A Geodyssey – Geoscience Text Analytics and Enterprise Search Research
A voluntary non-commercial research blog since 2015
Panning for fossils. Standing in a river digging the gravel with a shovel to sieve for fossils is a fun hobby. This is the latest hoard. Megalodon tooth in the center (7cm) to the left are various shark teeth (such as Lemon, Tiger, Sand Tiger), towards bottom left Stingray (and Eagle Ray) tooth plates. To... Continue Reading →
The video of my talk at the Artificial Intelligence for Geological Modelling and Mapping Conference at the University of Exeter back in May is now on YouTube. You can find it on the link below in the comments, along with the other excellent talks on this topic, in a superbly organised conference by Charlie Kirkwood... Continue Reading →
NASA and IBM Large Language Models for Earth Science (INDUS). The INDUS encoders were trained on a corpus of 60 billion tokens encompassing astrophysics, planetary science, Earth science, heliophysics, biological, and physical sciences data. According to IBM/NASA the models are freely available on Huggingface and they will be releasing benchmark datasets.Under the hood, from an Earth... Continue Reading →
I gave an overview of ethical AI for chatbots during this months SPE Data Science and Engineering Analytics Technical Section (DSEATS) Advisory Board. If you wish to find out more about DSEATS and its activities, link here: https://connect.spe.org/dsea/home
Artificial intelligence-driven assessment of salt caverns for underground hydrogen storage: Interesting paper from Derakhshani et al (2024) recently published, using various raster images and machine learning algorithms to predict what might be the most suitable areas for underground geological hydrogen storage in Poland (focusing on salt caverns). Derakhshani R, Lankof L, GhasemiNejad A, Zaresefat M.... Continue Reading →
Interesting article in The Guardian today. They picked up the Deep-time Digital Earth (DDE) Chinese chatbot article in the "Geoscientist" from last week, and have conducted some further research with geologists. Link to the article in The Guardian here: https://www.theguardian.com/technology/article/2024/jun/24/geologists-censorship-bias-chinese-chatbot-geogpt #Geology #Geoscience #ArtificialIntelligence #AI #LLM #GeoGPT #DDE #IUGS
My letter in the 'Geoscientist' raises serious concerns about Artificial Intelligence from the Deep-time Digital Earth (DDE) project, around lack of transparency, state censorship, and potential copyright infringement.DDE is an international geoscience Big Data and AI project endorsed by the International Union of Geological Sciences (IUGS) and led by China. The DDE accords were signed... Continue Reading →
Excellent paper published this month by Hinsby et al (2024) abstract below:Open access to harmonised digital data describing Earth’s surface and subsurface holds immense value for society. This paper highlights the significance of open access to digital geoscience data ranging from the shallow topsoil or seabed to depths of 5 km.Such data play a pivotal role... Continue Reading →
Issues can arise when chatbots intended for an international audience are based on current Chinese Large Language Models (LLM) which are state censored. These models include traits identified in testing such as refusing to answer certain questions, deflection by broadening a question, omission, and propaganda. These models, such as potentially Alibaba's "Qwen" are available in... Continue Reading →
At the University of Exeter Institute for Data Science and Artificial Intelligence (part of Turing Institute) at the AI for Geological Modelling and Mapping conference today. Great conference, discusions, excellent presentations from: Kristine Asch, BGR, Germany – Geology without frontiers: The pioneer project of the first digital International Geological Map of Europe (IGME 5000)Charlie Kirkwood,... Continue Reading →
A Geodyssey - Geoscience Text Analytics and Enterprise Search Research 