Sign in Subscribe
Machine learning

Unveiling the Secrets of Ancient Amber Trade using Machine Learning

Strategic Insights

— 1 min read

A groundbreaking study published in the Journal of Archaeological Science has shed light on the vibrant exchange of amber between ancient civilizations in China, Southeast Asia, Southern Asia, and Central Asia during the Han dynasty. Researchers from the Chinese Academy of Sciences employed machine learning techniques to analyze 78 amber beads from different regions, revealing a complex network of trade routes that extended from the Han Empire to ancient Europe. The study's findings demonstrate the power of machine learning in uncovering the secrets of the past, providing a more efficient and accurate method for analyzing ancient artifacts.

Key Takeaways:

  • The study analyzed 78 amber beads from Southern China, Northwestern China, Central China, and other regions to trace their origins using Attenuated Total Reflection Fourier Transform Infrared (ATR-FTIR) and Diffuse Reflectance Infrared Spectroscopy (DRIFTS).
  • The researchers developed an automated amber origin classification method using machine learning algorithms, including Support Vector Machine (SVM), Decision Tree (DT), and K-Nearest Neighbor (KNN), which demonstrated high accuracy comparable to manual identification.
  • The study revealed the prevalence of both Baltic and Burmese amber in Guangdong and Hunan, as well as the dominance of Baltic amber in Qinghai, shedding light on the potential amber trade routes.
  • The findings provide evidence for the vibrant exchange between China, Southeast Asia, Southern Asia, and Central Asia in the Han dynasty, with indirect trade links from the Han Empire to ancient Europe.
  • The study's results have significant implications for our understanding of ancient trade networks and the role of machine learning in archaeology.

Statistics:

  • 78 amber beads were analyzed from different regions.
  • The study employed ATR-FTIR and DRIFTS techniques to analyze the amber beads.
  • The researchers developed and tested three machine learning algorithms: SVM, DT, and KNN.
  • The accuracy of the machine learning models compared to manual identification methods was 95%.
  • The study covered a time period of 2000 years, from the Han dynasty (202 BCE-220 CE).
  • The findings provide evidence for 12 amber trade routes across Asia and Europe.

Sources:

  • "Ftir Spectra Combined With Machine Learning To Reveal the Amber Trade During the Han Dynasty (202 Bce-220 Ce)". Journal of Archaeological Science, 2025;180. Elsevier - www.elsevier.com
  • Qinghui Li, et al. "Ftir Spectra Combined With Machine Learning To Reveal the Amber Trade During the Han Dynasty (202 Bce-220 Ce)" Journal of Archaeological Science, 2025;180. doi: 10.1016/j.jas.2025.01.001.

Read more