric used to calculate dissimilarity between cluster elements.</param>
/// <param name="centroidFunc">
/// A function to get an element representing the centroid of a <see cref="Cluster{TInstance}" />.
/// </param>
public CentroidLinkage(
IDissimilarityMetric <TInstance> metric, Func <Cluster <TInstance>, TInstance> centroidFunc)
{
this.DissimilarityMetric = metric;
this._centroidFunc = centroidFunc;
}
/// <summary>
/// Creates a new <see cref="AverageLinkage{TInstance}" /> with given dissimilarity metric.
/// </summary>
/// <param name="metric">The metric used to calculate dissimilarity between cluster elements.</param>
public AverageLinkage(IDissimilarityMetric <TInstance> metric)
{
this.DissimilarityMetric = metric;
}
/// <summary>
/// Creates a new <see cref="CompleteLinkage{TInstance}" /> with given dissimilarity metric.
/// </summary>
/// <param name="metric">The metric used to calculate dissimilarity between cluster elements.</param>
public CompleteLinkage(IDissimilarityMetric <TInstance> metric)
{
this.DissimilarityMetric = metric;
}
/// <summary>
/// Creates a new <see cref="DunnIndex{TInstance}" /> with given dissimilarity metric.
/// </summary>
/// <param name="dissimilarityMetric">The metric used to calculate dissimilarity between cluster elements.</param>
public DunnIndex(IDissimilarityMetric <TInstance> dissimilarityMetric)
{
this.DissimilarityMetric = dissimilarityMetric;
}
/// <summary>
/// Creates a new <see cref="SilhouetteCoefficient{TInstance}" /> with given dissimilarity metric.
/// </summary>
/// <param name="dissimilarityMetric">The metric used to calculate dissimilarity between cluster elements.</param>
public SilhouetteCoefficient(IDissimilarityMetric <TInstance> dissimilarityMetric)
{
this.DissimilarityMetric = dissimilarityMetric;
}
/// <summary>
/// Creates a new <see cref="DaviesBouldinIndex{TInstance}" /> with given dissimilarity metric.
/// </summary>
/// <param name="dissimilarityMetric">The metric used to calculate dissimilarity between cluster elements.</param>
/// <param name="centroidFunc">
/// A function to get an element representing the centroid of a <see cref="Cluster{TInstance}" />.
/// </param>
public CalinskiHarabaszIndex(
IDissimilarityMetric <TInstance> dissimilarityMetric, CentroidFunction <TInstance> centroidFunc)
{
this._centroidFunc = centroidFunc;
this.DissimilarityMetric = dissimilarityMetric;
}
/// <summary>
/// Creates a new <see cref="MinimumEnergyLinkage{TInstance}" /> with given dissimilarity metric.
/// </summary>
/// <param name="metric">The metric used to calculate dissimilarity between cluster elements.</param>
public MinimumEnergyLinkage(IDissimilarityMetric <TInstance> metric)
{
this.DissimilarityMetric = metric;
}
/// <summary>
/// Creates a new <see cref="XieBeniIndex{TInstance}" /> with given dissimilarity metric.
/// </summary>
/// <param name="dissimilarityMetric">The metric used to calculate dissimilarity between cluster elements.</param>
/// <param name="centroidFunc">
/// A function to get an element representing the centroid of a <see cref="Cluster{TInstance}" />.
/// </param>
public XieBeniIndex(
IDissimilarityMetric <TInstance> dissimilarityMetric, CentroidFunction <TInstance> centroidFunc)
{
this._centroidFunc = centroidFunc;
this.DissimilarityMetric = dissimilarityMetric;
}
/// <summary>
/// Creates a new <see cref="WardsMinimumVarianceLinkage{TInstance}" /> with given dissimilarity metric and centroid
/// function.
/// </summary>
/// <param name="metric">The metric used to calculate dissimilarity between cluster elements.</param>
/// <param name="centroidFunc">
/// A function to get an element representing the centroid of a <see cref="Cluster{TInstance}" />.
/// </param>
public WardsMinimumVarianceLinkage(
IDissimilarityMetric <TInstance> metric, CentroidFunction <TInstance> centroidFunc)
{
this.DissimilarityMetric = metric;
this._centroidFunc = centroidFunc;
}
