Use of AI in Archaeology and its Impact.

Artificial Intelligence is being used in many fields nowadays and it has made the work of humans easier. AI is the ability of a machine to have human-like abilities like learning, reasoning, planning, and creativity. It enables the system to deal with and solve problems to achieve a specific goal. AI systems can adapt their behavior by analyzing the effect of previous actions and then working almost independently.

The field we would discuss is archaeology and the use of AI in archaeology. Archaeology is a profession where people perform excavations and study remains and fossils to understand past civilizations. Archaeologists are responsible for preserving ancient ruins and artifacts. They use their findings to explain the behavior of human beings in the past and try to understand the evolution of the species.

What are the types of archaeologists?

There are different fields in archaeology and the field you chose will determine what type of archaeologist you would be. Following are the most common types of archaeologists:

Environmental archaeologists:

Environmental archaeologists study the relationship between the natural environment and ancient people. They try to determine the animals and plants that were present during a time period and how the people of that time used them in their lives.

Underwater archaeologists

These archaeologists study the underwater remains and the evidence of shipwrecks buried underwater. The archaeologist must be familiar with the techniques needed to perform underwater operations.

Historical archaeologists

Historical archaeologists study the written evidence and the preserved artifacts about a particular time period in history. They study trade, industries, and cultural histories.

Classical archaeologists

The classical archaeologist is focused on the history that relates to Rome and Greece and how these ancient civilizations impacted the nearby cultures and other areas in the world.

Ethnoarchaeologists

They are concerned with finding clues on the habits and cultures of people. Ethnoarchaeologists study an ancient culture in a region and compare it to the way the modern people live in that area. It also involves examining how modern groups of individuals live and extrapolate to find how ancient people lived.

Use of AI in archaeology

Artificial intelligence is being used by archaeologists to find new sites for digging and making discoveries, thereby increasing the pace of research. AI and computer algorithms are being used to analyze satellite image data and automate the process of finding archaeological sites.

Due to the accumulation of image data collected through satellites, planes, and drones, they can check the area on Earth for possible archaeological excavation without visiting the site. This makes the work quicker and efficient as compared to manually going through images which is time-consuming. Some of the places where archaeology has been used are given in examples below:

1) By leveraging computer vision algorithms, archaeologists are analyzing satellite imagery and data from drones and planes. Many unidentified archaeological mound features evade detection due to heavily forested canopies that occupy a large region, making pedestrian and aerial surveys difficult.

To meet this challenge, object-based image analysis (OBIA) for analyzing light and radar and LiDAR (a 3D scanning technology ) data have been developed. The LiDAR pulses make maps of the forest floor containing information on the size, shape, and slope.

This was possible as the archaeological team taught the computer the shape and texture of the mounds, so it could identify potential sites from the 3D data that is analyzed. Over a course of a year, the AI was able to identify over 70 sites covering an area of more than 1000 square kilometers. This saved archaeologists the equivalent of several years of manual labor that would have been required for analyzing the LiDAR image by hand.

Archaeologists are forever looking at new ways to enhance the speed of the identification of archaeological sites. The rise in the sea level has threatened many potential archaeological sites with destruction and other climate change impacts, like deforestation, construction, and other activities. Therefore, machine learning is proving to be useful in archaeological research as traditional methods take several years.

2) The AI models developed to enhance archaeological research can have application beyond learning the history and culture of the early civilizations. Researchers at the ICAC used machine learning algorithms to reconstruct more than 20,000 kilometers of paleo-rivers in the Indus Valley Civilization which falls in Pakistan and India. The mapping would not be possible with the help of satellite images alone. This helped to locate many unknown archaeological sites and gave new insights into the Bronze Age cultures.

Moreover, analytics can help the government with water resource management. It can also help and provide information on important aspects, such as the variability of agricultural productivity and the influence of climate change on the expansion of deserts.

3) Artificial intelligence methods are being used to understand the chemical makeup of pottery and ceramics. It can help in understanding the past trading networks by identifying where the materials were made and how far they could be transported.

A team of Mexican archaeologist persons from the Philipps University of Marburg has shown the potential of AI in determining the origin of archaeological finds. Pieces of Obsidian stone glass were found in Xalasco in western Mexico. There were indications that the inhabitants of this pre-Columbian site had contacts with other regions of Mexico, which also included the exchange of raw materials.

The team examined the chemical composition of the artifacts with a portable X-ray fluorescence spectrometer, with which they examined 256 obsidian finds from Xalasco after obtaining samples from several Central American sites.

The team used machine learning and with the help of AI, the obsidian workpieces were classified into separate groups based on the origin of raw material. The findings suggest that the Xalasco culture had a certain exchange of materials from distant settlements, but most of the workpieces from Xalasco are made of material mined from near the site. This was discovered with the help of AI.

4) Using AI, we can learn when and where the languages evolved around the world, and this is being used by linguistic anthropologists. AI has helped to reconstruct ancient texts. Google’s DeepMind used a deep neural network called PYTHIA to find out missing inscriptions in Greek from damaged surfaces of objects made of ceramics. It takes a sequence of damaged text as input and predicts character sequences.

Two Israeli Universities employed artificial intelligence to restore broken words written in Akkadian language which is deemed as an extinct language, spoken in Mesopotamian empires of Babylon and Assyria some 2500 years before Christ and up to 600 BC.

The researcher team digitized clay tablets written in cuneiform, using a tool called Babylonian Engine. The digitized texts were analyzed and studied by machine learning algorithms. This helped in restoring damaged texts by figuring out the words to use in the missing sections.

In the same way, Chinese scientists applied a convolutional neural network (CNN) to find out about another ancient tongue found on turtle shells and ox bones. It was able to classify bone morphology and piece the fragments together that had inscriptions showing evidence of the earliest recorded Chinese history.

5) AI has been used to differentiate toolsets from the Middle and the Late Stone Ages. Researchers from the University of Liverpool have used artificial neural networks, the function of which is based on machine learning. There was a change in the culture of our ancestors during the transition from the Middle to the Late Stone Age.

It was difficult to assign archaeological tools of these periods. The researchers examined the presence and absence of certain tools, in tools sets to assign them to the Middle or Late Stone Age. For this, 16 different tool types from 92 toolsets from Africa were analyzed.

Instead of focusing on individual tools, the focus was on toolsets and certain tool shapes that often appeared together in sets. The artificial neural network was used to help to find out similarities and differences. A breakthrough was made that enabled them to see various patterns in the composition of tool kits that differed in the Middle and Late Stone Ages.

How to become an archaeologist?

The first step for the persons who aspire to be archaeologists is to complete a bachelor’s program in anthropology or a related field such a history or geography. Many budding archaeologists choose to participate in the internship programs during or after they complete a bachelor’s degree. Places that may offer internships include museums, government agencies, and archaeological organizations.

Though the entry-level archaeological positions only require a bachelor’s degree, most of the archaeological jobs require a master’s degree. The master’s degree offered in this profession includes anthropology and archaeology programs. Depending on the type of archaeologist you want to become, you can begin to seek employment.

We have considered a few examples where artificial intelligence is helping us in extracting information from past ruins. Machine learning algorithms help to parse through complex data quickly and result in solving problems and getting the work done faster. AI can’t replace humans as it lacks the subjective expertise of a human being but at the same time, it helps in solving many problems faced by archaeologists.

References :

https://www.analyticsinsight.net/artificial-intelligence-in-archaeology-can-it-be-the-next-indiana-jones/

https://www.unite.ai/ai-is-dramatically-changing-archaeology-discovering-new-sites-and-artifacts/