void ComputeCorrelationsUShortOverlap(IBooleanMatrix entity_data)
{
var overlap = Overlap.ComputeUShort(entity_data);
for (int x = 0; x < NumEntities; x++)
for (int y = 0; y < x; y++)
this[x, y] = ComputeCorrelationFromOverlap(overlap[x, y], entity_data.NumEntriesByRow(x), entity_data.NumEntriesByRow(y));
}
void ComputeCorrelationsUIntOverlap(IBooleanMatrix entity_data)
{
var overlap = Overlap.ComputeUInt(entity_data);
// compute correlations
for (int x = 0; x < num_entities; x++)
for (int y = 0; y < x; y++)
{
this[x, y] = ComputeCorrelationFromOverlap(overlap[x, y], entity_data.NumEntriesByRow(x), entity_data.NumEntriesByRow(y));
this[y, x] = ComputeCorrelationFromOverlap(overlap[x, y], entity_data.NumEntriesByRow(y), entity_data.NumEntriesByRow(x));
}
}
void ComputeCorrelationsUShortOverlap(IBooleanMatrix entity_data)
{
var overlap = Overlap.ComputeUShort(entity_data);
// compute correlation
for (int x = 0; x < num_entities; x++)
{
for (int y = 0; y < x; y++)
{
this[x, y] = ComputeCorrelationFromOverlap(overlap[x, y], entity_data.NumEntriesByRow(x), entity_data.NumEntriesByRow(y));
this[y, x] = ComputeCorrelationFromOverlap(overlap[x, y], entity_data.NumEntriesByRow(y), entity_data.NumEntriesByRow(x));
}
}
}
void ComputeCorrelationsUIntOverlap(IBooleanMatrix entity_data)
{
var overlap = Overlap.ComputeUInt(entity_data);
for (int x = 0; x < NumEntities; x++)
{
for (int y = 0; y < x; y++)
{
this[x, y] = ComputeCorrelationFromOverlap(overlap[x, y], entity_data.NumEntriesByRow(x), entity_data.NumEntriesByRow(y));
}
}
}
///
public override void ComputeCorrelations(IBooleanMatrix entity_data)
{
var transpose = entity_data.Transpose();
var overlap = new SparseMatrix <int>(entity_data.NumberOfRows, entity_data.NumberOfRows);
// go over all (other) entities
for (int row_id = 0; row_id < transpose.NumberOfRows; row_id++)
{
var row = ((IBooleanMatrix)transpose).GetEntriesByRow(row_id);
for (int i = 0; i < row.Count; i++)
{
int x = row[i];
for (int j = i + 1; j < row.Count; j++)
{
int y = row[j];
if (x < y)
{
overlap[x, y]++;
}
else
{
overlap[y, x]++;
}
}
}
}
// the diagonal of the correlation matrix
for (int i = 0; i < num_entities; i++)
{
this[i, i] = 1;
}
// compute cosine
foreach (var index_pair in overlap.NonEmptyEntryIDs)
{
int x = index_pair.First;
int y = index_pair.Second;
this[x, y] = (float)(overlap[x, y] / Math.Sqrt(entity_data.NumEntriesByRow(x) * entity_data.NumEntriesByRow(y)));
}
}
/// <summary>Optimizes the specified data</summary>
/// <param name="data">data</param>
/// <param name="inverse_data">data</param>
/// <param name="W">W</param>
/// <param name="H">H</param>
void Optimize(IBooleanMatrix data, IBooleanMatrix inverse_data, Matrix<double> W, Matrix<double> H)
{
var HH = new Matrix<double>(num_factors, num_factors);
var HC_minus_IH = new Matrix<double>(num_factors, num_factors);
var HCp = new double[num_factors];
var m = new MathNet.Numerics.LinearAlgebra.Matrix(num_factors, num_factors);
MathNet.Numerics.LinearAlgebra.Matrix m_inv;
// TODO speed up using more parts of that library
// TODO using properties gives a 3-5% performance penalty
// source code comments are in terms of computing the user factors
// works the same with users and items exchanged
// (1) create HH in O(f^2|Items|)
// HH is symmetric
for (int f_1 = 0; f_1 < num_factors; f_1++)
for (int f_2 = 0; f_2 < num_factors; f_2++)
{
double d = 0;
for (int i = 0; i < H.dim1; i++)
d += H[i, f_1] * H[i, f_2];
HH[f_1, f_2] = d;
}
// (2) optimize all U
// HC_minus_IH is symmetric
for (int u = 0; u < W.dim1; u++)
{
var row = data.GetEntriesByRow(u);
// prepare KDD Cup specific weighting
int num_user_items = row.Count;
int user_positive_weight_sum = 0;
foreach (int i in row)
user_positive_weight_sum += inverse_data.NumEntriesByRow(i);
double neg_weight_normalization = (double) (num_user_items * (1 + CPos)) / (Feedback.Count - user_positive_weight_sum);
// TODO precompute
// TODO check whether this is correct
// create HC_minus_IH in O(f^2|S_u|)
for (int f_1 = 0; f_1 < num_factors; f_1++)
for (int f_2 = 0; f_2 < num_factors; f_2++)
{
double d = 0;
foreach (int i in row)
//d += H[i, f_1] * H[i, f_2] * (c_pos - 1);
d += H[i, f_1] * H[i, f_2] * CPos;
HC_minus_IH[f_1, f_2] = d;
}
// create HCp in O(f|S_u|)
for (int f = 0; f < num_factors; f++)
{
double d = 0;
for (int i = 0; i < inverse_data.NumberOfRows; i++)
if (row.Contains(i))
d += H[i, f] * (1 + CPos);
else
d += H[i, f] * inverse_data.NumEntriesByRow(i) * neg_weight_normalization;
HCp[f] = d;
}
// create m = HH + HC_minus_IH + reg*I
// m is symmetric
// the inverse m_inv is symmetric
for (int f_1 = 0; f_1 < num_factors; f_1++)
for (int f_2 = 0; f_2 < num_factors; f_2++)
{
double d = HH[f_1, f_2] + HC_minus_IH[f_1, f_2];
if (f_1 == f_2)
d += Regularization;
m[f_1, f_2] = d;
}
m_inv = m.Inverse();
// write back optimal W
for (int f = 0; f < num_factors; f++)
{
double d = 0;
for (int f_2 = 0; f_2 < num_factors; f_2++)
d += m_inv[f, f_2] * HCp[f_2];
W[u, f] = d;
}
}
}
void ComputeCorrelationsUShortOverlap(IBooleanMatrix entity_data)
{
var transpose = entity_data.Transpose() as IBooleanMatrix;
var overlap = new SymmetricMatrix<ushort>(entity_data.NumberOfRows);
// go over all (other) entities
for (int row_id = 0; row_id < transpose.NumberOfRows; row_id++)
{
var row = transpose.GetEntriesByRow(row_id);
for (int i = 0; i < row.Count; i++)
{
int x = row[i];
for (int j = i + 1; j < row.Count; j++)
overlap[x, row[j]]++;
}
}
// the diagonal of the correlation matrix
for (int i = 0; i < num_entities; i++)
this[i, i] = 1;
// compute cosine
for (int x = 0; x < num_entities; x++)
for (int y = 0; y < x; y++)
{
long size_product = entity_data.NumEntriesByRow(x) * entity_data.NumEntriesByRow(y);
if (size_product > 0)
this[x, y] = (float) (overlap[x, y] / Math.Sqrt(size_product));
}
}
/// <summary>Optimizes the specified data</summary>
/// <param name="data">data</param>
/// <param name="inverse_data">data</param>
/// <param name="W">W</param>
/// <param name="H">H</param>
void Optimize(IBooleanMatrix data, IBooleanMatrix inverse_data, Matrix <double> W, Matrix <double> H)
{
var HH = new Matrix <double>(num_factors, num_factors);
var HC_minus_IH = new Matrix <double>(num_factors, num_factors);
var HCp = new double[num_factors];
var m = new MathNet.Numerics.LinearAlgebra.Matrix(num_factors, num_factors);
MathNet.Numerics.LinearAlgebra.Matrix m_inv;
// TODO speed up using more parts of that library
// TODO using properties gives a 3-5% performance penalty
// source code comments are in terms of computing the user factors
// works the same with users and items exchanged
// (1) create HH in O(f^2|Items|)
// HH is symmetric
for (int f_1 = 0; f_1 < num_factors; f_1++)
{
for (int f_2 = 0; f_2 < num_factors; f_2++)
{
double d = 0;
for (int i = 0; i < H.dim1; i++)
{
d += H[i, f_1] * H[i, f_2];
}
HH[f_1, f_2] = d;
}
}
// (2) optimize all U
// HC_minus_IH is symmetric
for (int u = 0; u < W.dim1; u++)
{
var row = data.GetEntriesByRow(u);
// prepare KDD Cup specific weighting
int num_user_items = row.Count;
int user_positive_weight_sum = 0;
foreach (int i in row)
{
user_positive_weight_sum += inverse_data.NumEntriesByRow(i);
}
double neg_weight_normalization = (double)(num_user_items * (1 + CPos)) / (Feedback.Count - user_positive_weight_sum);
// TODO precompute
// TODO check whether this is correct
// create HC_minus_IH in O(f^2|S_u|)
for (int f_1 = 0; f_1 < num_factors; f_1++)
{
for (int f_2 = 0; f_2 < num_factors; f_2++)
{
double d = 0;
foreach (int i in row)
{
//d += H[i, f_1] * H[i, f_2] * (c_pos - 1);
d += H[i, f_1] * H[i, f_2] * CPos;
}
HC_minus_IH[f_1, f_2] = d;
}
}
// create HCp in O(f|S_u|)
for (int f = 0; f < num_factors; f++)
{
double d = 0;
for (int i = 0; i < inverse_data.NumberOfRows; i++)
{
if (row.Contains(i))
{
d += H[i, f] * (1 + CPos);
}
else
{
d += H[i, f] * inverse_data.NumEntriesByRow(i) * neg_weight_normalization;
}
}
HCp[f] = d;
}
// create m = HH + HC_minus_IH + reg*I
// m is symmetric
// the inverse m_inv is symmetric
for (int f_1 = 0; f_1 < num_factors; f_1++)
{
for (int f_2 = 0; f_2 < num_factors; f_2++)
{
double d = HH[f_1, f_2] + HC_minus_IH[f_1, f_2];
if (f_1 == f_2)
{
d += Regularization;
}
m[f_1, f_2] = d;
}
}
m_inv = m.Inverse();
// write back optimal W
for (int f = 0; f < num_factors; f++)
{
double d = 0;
for (int f_2 = 0; f_2 < num_factors; f_2++)
{
d += m_inv[f, f_2] * HCp[f_2];
}
W[u, f] = d;
}
}
}
///
public override void ComputeCorrelations(IBooleanMatrix entity_data)
{
var transpose = entity_data.Transpose();
var overlap = new SparseMatrix<int>(entity_data.NumberOfRows, entity_data.NumberOfRows);
// go over all (other) entities
for (int row_id = 0; row_id < transpose.NumberOfRows; row_id++)
{
var row = ((IBooleanMatrix) transpose).GetEntriesByRow(row_id);
for (int i = 0; i < row.Count; i++)
{
int x = row[i];
for (int j = i + 1; j < row.Count; j++)
{
int y = row[j];
if (x < y)
overlap[x, y]++;
else
overlap[y, x]++;
}
}
}
// the diagonal of the correlation matrix
for (int i = 0; i < num_entities; i++)
this[i, i] = 1;
// compute cosine
foreach (var index_pair in overlap.NonEmptyEntryIDs)
{
int x = index_pair.First;
int y = index_pair.Second;
this[x, y] = (float) (overlap[x, y] / Math.Sqrt(entity_data.NumEntriesByRow(x) * entity_data.NumEntriesByRow(y) ));
}
}
///
public override void ComputeCorrelations(IBooleanMatrix entity_data)
{
var transpose = entity_data.Transpose() as IBooleanMatrix;
var overlap = new SymmetricMatrix<int>(entity_data.NumberOfRows);
// go over all (other) entities
for (int row_id = 0; row_id < transpose.NumberOfRows; row_id++)
{
var row = transpose.GetEntriesByRow(row_id);
for (int i = 0; i < row.Count; i++)
{
int x = row[i];
for (int j = i + 1; j < row.Count; j++)
{
int y = row[j];
overlap[x, y]++;
}
}
}
// the diagonal of the correlation matrix
for (int i = 0; i < num_entities; i++)
this[i, i] = 1;
// compute Jaccard index
for (int x = 0; x < num_entities; x++)
for (int y = 0; y < x; y++)
this[x, y] = (float) (overlap[x, y] / (entity_data.NumEntriesByRow(x) + entity_data.NumEntriesByRow(y) - overlap[x, y]));
}
