This course will be delivered ONLINE: 20 hours of online live lessons, plus 10 hours of recorded classes and assignments. A good internet connection is required to follow the course.
This course introduces students to the most advanced tools in Artificial Intelligence (AI); machine learning methods that make data mining and data processing a fascinating topic.
Obtaining and analyzing data is currently a very well developed field in computer science. Finding patterns in these data, or processing this information, is less straightforward and is sometimes subjected to biases. Data Mining has recently given way to Process Mining, in which powerful statistical and software tools are used in combination to correctly detect patterns and make reliable classifications of customers or products and make accurate predictions. For Paleobiology, these tools provide the most advanced computing technique for accurate classification and prediction.
This course offers a practical introduction to Machine Learning applied to Palaeontology and Archaeology. From class One, students will learn the use of these information-managing tools on their computers. After its completion, students will be prepared to understand the patterns hidden in any database, regardless of its size and complexity. For a practical demonstration, two types of taphonomic fields will be provided.
The study of bone surface modifications (BSM) has been one of the most difficult and controversial areas in taphonomic research. Only AI has provided a way to understands the subtleties of this type of analysis by yielding systematic identification rates of BSM with accuracy higher than 90% of the cases. This constitutes a major revolution in this field.
The second taphonomic field is biometric. As a practicum, metric properties of broken bones will be used to discern process (dry and green breaking) and agency (human or carnivore) in bone fragmentation.
Teaching will be done using R. In the last module involving computer vision and deep learning, both R and Python will be used.
Although students will benefit from having prior knowledge on statistics (namely, univariate and bivariate or multivariate statistics), the teaching system will not require them to have any statistical basis. Concepts will be explained from their basic foundation so that they are fully understood by students with different backgrounds.
All participants must have a personal computer (Windows, Macintosh), with webcam if possible, and a good internet connection.
Required textbook: James, G., Witten, D., Hastie, T., Tibshirani, R., 2013. An Introduction to Statistical Learning with applications in R. Springer. A pdf version is available for free HERE.