theplu::yat::utility::WeNNI Class Reference

Weighted Nearest Neighbour Imputation. More...

#include <yat/utility/WeNNI.h>

Inheritance diagram for theplu::yat::utility::WeNNI:

theplu::yat::utility::NNI

List of all members.

Public Member Functions

 WeNNI (const utility::Matrix &matrix, const utility::Matrix &weight, const unsigned int neighbours)
unsigned int estimate (void)
 Function doing WeNNI imputation.
const utility::Matriximputed_data_raw (void) const
const utility::Matriximputed_data (void) const
const std::vector< size_t > & not_imputed (void) const

Protected Member Functions

std::vector< std::pair< size_t,
double > > 
calculate_distances (const size_t) const
std::vector< size_t > nearest_neighbours (const size_t, const std::vector< std::pair< size_t, double > > &) const

Protected Attributes

const utility::Matrixdata_
utility::Matrix imputed_data_
unsigned int neighbours_
std::vector< size_t > not_imputed_
const utility::Matrixweight_


Detailed Description

Weighted Nearest Neighbour Imputation.

WeNNI is a continuous weights generalization of the (binary weights) kNNI algorithm presented by Troyanskaya et al. A reference to this paper is found in the NNI document referred to in the NNI class documentation. The NNI document also describes WeNNI in depth.

Note:
Missing values should be represented with a zero weight. WeNNI will treat the corresponding data values as zero, i.e., this implies that NaNs and Infs with zero weight will not have any impact on calculations.a
See also:
NNI and kNNI

Constructor & Destructor Documentation

theplu::yat::utility::WeNNI::WeNNI ( const utility::Matrix matrix,
const utility::Matrix weight,
const unsigned int  neighbours 
)

Constructor


Member Function Documentation

unsigned int theplu::yat::utility::WeNNI::estimate ( void   )  [virtual]

Function doing WeNNI imputation.

Perform WeNNI on data in matrix with continuous uncertainty weights in weight using neighbours for the new impute value.

The return value can be used as an indication of how well the imputation worked. The return value should be zero if proper pre-processing of data is done. An example of bad data is a matrix with a column of zero weights, another is a corresponding situation with a row with all weights zero.

Returns:
The number of rows that have at least one value not imputed.

Implements theplu::yat::utility::NNI.

const utility::Matrix& theplu::yat::utility::WeNNI::imputed_data_raw ( void   )  const [inline]

Returns:
A const reference to imputed_data_raw.

const utility::Matrix& theplu::yat::utility::NNI::imputed_data ( void   )  const [inherited]

Returns:
A const reference to the modified data.

const std::vector<size_t>& theplu::yat::utility::NNI::not_imputed ( void   )  const [inherited]

Returns:
indices of rows in data matrix not imputed

std::vector<std::pair<size_t,double> > theplu::yat::utility::NNI::calculate_distances ( const   size_t  )  const [protected, inherited]

$ d_{ij}^2=\frac {\sum_{k=1}^C w_{ik} w_{jk} (x_{ik}-x_{jk})^2 }{\sum_{k=l}^C w_{ik} w_{jk} } $ where C is the number of columns

std::vector<size_t> theplu::yat::utility::NNI::nearest_neighbours ( const   size_t,
const std::vector< std::pair< size_t, double > > &   
) const [protected, inherited]

Contributing nearest neighbours are added up to the user set number, and neighbours are disqualified if their element (column) weight is zero


Member Data Documentation

const utility::Matrix& theplu::yat::utility::NNI::data_ [protected, inherited]

original data matrix

data after imputation

unsigned int theplu::yat::utility::NNI::neighbours_ [protected, inherited]

number of neighbor to use

std::vector<size_t> theplu::yat::utility::NNI::not_imputed_ [protected, inherited]

which rows are not imputed due to lack of data

const utility::Matrix& theplu::yat::utility::NNI::weight_ [protected, inherited]

weight matrix


The documentation for this class was generated from the following file:

Generated on Tue Jan 18 02:20:11 2011 for yat by  doxygen 1.5.5