- Protohistory, Prehistoric Archaeology, Protohistoric Iberian Peninsula, Remote sensing and GIS applications in Landscape Research, Archaeological GIS, Computer Science, and 20 moreStatistical Modeling, Archaeology, Archaeological Predictive Modeling, Forecasting and Prediction Tools, Open Source and Free Software Studies, Logistic Regression, Archaeological Science, Digital Archaeology, Applied Statistics, Geographic Information Systems (GIS), Free Software, Bronze Age Europe (Archaeology), Iron Age, Spatial Analysis, Settlement Patterns, Archaeological Method & Theory, Spatial archaeology, Computer Applications & Quantitative Methods in Archaeology (CAA), Arqueología, and Arqueologiaedit
To better understand the factors that have contributed to the choice of locations of proto-historic settlements in the region of Beira Interior, various techniques of univariate statistical analysis, multivariate logistic regression and... more
To better understand the factors that have contributed to the choice of locations of proto-historic settlements in the region of Beira Interior, various techniques of univariate statistical analysis, multivariate logistic regression and Geographic Information Systems were used.
Fifteen environmental and cultural variables were analyzed throughout the territory of study, having been selected 4 that were included in the final logistic model.
The selection of these variables was confirmed by a resampling method, bootstrap, having been obtained the same model that achieved previously.
The final 4 variables highlight the importance of studying the visibility and defensibility for the entire area in question and not only for the known settlements.
It was demonstrated that most of the archaeological sites are situated to locally maximize these two factors.
To evaluate the final model a validation sample was used, consisting of 19 know settlements that were not used in its construction, having yielded a value of 0.975 for the Kvamme Gain Index.
A reduction of the high probability zones to only 1.5% of the total area under study was also achieved, showing good accuracy and precision values.
Applying this logistic model to the entire area of Beira Interior was possible to obtain a map of probability of proto-historic settlements, thus creating an auxiliary tool in its exploration and safeguard.
Fifteen environmental and cultural variables were analyzed throughout the territory of study, having been selected 4 that were included in the final logistic model.
The selection of these variables was confirmed by a resampling method, bootstrap, having been obtained the same model that achieved previously.
The final 4 variables highlight the importance of studying the visibility and defensibility for the entire area in question and not only for the known settlements.
It was demonstrated that most of the archaeological sites are situated to locally maximize these two factors.
To evaluate the final model a validation sample was used, consisting of 19 know settlements that were not used in its construction, having yielded a value of 0.975 for the Kvamme Gain Index.
A reduction of the high probability zones to only 1.5% of the total area under study was also achieved, showing good accuracy and precision values.
Applying this logistic model to the entire area of Beira Interior was possible to obtain a map of probability of proto-historic settlements, thus creating an auxiliary tool in its exploration and safeguard.
Research Interests:
The results of the study conducted in the region of Beira Interior (Central Portugal) are presented, aiming the creation of a predictive map that helps to identify probable areas of occurrence of proto-historic settlements, using only... more
The results of the study conducted in the region of Beira Interior (Central Portugal) are presented, aiming the creation of a predictive map that helps to identify probable areas of occurrence of proto-historic settlements, using only free software.
For a better understanding of some of the factors that led to the choice of certain inhabited places with this chronology, 15 environmental and cultural variables were analyzed, of which only 4 were selected for the final model, three of these were cultural, thus stressing the importance of including this type of predictors in archaeological models. This was achieved through a multivariate binary logistic regression with bootstrapping, having the resulting map four probability classes: low, medium, high and very high, corresponding the latest to 1.5% of the study area and yielding a value of 0.975 for the Kvamme gain Index.
A theoretical validation was performed using a sample of known sites, intending in the future to continue testing through field work.
For a better understanding of some of the factors that led to the choice of certain inhabited places with this chronology, 15 environmental and cultural variables were analyzed, of which only 4 were selected for the final model, three of these were cultural, thus stressing the importance of including this type of predictors in archaeological models. This was achieved through a multivariate binary logistic regression with bootstrapping, having the resulting map four probability classes: low, medium, high and very high, corresponding the latest to 1.5% of the study area and yielding a value of 0.975 for the Kvamme gain Index.
A theoretical validation was performed using a sample of known sites, intending in the future to continue testing through field work.
Os modelos preditivos ganharam destaque nas últimas décadas, mas em Portugal, na área da Arqueologia, poucos foram os arqueólogos que a eles se dedicaram desde a sua introdução, no ano de 2004, com os trabalhos de Helena Rua e Natália... more
Os modelos preditivos ganharam destaque nas últimas décadas, mas em Portugal, na área da Arqueologia, poucos foram os arqueólogos que a eles se dedicaram desde a sua introdução, no ano de 2004, com os trabalhos de Helena Rua e Natália Botica.
Ainda mal compreendidos e divulgados no nosso país, podem constituir uma ferramenta útil na identificação e protecção do património cultural, mas são por vezes erroneamente interpretados como “mapas do tesouro”.
Este texto reflecte sobre alguns dos diversos problemas relacionados com a sua finalidade, construção e resultados.
Predictive models have gained prominence in recent decades, but in Portugal, in the area of Archeology, few were the archaeologists who have dedicated themselves to them since its introduction in 2004, with the theses of Helena Rua and Natália Botica.
Still poorly understood and disseminated in our country, they can be a useful tool in the identification and protection of cultural heritage, but are sometimes erroneously interpreted as “treasure maps”.
This text reflects on some of the various problems related to its purpose, construction and results.
Ainda mal compreendidos e divulgados no nosso país, podem constituir uma ferramenta útil na identificação e protecção do património cultural, mas são por vezes erroneamente interpretados como “mapas do tesouro”.
Este texto reflecte sobre alguns dos diversos problemas relacionados com a sua finalidade, construção e resultados.
Predictive models have gained prominence in recent decades, but in Portugal, in the area of Archeology, few were the archaeologists who have dedicated themselves to them since its introduction in 2004, with the theses of Helena Rua and Natália Botica.
Still poorly understood and disseminated in our country, they can be a useful tool in the identification and protection of cultural heritage, but are sometimes erroneously interpreted as “treasure maps”.
This text reflects on some of the various problems related to its purpose, construction and results.
Research Interests:
Within her Master’s degree thesis (presented at the University of Coimbra), the author developed a predictive model for the Beira Interior region (in Central Portugal) in order to understand the factors that led to the choice of... more
Within her Master’s degree thesis (presented at the University of Coimbra), the author developed a predictive model for the Beira Interior region (in Central Portugal) in order to understand the factors that led to the choice of settlement locations in that area and to enable the identification of unknown archaeological sites. In this paper, she presents a summary of her work, the methodology used and the results she obtained.