When people ask me what I want to do with my life, I tell them that I want to be a computational archaeologist. But what exactly is Computational Archaeology? No, I do not want to dig computers out of the ground, like some thought when I originally mentioned the term. In short, it refers to the development or use of software or computer devices to aid archaeologists in their daily activities.
Many people think of Indiana Jones or Lara Croft when the word archaeologist is mentioned. As exciting as running away from a giant rolling rock in order to grab a treasure which is located between two streams of fire emerging from the statues of two dragons that shoot arrows out of their eyes sounds, archaeology is much calmer and involves a lot more than being fit.
The archaeological process can be divided into five main steps. These steps are:
- Discovery and Investigation
- Post-excavation analysis and lab work
- Dating Discoveries
- Reporting and Interpretation
We are all familiar with excavation, but as it can be seen above, this is only 1/5th of the archaeological process. A large amount of documentation takes place during each step. This is mainly where computers come in.
Computational archaeology can include a large number of things, such as:
- Data storage
- Data organisation
- 3D modelling of ruins
- Data Mining (determining similarities between data pattern which may previously have been unknown)
- Simulations of human behaviour
- Data representation
- and many more
My main focus is Computational Intelligence. This refers to a sub-branch of artificial intelligence which consists on a number of algorithms capable of learning and adapting to new environments. These algorithms often model biological processes or social behaviour of various living organisms.
How is this useful to archaeology? Well, as archaeologists deal with large numbers of data daily they may easily miss small similarities between certain data patterns due to having to look at too much information. Computational Intelligence algorithms can be used to mine this data and determine the similarities or differences as well as to discover trends among the data. Computational Intelligence algorithms can also be used to train system do certain things. The most common application of this, which is highly relevant to archaeology, is training a system to classify data patterns. For example, an archaeologist may classify pots into 3 different classes (A, B and C). He may train a system to classify pots correctly according to a set of attributes. Once the system is trained, he may then present new pots to it and it can determine what class the new pots belong to. This may be useful for cases where many objects need to be classified, or the differences between objects are very small for a human to determine.
My dream is to one day program robots that will be used in marine archaeology to help discover the past that the sea hides.