https://github.com/inseefrlab/rtauargus
τ-Argus from R
https://github.com/inseefrlab/rtauargus
statistical-disclosure-control technical-building-block
Last synced: 6 months ago
JSON representation
τ-Argus from R
- Host: GitHub
- URL: https://github.com/inseefrlab/rtauargus
- Owner: InseeFrLab
- License: mit
- Created: 2021-12-27T09:57:18.000Z (over 4 years ago)
- Default Branch: master
- Last Pushed: 2025-08-04T22:56:30.000Z (8 months ago)
- Last Synced: 2025-08-05T00:46:00.983Z (8 months ago)
- Topics: statistical-disclosure-control, technical-building-block
- Language: HTML
- Homepage: https://inseefrlab.github.io/rtauargus/
- Size: 28.5 MB
- Stars: 10
- Watchers: 5
- Forks: 5
- Open Issues: 10
-
Metadata Files:
- Readme: README.Rmd
- Changelog: NEWS.md
- License: LICENSE
Awesome Lists containing this project
README
---
output: github_document
---
```{r, echo = FALSE}
knitr::opts_chunk$set(
collapse = TRUE,
comment = "#>",
fig.path = "man/figures/README-"
)
```
# rtauargus 
## Run τ-Argus from R
The *rtauargus* package provides an **R** interface for **τ-Argus**.
It allows to:
- create inputs (rda, arb, hst and tab files) from data in R format ;
- generate the sequence of instructions to be executed in batch mode (arb file);
- launch a τ-Argus batch in command line;
- retrieve the results in R.
These different operations can be executed in one go, but also in a modular way.
They allow to integrate the tasks performed by τ-Argus in a processing chain written in R.
The package presents other **additional functionalities**, such as:
- managing the protection of several tables at once;
- creating a hierarchical variable from correspondence table.
It's possible to choose a tabular or microdata approach, but the tabular
one is, from now on, encouraged.
## Installation
* **most recent stable version** (recommended)
- For Insee agents:
```{r eval=FALSE}
install.packages(
"rtauargus",
repos = "https://nexus.insee.fr/repository/r-public",
type = "source"
)
```
- Elsewhere:
```{r eval=FALSE}
install.packages("remotes")
remotes::install_github(
"InseeFrLab/rtauargus",
build_vignettes = FALSE,
upgrade = "never"
)
```
* **version in development**
To install a specific version, add to the directory a reference
([commit](https://github.com/inseefrlab/rtauargus/commits/master) or
[tag](https://github.com/inseefrlab/rtauargus/tags)),
for example `"inseefrlab/rtauargus@v-0.4.1"`.
## Simple example
When loading the package, the console displays some information:
```{r}
library(rtauargus)
```
In particular, a plausible location for the τ-Argus software is
predefined. This can be changed for the duration of the R session, as follows:
```{r}
loc_tauargus <- "Y:/Logiciels/TauArgus/TauArgus4.2.2b1/TauArgus.exe"
options(rtauargus.tauargus_exe = loc_tauargus)
```
With this small adjustment done, the package is ready to be used.
For the following demonstration, a fictitious table will be used:
```{r}
act_size <-
data.frame(
ACTIVITY = c("01","01","01","02","02","02","06","06","06","Total","Total","Total"),
SIZE = c("tr1","tr2","Total","tr1","tr2","Total","tr1","tr2","Total","tr1","tr2","Total"),
VAL = c(100,50,150,30,20,50,60,40,100,190,110,300),
N_OBS = c(10,5,15,2,5,7,8,6,14,20,16,36),
MAX = c(20,15,20,20,10,20,16,38,38,20,38,38)
)
```
As primary rules, we use the two following ones:
- The n-k dominance rule with n=1 and k = 85
- The minimum frequency rule with n = 3 and a safety range of 10.
To get the results for the dominance rule, we need to specify the largest
contributor to each cell, corresponding to the `MAX` variable in the tabular data.
```{r}
ex1 <- tab_rtauargus(
act_size,
dir_name = "tauargus_files",
files_name = "ex1",
explanatory_vars = c("ACTIVITY","SIZE"),
safety_rules = "FREQ(3,10)|NK(1,85)",
value = "VAL",
freq = "N_OBS",
maxscore = "MAX",
totcode = c(ACTIVITY="Total",SIZE="Total")
)
```
By default, the function displays in the console the logbook content in which
user can read all steps run by τ-Argus. This can be retrieved in the logbook.txt file.
With `verbose = FALSE`, the function can be silenced.
By default, the function returns the original dataset with one variable more,
called `Status`, directly resulting from τ-Argus and describing the status of
each cell as follows:
-`A`: primary secret cell because of frequency rule;
-`B`: primary secret cell because of dominance rule (1st contributor);
-`C`: primary secret cell because of frequency rule (more contributors in case when n>1);
-`D`: secondary secret cell;
-`V`: valid cells - no need to mask.
```{r}
ex1
```
All the files generated by the function are written in the specified directory
(`dir_name` argument).
The default format for the protected table is csv but it can be changed.
All the τ-Argus files (.tab, .rda, .arb and .txt) are written in the
same directory, too. To go further, you can consult the latest version of the
τ-Argus manual is downloadable here:
[https://research.cbs.nl/casc/Software/TauManualV4.1.pdf](https://research.cbs.nl/casc/Software/TauManualV4.1.pdf).
**A detailed overview is available via `vignette("rtauargus")`.**
## Important notes
The functions of *rtauargus* calling τ-Argus require that this software be accessible from the workstation. The download of τ-Argus is done on the [dedicated page](https://github.com/sdcTools/tauargus/releases) of the *sdcTools* git repository.
_The package was developed on the basis of open source versions of τ-Argus
(versions 4.2 and above), in particular the version used for this version is
τ-Argus 4.2.3. It is not compatible with version 3.5.**_