Computational and Machine Learning Tools for Archaeological Site Modeling (Springer Theses)

✍ Scribed by Maria Elena Castiello

• Publisher: Springer
• Tongue: English
• Leaves: 304
• Category: Library

✦ Synopsis

This book describes a novel machine-learning based approach   to answer some traditional archaeological problems, relating to archaeological site detection and site locational preferences. Institutional data collected from six Swiss regions (Zurich, Aargau, Grisons, Vaud, Geneva and Fribourg) have been analyzed with an original conceptual framework based on the Random Forest algorithm. It is shown how the algorithm can assist in the modelling process in connection with heterogeneous, incomplete archaeological datasets and related cultural heritage information. Moreover, an in-depth review of past and more recent works of quantitative methods for archaeological predictive modelling is provided. The book guides the readers to set up their own protocol for: i) dealing with uncertain data, ii) predicting archaeological site location, iii) establishing environmental features importance, iv) and suggest a model validation procedure. It addresses both academics and professionals in archaeology and cultural heritage management, and offers a source of inspiration for future research directions in the field of digital humanities and computational archaeology.

 

✦ Table of Contents

Supervisor’s Foreword
Acknowledgements
Contents
Abbreviations
Section I
1 Introduction
1.1 Research Context
1.2 Swiss Archaeological Heritage Management
1.3 Motivation—Research Questions
1.4 Challenges and Objectives
1.5 Thesis Outline
References
Section II
2 Space, Environment and Quantitative Approaches in Archaeology
2.1 Landscape Archaeology: A Synopsis
2.2 Computational and Quantitative Approaches
2.3 Geographic Information Systems
References
3 Predictive Modeling
3.1 Theoretical Perspective and Model Definition
3.2 From Global to Local Scale: Indicative Case Studies and Experiences
3.2.1 Theoretical Framework
3.2.2 From Spatial Analysis to Machine Learning Applications: Case Studies
3.2.3 Delivering Uncertainty with Archaeological Predictive Models
3.2.4 A Swiss Case Study
3.3 Uncertainty and Vagueness in Archaeological Predictive Modelling
3.4 Data Mining and Machine Learning Techniques
3.5 Random Forest: Classification and Regression Trees
3.6 Towards New Perspectives in Archaeological Practices
References
Section III
4 Materials and Data
4.1 Premise
4.2 The Concept of ‘Archaeological Site’ and Relative Issues
4.3 Historical Framework of the Research Areas
4.4 Canton of Zurich
4.4.1 General Framework of the Region
4.4.2 Archaeological Dataset
4.5 Canton of Aargau
4.5.1 General Framework of the Region
4.5.2 Archaeological Dataset
4.6 Canton of Grisons
4.6.1 General Framework of the Region
4.6.2 Archaeological Dataset
4.7 Canton of Fribourg/Freiburg
4.7.1 General Framework of the Region
4.7.2 Archaeological Dataset
4.8 Canton of Vaud
4.8.1 General Framework of the Region
4.8.2 Archaeological Dataset
4.9 Canton of Geneva
4.9.1 General Framework of the Region
4.9.2 Archaeological Dataset
4.10 Geo-Environmental Predictors
4.10.1 Topography
4.10.2 Hydrology
4.10.3 Soil and Agriculture Suitability Map
4.10.4 Geology
References
5 Modeling Approach
5.1 GIS Preprocessing
5.1.1 Conceptual Modeling for the Archaeological Database
5.2 Mapping Uncertainty
5.3 Preparing the Environmental Variables
5.3.1 Topography
5.3.2 Hydrology—Distance to Water
5.3.3 Soil Map—Agricultural Suitability
5.3.4 Geology
5.4 Locational Preference Analysis
5.4.1 All Cantons
5.4.2 Canton of Zurich
5.4.3 Canton of Aargau
5.4.4 Canton of Fribourg
5.4.5 Canton of Geneva
5.4.6 Canton of Vaud
5.4.7 Canton of Grisons
5.5 Random Forest Based Approach
References
Section IV
6 Results and Discussion
6.1 Zurich
6.2 Aargau
6.3 Grisons
6.4 Vaud
6.5 Geneva
6.6 Fribourg
6.7 Switzerland
6.8 Validity Assessment
6.9 Previous Knowledge—New Knowledge
6.10 Limitations and Advantages
References
7 Conclusions
7.1 Main Achievements and Conclusions
7.2 Research Perspectives
References
Appendix
A.1 Canton of Aargau
A.1.1 Database Reclassification
A.1.2 Environmental Variables
A.1.3 Locational Preference Analysis
A.1.4 RF Classification Model Results
A.1.5 Comparative Analysis of Predicted High Probability Areas—RF Classification
A.1.6 RF Regression Model Results (Single Finds)
A.2 Canton of Fribourg
A.2.1 Database Reclassification
A.2.2 Environmental Variables
A.2.3 Locational Preference Analysis
A.2.4 RF Classification Model Results
A.2.5 Comparative Analysis of Predicted High Probability Areas—RF Classification
A.3 Canton of Geneva
A.3.1 Database Reclassification
A.3.2 Environmental Variables
A.3.3 Locational Preference Analysis
A.3.4 RF Classification Model Results
A.3.5 Comparative Analysis of Predicted High Probability Areas—RF Classification
A.3.6 RF Regression Model Results
A.3.7 Comparative Analysis of Predicted High Probability Areas—RF Regression
A.4 Canton of Grisons
A.4.1 Database Reclassification
A.4.2 Environmental Variables
A.4.3 Locational Preference Analysis
A.4.4 RF Classification Model Results
A.4.5 Comparative Analysis of Predicted High Probability Areas—RF Classification
A.5 Canton of Vaud
A.5.1 Database Reclassification
A.5.2 Environmental Variables
A.5.3 Locational Preference Analysis
A.5.4 RF Classification Model Results
A.5.5 Comparative Analysis of Predicted High Probability Areas—RF Classification
A.6 Canton of Zurich
A.6.1 Database Reclassification
A.6.2 Environmental Variables
A.6.3 Locational Preference Analysis
A.6.4 RF Classification Model Results
A.6.5 Comparative Analysis of Predicted High Probability Areas—RF Classification
A.7 Switzerland
A.7.1 Environmental Variables
A.7.2 Locational Preference Analysis
A.7.3 RF Classification Model Results
A.7.4 Comparative Analysis of Predicted High Probability Areas—RF Classification
About the Author
References