A voluntary non-commercial research blog since 2015
Using AI to describe fossils. I've been experimenting with GPT-4o to see how well it can describe fossils from smartphone photographs. At a high level it seems to perform reasonably well given most fossils found are not complete specimens.In my unscientific experiment using 10 commonly found fossils from the Jurassic Coast in England, it was... Continue Reading →
Free open-source tools: Automated petrographic image analysis by supervised and unsupervised machine learning methods. Interesting paper from Azzam, Blaise and Brigaud (2024).“We present two novel software tools: GrainSight, which utilizes a supervised deep learning model (FastSAM) for automated grain detection and morphological characterization; and PetroSeg, which employs an unsupervised segmentation approach to explore rock properties... Continue Reading →
A Geological Aware Large Language Model to be released November 14th 2024: University researchers have experimented using open-access geoscience papers to guide OpenAI’s GPT-4, a system they call “GeologyOracle”.The authors, Baucon and Neto de Carvalho, are from the University of Genova and Lisbon respectively, the latter also from UNESCO Global Geoparks. As the paper was... Continue Reading →
I've been experimenting taking a large volume of text, building embeddings and using PCA for dimensionality reduction. These data can be an input for clustering e.g. k-means. In this example I've used thousands of minerals and lithologies. I've highlighted some of the associations to illustrate. Where there are associations (complex word association co-occurrence) that are... Continue Reading →
Some great work at Queen Mary University of London by Cedric John and Harriet Dawson using machine vision to automatically detect skeletal components in carbonates."The subsurface is a fantastic resource to store and extract freshwater, clean hydrogen fuel, store carbon emissions, or even extract heat from geothermal energy. But to do this requires a meticulous... Continue Reading →
Using Large Language Models (LLM) for automated geoscience image descriptions. There are some interesting techniques being applied to geoscience information using Frontier AI.When a large volume of images exists, having automated techniques may help us geologists sift through it all, gain insights or pointers to potentially interesting features or areas that may warrant further investigation.Its... Continue Reading →
Excellent presentation yesterday using Large Language Models to query seismic interpretations and data using natural language. Nate Suurmeyer from Onward and Anna Dubovik from WAIW ran through the possibilities with their framework, using natural language to ask questions of their structured data such as geological horizons and the sea floor, with basic data and analogues.... Continue Reading →
I thought I'd plot where the readers are of my free blog on Digital Geoscience. A total of 144 countries (45 in Europe, 32 in Asia, 30 in Africa, 26 in the Americas, 11 in the Middle East). I was having some challenges with the georeferencing and some countries don't appear but its good enough!I've... Continue Reading →
Serious concerns on Deep-time Digital Earth AI: “A scientific chatbot prohibiting certain questions because they may tarnish the image of a government, poses a serious risk to academic freedom of expression”.The development of Artificial Intelligence (AI) in the geosciences continues at pace. There is much to be positive about, but with AI comes a responsibility to ensure these powerful emerging tools are... Continue Reading →
Subsurface Data: New foresight report from the UK Government Office for Science. The subsurface is a space much the same as the above ground whose use needs to be monitored, planned, and managed. But the data, technologies, and policy tools needed for this lag behind those available for the surface.Quotes on data include: "Evidence gathered... Continue Reading →
A Geodyssey - Geoscience Text Analytics and Enterprise Search Research 