Data Driven Humanities. May 18th 2018. Data Driven Innovation 2018. Engineering Department, University of Roma Tre

1. https://www.flickr.com/photos/wwworks/
Data Driven Archaeology
Gabriele Gattiglia
Università di Pisa

2. i s a research Lab of the University of Pisa, in which archaeologists,
mathematicians and geologists deal with:
• mathematical models for archaeology (www.mappaproject.org)
• open data (www.mappaproject.org/mod)
and now they are starting to explore risks and potentiality of a Big Data
approach to archaeology.
MAPPA Lab

5. ArchAIDE
Archaeological Automatic Interpretation and
Documentation of cEramics
[Datafication]

6. From digitisation….
Digitisation has changed archaeology deeply. It has boosted the volume of data that can be
analysed.
….to datafication
• to transform objects, processes, etc. in a quantified format so they can be tabulated and
analysed.
• to record new information;
• to produce a constant flow of data, starting from the data produced by archaeological
practice, such as locations and relationships between finds and sites, that the archaeological
community should have available with minimum time delay, to process again and again.
• to fit Big Data approach and permit mathematical analyses to identify non‐linear
relationships
These new data can modify the way we conduct our analyses, increase our capacity to process
and visualise information in novel ways, and more decisively, provide new ways of doing
archaeological research.
“Archaeology is a place within the social sciences and Humanities where the nature
of the work deals with Big Data”
As archaeologists, we are well‐aware that it is easier to create new datasets than transform old
ones.

7. ArchAIDE
Archaeological Automatic Interpretation and
Documentation of cEramics
*Data recording

8. ArchAIDE
Archaeological Automatic Interpretation and
Documentation of cEramics
*access to data

9. ArchAIDE
Archaeological Automatic Interpretation and
Documentation of cEramics
*Data Management

10. ArchAIDE
Archaeological Automatic Interpretation and
Documentation of cEramics
*Data Analysis

11. https://www.flickr.com/photos/wwworks/
[Distribution Map]
[Data Visualisation]

12. Predictive Map of Archaeological
Potential
Archaeological Map
Paleogeographical Map
Mathematical model Open digital archaeological archive
Open Data
Cooperation
Transparency
Geology
Archaeology
Mathematics
Pisa, Italy

13. DATA MODEL
• Urban data
• Historical
cartography data
• Geographical/
geomorphological
data
• Archaeological
data
PRIMARY DATA
• available data
SECONDARY DATA

14. MAPPAGIS
www.mappaproject.org/webgis

15. 7. Report,
explain,
predict
1. Real-world
problem
3. Formulate
the abstract
problem
4. Solve the
abstract problem
5. Interpret
the
solution
6. Verify
the model
TESTING
2. Make assumptions
BASIC RULES governing the system
PROBABILITY DISTRIBUTIONS
MATHEMATICAL
MODELS
STATISTICAL
MODELS

16. A page pointing to other pages distributes its
importance to those pages








13
312
21
3/1
3/2
PP
PPP
PP
PageRank
briefly
Pinitial provides the initial potential
values, given by the available data
The matrix W assigns values of links
between cells
1
1/32/3
1











003/1
103/2
010
W
 321initialP

17. Creation of functional areas
Using the archaeological
findings we draw up the
functional areas, i.e. levels of
spatial and functional
organization (e.g. urban,
suburban, rural areas) in which
the urban space is organized.
Archaeological
findings
Functional area

18. Geomorphological, stratigraphic geophisical and archaeological
data have been processed using geostatistical techniques in order
to get temporal DEMs for 7 periods from Protohistory to
Contemporary Age and to create
paleogeographical maps
Creation of paleogeographical maps

19. Contemporary Age
Modern Age
Late Medieval period
Early Medieval period
Roman period
Etruscan period
Protohistory

20. THE MAP OF ARCHAEOLOGICAL POTENTIAL OF THE URBAN
AREA OF PISA
The map of archaeological potential is given by
the composition of the 7 layers.
It’s a weighted sum in which archaeological
periods with less information available are taken
more into account.

21. [www.archaide.eu]
The ArchAIDE project goes exactly in this direction.

22. ArchAIDE aims to support the classification and interpretation work of the
archaeologists with innovative computer‐based tools, able to provide the user with
features for matching of each discovered sherd over the huge existing ceramic
catalogues.
Goals
from to

23. [How?]

24. From drawings digitisationto 3D models

25. [algorythms][algorythms]

26. Appearance‐based recognition
Training set of 15.000images
On 77%the first result is the right one
On 95%the right one is within the top 5results
gg1

27. Diapositiva 26
gg1 gabriele gattiglia; 12/02/2018

28. 1 3
Ranking of
pottery
classes by
relevance
2
Shape‐basedrecognition
The neural network has been trained both with real and syntetic
data
Preliminary results say that the right result is within the top 5
on the
78%

29. ArchAIDE
Archaeological Automatic Interpretation and
Documentation of cEramics

30. ArchAIDE
Archaeological Automatic Interpretation and
Documentation of cEramics

31. Just one minute…..

32. ArchAIDE
Archaeological Automatic Interpretation and
Documentation of cEramics
[Big Data]

36. https://www.flickr.com/photos/wwworks/

38. This project has received funding from the European Union’s Horizon 2020 research
and innovation programme under grant agreement N.693548
The views and opinions expressed in this presentation are
the sole responsibility of the authors and do not necessarily
reflect the views of the European Commission.
[Thank you for your attention]
www.archaide.eu