diff --git a/DESCRIPTION b/DESCRIPTION
index e08c3cfb03d1fe70a2dff30420ac2a99966755cc..ab69e371aa9bf136c933121a790bf040500c0b1b 100644
--- a/DESCRIPTION
+++ b/DESCRIPTION
@@ -5,12 +5,19 @@ Authors@R: c(
person("Bertrand", "Cloez", email = "bertrand.cloez@inrae.fr", role = c("aut")),
person("Isabelle", "Sanchez", email = "isabelle.sanchez@inrae.fr", role = c("aut", "cre")),
person("Benedicte", "Fontez", email = "benedicte.fontez@supagro.fr", role = c("ctr")))
-Author: Bertrand Cloez [aut, cre],
- Isabelle Sanchez [ctr],
+Author: Bertrand Cloez [aut],
+ Isabelle Sanchez [aut, cre],
Benedicte Fontez [ctr]
Maintainer: Isabelle Sanchez <isabelle.sanchez@inrae.fr>
Description: A method for detecting outliers with a Kalman filter on impulsed
- noised outliers and prediction on cleaned data.
+ noised outliers and prediction on cleaned data. kfino is a robust sequential
+ algorithm allowing to filter data with a large number of outliers. This
+ algorithm is based on simple latent linear Gaussian processes as in the
+ Kalman Filter method and is devoted to detect impulse-noised outliers. These
+ are data points that differ significantly from other observations. `ML`
+ (Maximization Likelihood) and `EM` (Expectation-Maximization algorithm)
+ algorithms were implemented in `kfino`. The method is described in full
+ details in the following arxiv preprint: <https://arxiv.org/abs/2208.00961>.
License: GPL-3
Depends: R (>= 4.1.0)
Encoding: UTF-8
diff --git a/R/utils_functions.R b/R/utils_functions.R
index 0d94b02e21964623129cfcd50f1ced5299d5dc84..486944010ca5459e59e4f41caa71c429ceeef560 100644
--- a/R/utils_functions.R
+++ b/R/utils_functions.R
@@ -6,7 +6,7 @@
# KBO_EM()
#-------------------------------------------------------------------
-#' doutlier This function defines an outlier distribution (Surface of a
+#' doutlier defines an outlier distribution (Surface of a
#' trapezium) and uses input parameters given in the main function kfino_fit()
#'
#' @param y numeric, point
@@ -15,7 +15,14 @@
#' @param expertMax numeric, the maximal weight expected by the user
#'
#' @details this function is used to calculate an outlier distribution
-#' following a trapezium shape
+#' following a trapezium shape.
+#' \eqn{y \mapsto \text{doutlier}(y,K,\text{expertMin},\text{expertMax})}
+#' is the probability density function on
+#' \eqn{[\text{expertMin},\text{expertMax}]} which is linear and verifies
+#' \eqn{\text{doutlier}(\text{expertMax},K,\text{expertMin},\text{expertMax})
+#' =K*\text{doutlier}(\text{expertMin},K,\text{expertMin},\text{expertMax}).}
+#' In particular, when $K=1$ this corresponds to the uniform distribution.
+#'
#' @return a numeric value
#' @export
#'
@@ -34,7 +41,7 @@ doutlier<-function(y,
#' KBO_known a function to calculate a likelihood on given parameters
#'
#' @param param list, see initial parameter list in \code{kfino_fit}
-#' @param threshold numeric, threshold for CI, default 0.5
+#' @param threshold numeric, threshold for confidence interval, default 0.5
#' @param kappa numeric, truncation setting for likelihood optimization,
#' default 10
#' @param Y character, name of the numeric variable to predict in the
@@ -50,8 +57,8 @@ doutlier<-function(y,
#' \item{prediction}{vector, the prediction of weights}
#' \item{label}{vector, probability to be an outlier}
#' \item{likelihood}{numeric, the calculated likelihood}
-#' \item{lwr}{vector of lower bound CI of the prediction }
-#' \item{upr}{vector of upper bound CI of the prediction }
+#' \item{lwr}{vector of lower bound confidence interval of the prediction }
+#' \item{upr}{vector of upper bound confidence interval of the prediction }
#' \item{flag}{char, is an outlier or not}
#' }
#' @export
diff --git a/man/KBO_known.Rd b/man/KBO_known.Rd
index e28e51fafc0f7ee221d2d15ee76b7935cb19fa78..fe48f4c112bb7d00259f1d5ebe5c219128c6b8ee 100644
--- a/man/KBO_known.Rd
+++ b/man/KBO_known.Rd
@@ -9,7 +9,7 @@ KBO_known(param, threshold, kappa = 10, Y, Tps, N)
\arguments{
\item{param}{list, see initial parameter list in \code{kfino_fit}}
-\item{threshold}{numeric, threshold for CI, default 0.5}
+\item{threshold}{numeric, threshold for confidence interval, default 0.5}
\item{kappa}{numeric, truncation setting for likelihood optimization,
default 10}
@@ -29,8 +29,8 @@ a list
\item{prediction}{vector, the prediction of weights}
\item{label}{vector, probability to be an outlier}
\item{likelihood}{numeric, the calculated likelihood}
- \item{lwr}{vector of lower bound CI of the prediction }
- \item{upr}{vector of upper bound CI of the prediction }
+ \item{lwr}{vector of lower bound confidence interval of the prediction }
+ \item{upr}{vector of upper bound confidence interval of the prediction }
\item{flag}{char, is an outlier or not}
}
}
diff --git a/man/doutlier.Rd b/man/doutlier.Rd
index efd7eaa4a3348bd58deace1e7ef4401e3586dfcd..58f1886141154c7be8f76fea88509dafefafb7b4 100644
--- a/man/doutlier.Rd
+++ b/man/doutlier.Rd
@@ -2,7 +2,7 @@
% Please edit documentation in R/utils_functions.R
\name{doutlier}
\alias{doutlier}
-\title{doutlier This function defines an outlier distribution (Surface of a
+\title{doutlier defines an outlier distribution (Surface of a
trapezium) and uses input parameters given in the main function kfino_fit()}
\usage{
doutlier(y, K, expertMin, expertMax)
@@ -20,12 +20,18 @@ doutlier(y, K, expertMin, expertMax)
a numeric value
}
\description{
-doutlier This function defines an outlier distribution (Surface of a
+doutlier defines an outlier distribution (Surface of a
trapezium) and uses input parameters given in the main function kfino_fit()
}
\details{
this function is used to calculate an outlier distribution
- following a trapezium shape
+ following a trapezium shape.
+ \eqn{y \mapsto \text{doutlier}(y,K,\text{expertMin},\text{expertMax})}
+ is the probability density function on
+ \eqn{[\text{expertMin},\text{expertMax}]} which is linear and verifies
+ \eqn{\text{doutlier}(\text{expertMax},K,\text{expertMin},\text{expertMax})
+ =K*\text{doutlier}(\text{expertMin},K,\text{expertMin},\text{expertMax}).}
+ In particular, when $K=1$ this corresponds to the uniform distribution.
}
\examples{
doutlier(2,5,10,45)
diff --git a/man/kfino.Rd b/man/kfino.Rd
index 19a0444fe19ccd3c4668e44495b10f7875dcd251..a2ec5bf77f3b1d02d8953e9adbe37b334cce01f3 100644
--- a/man/kfino.Rd
+++ b/man/kfino.Rd
@@ -8,7 +8,7 @@
\description{
\if{html}{\figure{logo.png}{options: style='float: right' alt='logo' width='120'}}
-A method for detecting outliers with a Kalman filter on impulsed noised outliers and prediction on cleaned data.
+A method for detecting outliers with a Kalman filter on impulsed noised outliers and prediction on cleaned data. kfino is a robust sequential algorithm allowing to filter data with a large number of outliers. This algorithm is based on simple latent linear Gaussian processes as in the Kalman Filter method and is devoted to detect impulse-noised outliers. These are data points that differ significantly from other observations. `ML` (Maximization Likelihood) and `EM` (Expectation-Maximization algorithm) algorithms were implemented in `kfino`. The method is described in full details in the following arxiv preprint: \url{https://arxiv.org/abs/2208.00961}.
}
\details{
xxxxxxxx xxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxx.