this function is used as a kernel-function-identifier [ takes the distances and a weights-kernel (in form of a function) and returns weights ]

Description

this function is used as a kernel-function-identifier [ takes the distances and a weights-kernel (in form of a function) and returns weights ]

Usage

FUNCTION_weights(W_dist_matrix, weights_function, eps = 1e-06)

Australian Credit Approval

Description

This is the famous Australian Credit Approval dataset, originating from the StatLog project. It concerns credit card applications. All attribute names and values have been changed to meaningless symbols to protect the confidentiality of the data.

Usage

data(australian)

Format

A data frame with 690 Instances and 15 attributes (including the class attribute, "Class")

Details

There are 6 numerical and 8 categorical attributes, all normalized to [-1,1]. The original formatting was as follows: A1: A,B class attribute (formerly: +,-) A2: 0,1 CATEGORICAL (formerly: a,b) A3: continuous. A4: continuous. A5: 1,2,3 CATEGORICAL (formerly: p,g,gg) A6: 1, 2,3,4,5, 6,7,8,9,10,11,12,13,14 CATEGORICAL (formerly: ff,d,i,k,j,aa,m,c,w, e, q, r,cc, x) A7: 1, 2,3, 4,5,6,7,8,9 CATEGORICAL (formerly: ff,dd,j,bb,v,n,o,h,z) A8: continuous. A9: 1, 0 CATEGORICAL (formerly: t, f) A10: 1, 0 CATEGORICAL (formerly: t, f) A11: continuous. A12: 1, 0 CATEGORICAL (formerly t, f) A13: 1, 2, 3 CATEGORICAL (formerly: s, g, p) A14: continuous. A15: continuous.

Source

Confidential. Donated by Ross Quinlan

References

[LibSVM] (https://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/binary.html), UCI - 1987

Examples

data(australian)

X = australian[, -1]

y = australian[, 1]

stratified folds (in classification) [ detailed information about class_folds in the FeatureSelection package ]

Description

this function creates stratified folds in binary and multiclass classification

Usage

class_folds(folds, RESP)

this function compute entropy

Description

this function compute entropy

Usage

entropy(x)

Fast Class Noise Detector with Multi-Factor-Based Learning

Description

This function computes the noise score for each observation

Usage

fmf(x, ...)

## S3 method for class 'formula'
fmf(formula, data, ...)

## Default S3 method:
fmf(
  x,
  knn = 5,
  classColumn = 1,
  boxplot_range = 1,
  iForest = TRUE,
  threads = 1,
  ...
)

Arguments

Value

an object of class filter, which is a list with four components:

• cleanData is a data frame containing the filtered dataset.

• remIdx is a vector of integers indicating the indexes for removed instances (i.e. their row number with respect to the original data frame).

• noise_score is a vector of values indicating the optential of being a noise.

• call contains the original call to the filter.

Author(s)

Wanwan Zheng

Examples

data(iris)
out = fmf(Species~.,iris)

OPTION to convert categorical features TO either numeric [ if levels more than 32] OR to dummy variables [ if levels less than 32 ]

Description

OPTION to convert categorical features TO either numeric [ if levels more than 32] OR to dummy variables [ if levels less than 32 ]

Usage

func_categorical_preds(prepr_categ)

shuffle data

Description

this function shuffles the items of a vector

Usage

func_shuffle(vec, times = 10)

this function returns a table of probabilities for each label

Description

this function returns a table of probabilities for each label

Usage

func_tbl(DF, W, labels)

this function returns the probabilities in case of classification

Description

this function returns the probabilities in case of classification

Usage

func_tbl_dist(DF, Levels)

Iris Data Set

Description

This is perhaps the best known database to be found in the pattern recognition literature. Fisher's paper is a classic in the field and is referenced frequently to this day. (See Duda & Hart, for example.) The data set contains 3 classes of 50 instances each, where each class refers to a type of iris plant. One class is linearly separable from the other 2; the latter are NOT linearly separable from each other.

Usage

data(iris)

Format

A data frame with 150 Instances and 4 attributes (including the class attribute, "Species") In this package, the iris dataset has been normalized by the max-min normalization.

Details

Fisher,R.A. "The use of multiple measurements in taxonomic problems" Annual Eugenics, 7, Part II, 179-188 (1936); also in "Contributions to Mathematical Statistics" (John Wiley, NY, 1950).

Predicted attribute: class of iris plant.

This is an exceedingly simple domain.

This data differs from the data presented in Fishers article (identified by Steve Chadwick, spchadwick '@' espeedaz.net ). The 35th sample should be: 4.9,3.1,1.5,0.2,"setosa" where the error is in the fourth feature. The 38th sample: 4.9,3.6,1.4,0.1,"setosa" where the errors are in the second and third features.

Source

Creator:

R.A. Fisher

Donor:

Michael Marshall

References

https://archive.ics.uci.edu/ml/datasets/iris

Examples

data(iris)

X = iris[, -1]

y = iris[, 1]

kernel k-nearest-neighbors

Description

This function utilizes kernel k nearest neighbors to predict new observations

Usage

kernelknn(
  data,
  TEST_data = NULL,
  y,
  k = 5,
  h = 1,
  method = "euclidean",
  weights_function = NULL,
  regression = FALSE,
  transf_categ_cols = FALSE,
  threads = 1,
  extrema = FALSE,
  Levels = NULL
)

Arguments

Details

This function takes a number of arguments and it returns the predicted values. If TEST_data is NULL then the predictions for the train data will be returned, whereas if TEST_data is not NULL then the predictions for the TEST_data will be returned. There are three possible ways to specify the weights function, 1st option : if the weights_function is NULL then a simple k-nearest-neighbor is performed. 2nd option : the weights_function is one of 'uniform', 'triangular', 'epanechnikov', 'biweight', 'triweight', 'tricube', 'gaussian', 'cosine', 'logistic', 'gaussianSimple', 'silverman', 'inverse', 'exponential'. The 2nd option can be extended by combining kernels from the existing ones (adding or multiplying). For instance, I can multiply the tricube with the gaussian kernel by giving 'tricube_gaussian_MULT' or I can add the previously mentioned kernels by giving 'tricube_gaussian_ADD'. 3rd option : a user defined kernel function

Value

a vector (if regression is TRUE), or a data frame with class probabilities (if regression is FALSE)

Author(s)

Lampros Mouselimis edit by Wanwan Zheng

The Max-Min Normalization

Description

This function normalizes the data using the max-min normalization

Usage

normalization(x, margin = 2)

Arguments

Author(s)

Wanwan Zheng

Examples

data(ozone)
scaled.data = normalization(ozone[,-1])
ozone.scale = data.frame(y = as.character(ozone[,1]), scaled.data[,-1])

this function normalizes the data

Description

this function normalizes the data

Usage

normalized(x)

Ozone Level Detection Data Set

Description

Forecasting skewed biased stochastic ozone days: analyses, solutions and beyond, Knowledge and Information Systems, Vol. 14, No. 3, 2008. Discusses details about the dataset, its use as well as various experiments (both cross-validation and streaming) using many state-of-the-art methods.

A shorter version of the paper (does not contain some detailed experiments as the journal paper above) is in: Forecasting Skewed Biased Stochastic Ozone Days: Analyses and Solutions. ICDM 2006: 753-764

Usage

data(ozone)

Format

A data frame with 2536 Instances and 73 attributes (including the class attribute, "Class": ozone day, normal day)

Details

The following are specifications for several most important attributes that are highly valued by Texas Commission on Environmental Quality (TCEQ). More details can be found in the two relevant papers.

– O 3 - Local ozone peak prediction – Upwind - Upwind ozone background level – EmFactor - Precursor emissions related factor – Tmax - Maximum temperature in degrees F – Tb - Base temperature where net ozone production begins (50 F) – SRd - Solar radiation total for the day – WSa - Wind speed near sunrise (using 09-12 UTC forecast mode) – WSp - Wind speed mid-day (using 15-21 UTC forecast mode)

Source

Kun Zhang zhang.kun05 '@' gmail.com Department of Computer Science, Xavier University of Lousiana

Wei Fan wei.fan '@' gmail.com IBM T.J.Watson Research

XiaoJing Yuan xyuan '@' uh.edu Engineering Technology Department, College of Technology, University of Houston

References

https://archive.ics.uci.edu/ml/datasets/Ozone+Level+Detection

Examples

data(ozone)

X = ozone[, -1]

y = ozone[, 1]

PCA Plot of the Noise Score of Each Individual

Description

This function plots the noise score for each observation

Usage

plot(
  score,
  data,
  cl,
  geom.ind = "text",
  labelsize = 3,
  geom_point_size = 3,
  ...
)

Arguments

Value

an plot of PCA with the noise score of each observation

Author(s)

Wanwan Zheng

Examples

data(iris)
out = fmf(Species~.,iris)
plot(out$noise_score, iris[,-1], iris[,1])

create folds (in regression) [ detailed information about class_folds in the FeatureSelection package ]

Description

this function creates both stratified and non-stratified folds in regression

Usage

regr_folds(folds, RESP)

Arithmetic operations on lists

Description

Arithmetic operations on lists

Usage

switch.ops(LST, MODE = "ADD")

Tuning For Fast Class Noise Detector with Multi-Factor-Based Learning

Description

This function tunes the hyper-parameters the threshold and the k of k-NN

Usage

tuning(x, ...)

## S3 method for class 'formula'
tuning(formula, data, ...)

## Default S3 method:
tuning(
  x,
  knn_k = seq(3, 7, 2),
  classColumn = 1,
  boxplot_range = seq(0.1, 1.1, 0.2),
  repeats = 10,
  method = "svm",
  iForest = TRUE,
  threads = 1,
  ...
)

Arguments

Value

An object of class filter, which is a list with two components:

• summary is the a vector of values when different hyper-parameter is set.

• call contains the original call to the filter.

Author(s)

Wanwan Zheng

Examples

data(iris)
out = tuning(Species~.,iris)