This paper deals with analytical analyses of samples from one of the most important prehistoric sites, which is Riparo Mochi in Ventimiglia (Imperia - Italy), that aim at obtaining a reliable reconstruction of the combustion structures at the site. The adopted multi-analytical approach relies on Fourier Transform Infrared (FTIR) pectroscopy and X-ray powder diffraction (XRPD). The qualitative analysis allowed the main mineralogical phases present in the findings extracted at the prehistoric site to be identified. To improve the quantitative performance in the materials’ analysis, the available data was post-processed taking advantage of Artificial Intelligence (AI) strategies. A data-set of FTIR spectra that refer to pellets of known concentration of three of the identified substances, which are calcite, silicate and quartz, was built. Such a data-set was used to train and validate a neural network, which showed good performance in predicting the concentration of two out of three of the investigated substances. The low performance of the neural network in predicting the concentration of the third compound was directly related to the poor reproducibility of the pellet production process, which remains an important challenge for improving the network performance.
A Multi-Disciplinary Study based on Archaeometry and Artificial Intelligence: A New Approach for the Investigation of Hearths at the Riparo Mochi Paleolithic Site / Guglielmi, Vittoria; Corbellini, Simone; Grimaldi, Stefano; Lombardo, Luca; Santaniello, Fabio; Santiglia, Alessia; Tassi, Anna Laura; Sento, Marco; Carullo, Alessio. - ELETTRONICO. - (2024), pp. 75-80. (Intervento presentato al convegno IEEE International Workshop on Metrology for Living Environment (MetroLivEnv) tenutosi a Chania (Greece) nel 12-14 June 2024) [10.1109/metrolivenv60384.2024.10615925].
A Multi-Disciplinary Study based on Archaeometry and Artificial Intelligence: A New Approach for the Investigation of Hearths at the Riparo Mochi Paleolithic Site
Corbellini, Simone;Lombardo, Luca;Sento, Marco;Carullo, Alessio
2024
Abstract
This paper deals with analytical analyses of samples from one of the most important prehistoric sites, which is Riparo Mochi in Ventimiglia (Imperia - Italy), that aim at obtaining a reliable reconstruction of the combustion structures at the site. The adopted multi-analytical approach relies on Fourier Transform Infrared (FTIR) pectroscopy and X-ray powder diffraction (XRPD). The qualitative analysis allowed the main mineralogical phases present in the findings extracted at the prehistoric site to be identified. To improve the quantitative performance in the materials’ analysis, the available data was post-processed taking advantage of Artificial Intelligence (AI) strategies. A data-set of FTIR spectra that refer to pellets of known concentration of three of the identified substances, which are calcite, silicate and quartz, was built. Such a data-set was used to train and validate a neural network, which showed good performance in predicting the concentration of two out of three of the investigated substances. The low performance of the neural network in predicting the concentration of the third compound was directly related to the poor reproducibility of the pellet production process, which remains an important challenge for improving the network performance.File | Dimensione | Formato | |
---|---|---|---|
1571006716 final.pdf
accesso aperto
Descrizione: A Multi-Disciplinary Study Based on Archaeometry and Artificial Intelligence: A New Approach for the Investigation of Hearths at the Riparo Mochi Paleolithic Site
Tipologia:
2. Post-print / Author's Accepted Manuscript
Licenza:
PUBBLICO - Tutti i diritti riservati
Dimensione
419.16 kB
Formato
Adobe PDF
|
419.16 kB | Adobe PDF | Visualizza/Apri |
Sento-AMulti.pdf
non disponibili
Tipologia:
2a Post-print versione editoriale / Version of Record
Licenza:
Non Pubblico - Accesso privato/ristretto
Dimensione
746.5 kB
Formato
Adobe PDF
|
746.5 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/11583/2992617