Excellent paper recently published: Using a 3D heat map to explore the diverse correlations among elements and mineral species exploiting the open API to the superb Mindat, one of the largest mineral databases in the world. To explore new associations and knowledge hidden in the big geoscience literature data. They conclude "This study demonstrates how... Continue Reading →

A couple of weeks ago Lin et al (2023) unveiled their geoscience fine tuned Large Language Model (LLM) a 30B parameter geoscience fine tuned version of Meta AI's 'OpenSource' Galactica LLM - to create GeoGalactica. This was as part of the Deep-time Digital Earth (DDE) initiative funded by NSF China. They scraped over 6 million... Continue Reading →

Using AI to detect Natural Hydrogen: Back in 2021 I worked on a project using text mining applied to old oil & gas reports to detect both explicit and implicit clues for overlooked H2 occurrences. I found it interesting to read a recent paper published last few weeks by Herreid et al (2023) from the... Continue Reading →

It is 100 years since Paul Otlet and Henri La Fontaine created the ‘Mundaneum’ - the prophetic conceptual precursor of today’s Internet.  The utopian Mundaneum (renamed from the Palais Mondial in 1924) was essentially a ‘Google by telegram'. It has been described by Le Monde as ‘A paper Google’, by the New York Times as... Continue Reading →

I was invited to No 10 Downing Street today for the winter reception. It was lovely to meet other guests from different business sectors hosted by the Chancellor of the Exchequer. A lot of interest in the subsurface, geoscience, data and AI. Merry Christmas and Happy New Year everyone. #geoscience #digital #technology #artificialintelligence #business #government

Link to presentation here: https://www.youtube.com/watch?v=sR96AGqNxhM&list=PLKBhokJ0qd39DKVumBWmeD0y17cJa_4n5&index=9

I've been assessing the potential of using patterns of words in large volumes of text to map geology. A hypothesis could be that there are subtle word association patterns in reports that might be useful in some way for geoscience. Perhaps by impacting uncertainty in our existing models or highlight differences that may warrant further... Continue Reading →

Congratulations to all those that took part in the first subsurface Large Language Model (LLM) Hackathon last week using data from the UK, Norway and The Netherlands government repositories. The event coincided with the anniversary 1 year ago of the launch of ChatGPT. This is an area where the first practical deployments of LLM's are... Continue Reading →

I've now publicly shared the deep learning model to detect ammonites which you can use on a Smartphone. This was built by labelling 300 images of ammonites (over 800 annotations in total) using Datature's platform free trial version. See my previous posts for more details. It is meant as a bit of fun to perhaps... Continue Reading →

I believe this may be the first research published on what geoscientists think of Generative AI responses. The experiment tested the impact of enriching text chunks generated from 100 public domain geoscience reports using Retrieval Augmented Generation (RAG). The tagging had the effect of influencing the top text chunk candidates from the vector database used... Continue Reading →