tesselle / kairos

:hourglass: Analysis of Chronological Patterns from Archaeological Count Data

tesselle

Overview

The tesselle suite is a collection of packages for research and teaching in archaeology. These packages focus on quantitative analysis methods developed for archaeology. The tesselle packages are designed to work seamlessly together and to complement general-purpose and other specialized statistical packages. These packages can be used to explore and analyze common data types in archaeology: count data, compositional data and chronological data.

The tesselle package is designed to make it easy to install and load key packages from the tesselle suite in a single step.

To cite tesselle in publications use:

  Frerebeau N (2024). _tesselle: Easily Install and Load 'tesselle'
  Packages_. Université Bordeaux Montaigne, Pessac, France.
  doi:10.5281/zenodo.6500491 <https://doi.org/10.5281/zenodo.6500491>,
  R package version 1.5.0, <https://packages.tesselle.org/tesselle/>.

This package is a part of the tesselle project
<https://www.tesselle.org>.

Installation

You can install the released version of tesselle from CRAN with:

install.packages("tesselle")

And the development version from GitHub with:

# install.packages("remotes")
remotes::install_github("tesselle/tesselle")

Usage

library(tesselle) will load the core packages:

• tabula: analysis and visualization of archaeological count data;
• kairos: analysis of chronological patterns from archaeological count data;
• nexus: analysis of compositional data;

And two companion packages:

• dimensio: multivariate analysis;
• isopleuros: ternary plots.

library(tesselle)
#> --- Attaching packages -------------------------------------- tesselle 1.5.0 ---
#> * dimensio    0.6.0
#> * isopleuros  1.2.0
#> * kairos      2.1.0
#> * nexus       0.2.0
#> * tabula      3.0.1

Contributing

Please note that the tesselle project is released with a Contributor Code of Conduct. By contributing to this project, you agree to abide by its terms.

kairos's Issues

Investigate Collins-Elliott (2019)

Collins-Elliott, S. A. (2019). Quantifying artefacts over time: Interval estimation of a Poisson distribution using the Jeffreys prior. Archaeometry, 61:1207-1222. https://doi.org/10.1111/arcm.12481.

Investigate Bellanger L., Coulon A. and Husi P. (2021)

Bellanger L., Coulon A. and Husi P. (2021b). Une méthode de classification ascendante hiérarchique par compromis : hclustcompro. 9e Conférence Internationale Francophone sur la Science des Données (CIFSD), CIFSD, Mohamed QUAFAFOU, Jun 2021, Marseille, France ⟨hal-03280918⟩.

Check all data references

Check that the data used in the examples are all properly cited.

Add seriation methods

Move seriation methods from tabula to kairos.

temporal blocks vs fixed dates

Hi,
Many thanks for a great package. I have been learning aoristic analysis comparing R packages and have noticed a different implementation of the method. It seems that kairos and aoristAAR is calculating the aoristic sum at fixed dates of temporal blocks.
Whereas, archSeries calculates weights within the temporal blocks. The archSeries output is also consistent with methodology and code by Crema and Palmisano. I hope the example below explain what I mean.

I use as data example from figure 3 from Palmisano et al, 2017

`
library(kairos)

dataOne = data.frame(
id = c("a", "b", "c", "d"),
from = c(-2200, -2000, -1600, -2000),
to = c(-1400, -1600, -1400, -1400)
)

aoristOne = kairos::aoristic(dataOne, step = 200, weight = TRUE)

kairos::get_dates(aoristOne)
[1] -2200 -2000 -1800 -1600 -1400

kairos::get_weights(aoristOne)
[,1] [,2] [,3] [,4] [,5]
[1,] 0.25 0.2500000 0.2500000 0.2500000 0.2500000
[2,] 0.00 0.5000000 0.5000000 0.5000000 0.0000000
[3,] 0.00 0.0000000 0.0000000 1.0000000 1.0000000
[4,] 0.00 0.3333333 0.3333333 0.3333333 0.3333333

[email protected]
[,1]
[1,] 0.250000
[2,] 1.083333
[3,] 1.083333
[4,] 2.083333
[5,] 1.583333

`

`

library(archSeries)

dataTwo = data.frame(
id = c("a", "b", "c", "d"),
Start = c(-2200, -2000, -1600, -2000),
End = c(-1400, -1600, -1400, -1400)
)

aoristTwo = archSeries::aorist(dataTwo, bin.width = 200)
aoristTwo
bin bin.no aorist
1: -2200--2000 1 0.2500000
2: -2000--1800 2 1.0833333
3: -1800--1600 3 1.0833333
4: -1600--1400 4 0.5833333

`

The kairos package returned 5 aoristic sums for each date break but archSeries calculated 4 sums for each temporal bin.

I would appreciate any help with understanding the method implemented in kairos package.
Especially, that the kairos::roc() function returns 4 bins for the same dataset.

Investigate Bellanger L., Coulon A. and Husi P. (2020)

Bellanger L., Coulon A. and Husi P. (2020) PerioClust: a new Hierarchical agglomerative clustering method including temporal or spatial ordering contraints. In Chatzipantelis Th. et al. (ed.), Data Analysis and Rationality in a Complex World, XXIII. Springer International Publishing (Springer Series, Studies in Classification, Data Analysis, and Knowledge Organization) ⟨ISBN:978-3-030-60104-4⟩.

Investigate Collins-Elliott (2017)

Collins-Elliott, S. A. (2017). Bayesian inference with Monte Carlo approximation: Measuring regional differentiation in ceramic and glass vessel assemblages in Republican Italy, ca. 200 BCE-20 CE. Journal of Archaeological Science, 80:37-49. https://doi.org/10.1016/j.jas.2017.01.006.

Seriation by constrained correspondence analysis

Poblome, Jeroen, and Patrick J.F. Groenen. 2003. "Constrained Correspondence Analysis for Seriation of Sagalassos Tablewares". In The Digital Heritage of Archaeology, edited by Martin Doerr and Apostolis Sarris. Athens: Hellenic Ministry of Culture.

Velden, Michel van de, Patrick J.F. Groenen, and Jeroen Poblome. 2009. ‘Seriation by Constrained Correspondence Analysis: A Simulation Study’. Computational Statistics & Data Analysis, 53 (8), p. 3129-38. https://doi.org/10.1016/j.csda.2008.08.020.

• #2

Test the significance of seriation solutions

Marko Porčić (2013). The goodness of fit and statistical significance of seriation solutions. Journal of Archaeological Science, 40(12): 4552-4559. https://doi.org/10.1016/j.jas.2013.07.013.

• #2