Hans Ket
@hansket.bsky.social
Medical Information Specialist, Medical Library, Vrije Universiteit Amsterdam. Find my co-authored publications on ORCID: https://orcid.org/0000-0002-1909-3150 or on PubMed: https://pubmed.ncbi.nlm.nih.gov/?term=ket+h+OR+ket+jc&sort=pubdate&sort_order=asc
Reposted by Hans Ket
Reporting databases with their platforms is a MINIMUM REQUIREMENT of PRISMA and PRISMA-S.
December 22, 2025 at 3:27 PM
Reporting databases with their platforms is a MINIMUM REQUIREMENT of PRISMA and PRISMA-S.
Reposted by Hans Ket
As we wrap up 2025, I wanted to share some of the videos I've made that strike on my pet peeves in #LiteratureReviews, #SystematicReviews, #searching and #EvidenceSynthesis.
A thread. 🧵 #medlibs
First up, report your databases with their platforms correctly.
youtu.be/lSn3IVxGG7M?...
A thread. 🧵 #medlibs
First up, report your databases with their platforms correctly.
youtu.be/lSn3IVxGG7M?...
Report Your Databases With Their Platforms | Methods Monday | SR Education
YouTube video by Carrie Price
youtu.be
December 22, 2025 at 3:26 PM
As we wrap up 2025, I wanted to share some of the videos I've made that strike on my pet peeves in #LiteratureReviews, #SystematicReviews, #searching and #EvidenceSynthesis.
A thread. 🧵 #medlibs
First up, report your databases with their platforms correctly.
youtu.be/lSn3IVxGG7M?...
A thread. 🧵 #medlibs
First up, report your databases with their platforms correctly.
youtu.be/lSn3IVxGG7M?...
Reposted by Hans Ket
A fallacy in my thinking. I assumed RAG properly done will reduce ghost ref to near zero. But this is not true if your source you do retrieval over can get contaminated which is easy if you just search the web! Elicit, Consensus searching indexes like Semantic Scholar should be less vulnerable
Eg of chatgpt instant saying a [Citation] Google Scholar entry exists by finding a paper reference chatgpt.com/share/69466f... . Here's a stronger model saying its likely a ghost reference chatgpt.com/share/694670...
ChatGPT - Education governance paper
Shared via ChatGPT
chatgpt.com
December 20, 2025 at 10:45 AM
A fallacy in my thinking. I assumed RAG properly done will reduce ghost ref to near zero. But this is not true if your source you do retrieval over can get contaminated which is easy if you just search the web! Elicit, Consensus searching indexes like Semantic Scholar should be less vulnerable
Reposted by Hans Ket
"When health librarians step into the light, everyone in health service benefits, especially patients." 'Light a Big Fire', a terric call to action in this guest blog by Bennery Rickard @hselive.bsky.social.
I've been called many things; I'm honoured by this mention
www.libfocus.com/2025/12/hse-...
I've been called many things; I'm honoured by this mention
www.libfocus.com/2025/12/hse-...
HSE Library Day 24th September 2025: Light a big fire
This guest post is by Bennery Rickard, Area Manager in the HSE Library East. Health Professionals in Galway University Hospital. Image from ...
www.libfocus.com
December 20, 2025 at 11:26 AM
"When health librarians step into the light, everyone in health service benefits, especially patients." 'Light a Big Fire', a terric call to action in this guest blog by Bennery Rickard @hselive.bsky.social.
I've been called many things; I'm honoured by this mention
www.libfocus.com/2025/12/hse-...
I've been called many things; I'm honoured by this mention
www.libfocus.com/2025/12/hse-...
Reposted by Hans Ket
I am constantly analysing in an informal way & analysing to work out why my systematic searches are doing what they’re doing. But have not done analysis as the later stage in terms of analysing primary or secondary quantitative or qualitative data. I look forward to doing this #ResearchAdvent #Day19
December 19, 2025 at 11:17 PM
I am constantly analysing in an informal way & analysing to work out why my systematic searches are doing what they’re doing. But have not done analysis as the later stage in terms of analysing primary or secondary quantitative or qualitative data. I look forward to doing this #ResearchAdvent #Day19
Reposted by Hans Ket
The ACM Digital Library, where a LOT of computing-related research is published (I'd say at least 75% of my own publications), is now not only providing (without consent of the authors and without opt-in by readers) AI-generated summaries of papers, but they appear as the *default* over abstracts.
December 16, 2025 at 11:31 PM
The ACM Digital Library, where a LOT of computing-related research is published (I'd say at least 75% of my own publications), is now not only providing (without consent of the authors and without opt-in by readers) AI-generated summaries of papers, but they appear as the *default* over abstracts.
Reposted by Hans Ket
we knew that the Google Scholar enshittification was coming...
telling students to not cite papers, chapters, books, etc they haven't read has always been ab ensuring references' accuracy & appropriateness - now, apparently, there's another reason: to make sure references exist in the first place
telling students to not cite papers, chapters, books, etc they haven't read has always been ab ensuring references' accuracy & appropriateness - now, apparently, there's another reason: to make sure references exist in the first place
Grading and googling hallucinated citations, as one does nowadays, and now that LLMs have been around for a while, I've discovered new horrors: hallucinated journals are now appearing in Google Scholar with dozens of citations bc so many people are citing these fake things
December 15, 2025 at 9:31 PM
we knew that the Google Scholar enshittification was coming...
telling students to not cite papers, chapters, books, etc they haven't read has always been ab ensuring references' accuracy & appropriateness - now, apparently, there's another reason: to make sure references exist in the first place
telling students to not cite papers, chapters, books, etc they haven't read has always been ab ensuring references' accuracy & appropriateness - now, apparently, there's another reason: to make sure references exist in the first place
Reposted by Hans Ket
New 📄! Scoping Review of Transformative Agreement Research in Evidence Based Library and Information Practice dx.doi.org/10.18438/ebl....
Scoping Review of Transformative Agreement Research
| Evidence Based Library and Information Practice
dx.doi.org
December 15, 2025 at 10:25 PM
New 📄! Scoping Review of Transformative Agreement Research in Evidence Based Library and Information Practice dx.doi.org/10.18438/ebl....
Christmas & end of year greetings from Amsterdam. "Dona nobes pacem"!
December 14, 2025 at 5:04 PM
Christmas & end of year greetings from Amsterdam. "Dona nobes pacem"!
Reposted by Hans Ket
Headline: “Exclusive: #Springer #Nature #retracts, removes nearly 40 publications that trained neural networks on ‘bonkers’ dataset.” www.thetransmitter.org/retraction/e...
Springer Nature retracts, removes nearly 40 publications
The dataset contains images of children’s faces downloaded from websites about autism, which sparked concerns at Springer Nature about consent and reliability.
www.thetransmitter.org
December 9, 2025 at 2:17 PM
Headline: “Exclusive: #Springer #Nature #retracts, removes nearly 40 publications that trained neural networks on ‘bonkers’ dataset.” www.thetransmitter.org/retraction/e...
Reposted by Hans Ket
*How to extract, analyse and present data in scoping reviews*
This JBI LIVE webinar recording provides a practical approach to extracting, analysing and presenting data within scoping reviews, with step-by-step examples.
Watch: https://ow.ly/Z42950V97JP
#EvidenceSynthesis
This JBI LIVE webinar recording provides a practical approach to extracting, analysing and presenting data within scoping reviews, with step-by-step examples.
Watch: https://ow.ly/Z42950V97JP
#EvidenceSynthesis
December 10, 2025 at 2:50 AM
*How to extract, analyse and present data in scoping reviews*
This JBI LIVE webinar recording provides a practical approach to extracting, analysing and presenting data within scoping reviews, with step-by-step examples.
Watch: https://ow.ly/Z42950V97JP
#EvidenceSynthesis
This JBI LIVE webinar recording provides a practical approach to extracting, analysing and presenting data within scoping reviews, with step-by-step examples.
Watch: https://ow.ly/Z42950V97JP
#EvidenceSynthesis
Reposted by Hans Ket
i have one query that a surprisingly large number of ai seach cant pass. And passing it doesnt mean you are the best, but likely somewhat decent. (2)
December 10, 2025 at 3:23 AM
i have one query that a surprisingly large number of ai seach cant pass. And passing it doesnt mean you are the best, but likely somewhat decent. (2)
Reposted by Hans Ket
If I had ONE tip for librarians testing "Ai search" products is focus on relevancy. If you had time for just one query , test with a "tricky query". There is no one defintion of tricky, but a query that can easily return dozens/hundreds of relevant articles is definitely pointless to test! (1)
December 10, 2025 at 3:20 AM
If I had ONE tip for librarians testing "Ai search" products is focus on relevancy. If you had time for just one query , test with a "tricky query". There is no one defintion of tricky, but a query that can easily return dozens/hundreds of relevant articles is definitely pointless to test! (1)
Reposted by Hans Ket
MeSH and PubMed Baseline Release Scheduled for January 2026
待ってました🥺🥺🙌
www.nlm.nih.gov/pubs/techbul...
#medlibs
待ってました🥺🥺🙌
www.nlm.nih.gov/pubs/techbul...
#medlibs
MeSH and PubMed Baseline Release Scheduled for January 2026. NLM Technical Bulletin. 2025 Nov–Dec
Download 2026 MeSH from links on the NLM Data Distribution page for MeSH Data the week of Jan 5, 2026. The updated MeSH 2026 DTDs are available for download at Download MeSH Data.
www.nlm.nih.gov
December 9, 2025 at 9:23 PM
MeSH and PubMed Baseline Release Scheduled for January 2026
待ってました🥺🥺🙌
www.nlm.nih.gov/pubs/techbul...
#medlibs
待ってました🥺🥺🙌
www.nlm.nih.gov/pubs/techbul...
#medlibs
Reposted by Hans Ket
The grievance blog masquerading as a "journal" founded by Trump HHS appointees (The "Journal" of the "Academy" of Public Health) is now publishing lengthy screeds about how HHS secretary RFK Jr is being treated unfairly by the BMJ and calling it a literature synthesis...
December 9, 2025 at 5:08 PM
Reposted by Hans Ket
And I often find people who feel the need to harp on this , dont actually understand the tech as well as they think. Cos half the webinar was showing use of LLM/GPT to screen for relevant papers aka "tech that generate content" IS used for evidence retrieval. Such a dumb comment. (3)
November 23, 2025 at 6:51 PM
And I often find people who feel the need to harp on this , dont actually understand the tech as well as they think. Cos half the webinar was showing use of LLM/GPT to screen for relevant papers aka "tech that generate content" IS used for evidence retrieval. Such a dumb comment. (3)
Reposted by Hans Ket
Watching webinar & "comment not a question" come in.. lecture abt diff between ML, DL vs GPT/ LLMs & says tech that generates content does not improve evidence retrieval - Do these people think they only ones who know the diff? (2)
November 23, 2025 at 6:49 PM
Watching webinar & "comment not a question" come in.. lecture abt diff between ML, DL vs GPT/ LLMs & says tech that generates content does not improve evidence retrieval - Do these people think they only ones who know the diff? (2)
Reposted by Hans Ket
More than 200 Korean papers retracted over AI use www.donga.com/en/article/a...
More than 200 Korean papers retracted over AI use
More than 200 academic papers in South Korea have been retracted over suspicions that artificial in…
www.donga.com
November 22, 2025 at 11:53 PM
More than 200 Korean papers retracted over AI use www.donga.com/en/article/a...
Reposted by Hans Ket
"An author is accountable for the content, methods, and findings of their evidence synthesis, including the decision to use AI" 10.11124/JBIES-25-00480 @cochrane.org @jbiebhc.bsky.social @envevidence.bsky.social
November 12, 2025 at 2:39 PM
"An author is accountable for the content, methods, and findings of their evidence synthesis, including the decision to use AI" 10.11124/JBIES-25-00480 @cochrane.org @jbiebhc.bsky.social @envevidence.bsky.social
Reposted by Hans Ket
General deep research are also slower cos many open a virtual browser and query it like a human. The other disadvantage using things like openai/Gemini DR is you know it isn't customized for academic in the way it does citations & it seems to create more "ghost references" for some reason (3)
November 23, 2025 at 8:52 AM
General deep research are also slower cos many open a virtual browser and query it like a human. The other disadvantage using things like openai/Gemini DR is you know it isn't customized for academic in the way it does citations & it seems to create more "ghost references" for some reason (3)
Reposted by Hans Ket
The main advantage has always been these specialised academic ai tools directly searched academic content while general deep research searches the web and needs to be prompted and be smart enough to know where to search and to prefer to cite. (2)
November 23, 2025 at 8:49 AM
The main advantage has always been these specialised academic ai tools directly searched academic content while general deep research searches the web and needs to be prompted and be smart enough to know where to search and to prefer to cite. (2)
Reposted by Hans Ket
Thinking about the business case for using specialised academic deep research eg elicit, Undermind, Consensus, Scispace, Scopus DR etc Vs general deep research (openai, Gemini, Claude research etc). (1)
November 23, 2025 at 8:47 AM
Thinking about the business case for using specialised academic deep research eg elicit, Undermind, Consensus, Scispace, Scopus DR etc Vs general deep research (openai, Gemini, Claude research etc). (1)
Reposted by Hans Ket
I've always been skeptical of "AI Scientist" claims but maybe there is something there eg Kosmos, Google Ai co scientist blog.google/feed/google-.... (3)
We’re launching a new AI system for scientists.
Today Google is launching an AI co-scientist, a new AI system built on Gemini 2.0 designed to aid scientists in creating novel hypotheses and research plans. Researchers…
blog.google
November 21, 2025 at 12:05 PM
I've always been skeptical of "AI Scientist" claims but maybe there is something there eg Kosmos, Google Ai co scientist blog.google/feed/google-.... (3)
Reposted by Hans Ket
A while ago I fell down the rabbit hole reading about Literature-based discovery while made connections in literature or suggest hypotheses & last month Elicit wrote a very detailed post on this elicit.com/blog/literat... (2)
Looking for Hidden Gems in Scientific Literature - Elicit
Scientific literature is vast and contains within it as yet unnoticed connections. Literature-based discovery is an attempt to bring them to light.
elicit.com
November 21, 2025 at 12:02 PM
A while ago I fell down the rabbit hole reading about Literature-based discovery while made connections in literature or suggest hypotheses & last month Elicit wrote a very detailed post on this elicit.com/blog/literat... (2)
Reposted by Hans Ket
I actually have quite a few posts queued up on latest trends in "ai academic search" eg mcp/connectors etc and sure all these features will make literature review more efficient but I can't help but feel we scratching the surface, remaining at the Horseless Carriage Syndrome stage (1)
November 21, 2025 at 11:58 AM
I actually have quite a few posts queued up on latest trends in "ai academic search" eg mcp/connectors etc and sure all these features will make literature review more efficient but I can't help but feel we scratching the surface, remaining at the Horseless Carriage Syndrome stage (1)