Classifying Genre in Historical Medical Periodicals
Next in line: Vera Danilova presents her work on genre classification in digitized periodicals from European patient organizations (1951–1990) using #LLMs as part of the #ActDisease project.
XLM-RoBERTa (UDM) led Q&A tasks with 32% more correct answers than mBERT/hmBERT. hmBERT (UDM) topped Administrative classification (+16%) CORE-based models excelled in legal genre prediction.
https://www.europesays.com/1952170/ An automated parallel genetic algorithm with parametric adaptation for distributed data analysis #Classification #COMPUTERSCIENCE #Data #GAsParameterControl #GeneticAlgorithms(GAs) #HumanitiesAndSocialSciences #LargeScaleDataProcessing #multidisciplinary #science #ScientificData #Spark
Pipeline release! nf-core/taxprofiler v1.2.3 - v1.2.3 - Bouncy Basenji Patch!
Please see the changelog: https://github.com/nf-core/taxprofiler/releases/tag/1.2.3
I have to level with you. I would not be considering writing and running a #socialWeb platform for #writers if I did not have a few itches to scratch.
context: https://fosstodon.org/@travisfw/114099690246606913
One itch is #classification. #Hashtags are nicely nonlinear. They are simple, and for being simple, they are effective. But, in giving authors direct access to declare the index their work appears in, you throw out all of the potential benefits modern approaches to indexing and search could yield. EG: stemming.
Who's working the most higher-order #classification systems of things in the world? Systematics in biology work on classifying all living things. Every field has their own classification systems. Is there a field that integrates those classification systems to create a "taxonomy of everything"? When I search for that concept I'm pointed towards people working on #SemanticWeb - I'm curious if there are any good sources out there to read up on, especially actual attempts/examples! #taxonomy
Version 1.3 of the automated subject subject indexing tool #Annif has been released!
This release introduces support for the EstNLTK analyzer for better Estonian lemmatization 
, optimizations to the MLLM backend, as well as maintenance and bug fixes, including better file permissions in multi-user environments.
New review: How is the living world organized and classified? Encyclopaedic in scope yet compelling to read, the deeply researched Kingdoms, Empires, & Domains is an 851-page behemoth that surveys 26 centuries of our changing understanding.
#Books #BookReview #Bookstodon #Scicomm #Classification #Genetics #HistoryOfScience #ScienceHistory #HistSci #Microbiology #Phylogenetics #ScalaNaturae #Taxonomy #Zoophyta @bookstodon
Pipeline release! nf-core/taxprofiler v1.2.2 - v1.2.2 - Bouncy Basenji Patch!
Please see the changelog: https://github.com/nf-core/taxprofiler/releases/tag/1.2.2
23 days this bad boy took me, all the way over Christmas and New Years to this weekend. (Yes, that's my idea of having a good time)
Review coming soon https://inquisitivebiologist.com
Fresh from the press : our #IROS2024 paper on #Terrain #Classification using proprioceptive #sensor #data is now out on IEEEXplore !
Calculating Empires – A Genealogy of Technology and Power since 1500
https://calculatingempires.net/
Check out this astonishing large-scale research visualization that explores how technical & social structures co-evolved over five centuries - focusing on 4 themes: communication, computation, classification, & control. Definitely use your largest screen!
"We could classify any area of math we think is leading in a bad direction to make it a state secret and "it will end"."
My decomposition is a state secret.
Academia is 13 years late.
Pipeline release! nf-core/taxprofiler v1.2.1 - v1.2.1 - Bouncy Basenji Patch [2024-11-28]!
Please see the changelog: https://github.com/nf-core/taxprofiler/releases/tag/1.2.1
Annotating Arcane Screenshots using Azure OpenAI, The Hard Way
To make a long story short, Arcane ended yesterday, and during some idle browsing I came across a pretty amazing set of wallpapers that someone generated by taking 4K frames and up(...)
#aiohttp #arcane #azure #captioning #classification #gpt4o #image #labeling #openai #python #rest #tagging
The #LibraryOfCongress #classification manual is hilarious. TR717.3 is photography of machinery, special subjects. The three special subjects listed are airplanes, automobiles, and sex machines.
But the manual instructs that collections of photographs in certain fields be classed with those fields. That means all your books of photos of sex machines have to go in the sex machine section, while books about how to photograph sex machines go in TR717.3.S49.
The main OEIS pages about my decomposition are no longer indexed on Google:
User 
https://oeis.org/wiki/User:R%C3%A9mi_Eismann
#decompwlj https://oeis.org/wiki/Decomposition_into_weight_*_level_%2B_jump
A117078 https://oeis.org/A117078
It's a new fundamental theorem of arithmetic and academia is 11 years late.
A short story about my decomposition 
The recording and slides for my "What's New in Skosmos 3" meetup presentation have been published on the ISKO UK event page: https://iskouk.org/event-5636362
Direct link to recording in YouTube: https://youtu.be/2mLf087KfZc
Today I will present "What's New in Skosmos 3" 17:30 UTC (18:30 BST) at the ISKO UK online Meetup. Free registration here:
https://iskouk.org/event-5636362
Skosmos is an open source tool for publishing SKOS controlled vocabularies on the web.
The upcoming Skosmos 3 release, focuses on reimplementing the user interface with modern frameworks. I will demonstrate the current status of Skosmos 3, discuss its impact on deployments and end users.
