Page 187 - AC/E Digital Culture Annual Report

on which to explore issues common to many cultural heritage artefacts, such as design styles, patterns and motifs, which in this case were relevant to a particular time and place.An ontology was developed for documenting architectural mouldings decorated with di erent ornament, making it possible to retrieve them with the assistance of the system that identi ed similar mouldings. The proposed ontology thus allowed 3D digital models to be catalogued based on information on their artistic style and method of execution.A shape analysis method was also designed to improve the information automatically extracted from 3D digital content, allowing di erent users to consult a particular repository using the semantic connections underlying each element.The idea is to make this system applicable to other heritage objects, as the shape analysis algorithm developed here could also be applied to a variety of heritage shapes in order to learn about their semantic signi cance.Soft computing tools in archaeologyContinuing with the diversi cation and appli- cability of tasks using arti cial intelligence, we nd a research project entitled “Aplicación de t cnicas soft computing y modelado 3D en el tratamiento masivo de datos en arqueología” (Application of soft computing and 3D modelling techniques in the mass handling of data in archaeology, 2015–2017), which set out to explore the potential of soft computing techniques – a branch of arti cial intelligence – for analysing archaeological data.The issue examined here was the increasing amount of data obtained from archaeological excavations. The sheer quantity of variables to be considered and their interaction subject to multiple combinations prevents them from being analysed and classi ed in greater depth. For this purpose the use of soft computing techniqueswas proposed. Based on advances in arti cial intelligence, these techniques make it possible to manage a large amount of partial information and nd classi catory and/or statistical parame- ters.To test the use of soft computing, three archae- ological sites dating from prehistory, early history and antiquity were chosen.The rst was the Palaeolithic settlement of El Salt (Alicante), where evidence was documented of the existence of three game species during the period: deer, horses and mountain goats. The many small fragments found make it di cult to recognise the species, and zooarchaeological identi cation rates were low. Computational analysis based on soft computing can facilitate this task by establishing parameters for identi- cation and saving time with the classi cation and cataloguing of the fragments.The second case was the Theban Tomb 209 (Luxor) belonging to the 25th dynasty, where evidence had been found of di erences in the execution of decorative funerary inscriptions, suggesting the hand of at least two scribes. Us- ing computational tools, it is therefore possible to analyse types of hieroglyphic writing in depth, paying special attention to elements that could otherwise go unnoticed, such as the distance between signs or the ratio of sign size to column width, which would enable the di erent author- ship to be identi ed.The third case was the Roman Punic city of Carteia (San Roque, Cádiz), where the aim was to study typological characteristics such as con- structional and stylistic patterns for the surviving Punic and Roman architectural remains. Ceramic ware was also taken into account as an import- ant area where soft computing can be particular useful: the variety of types and the fragmented state of the pieces made them extremely di cult to classify and catalogue on the basis of variables such as geometric shapes, colours, textures, decoration or modelling, among others.AC/E DIGITAL CULTURE ANNUAL REPORT 2017187The use of digital technologies in the conservation, analysis and dissemination of cultural heritage