[Sigmetrics] 1findr: research discovery & analytics platform
David Wojick
dwojick at craigellachie.us
Wed Apr 25 11:28:16 EDT 2018
Well, Kevin, having done quite a bit of research on this issue, I stand by
my claim. It is almost impossible to write a research report on X without
using the language that describes X. In fact this is why search works.
Moreover, language provides a lot more information than citations. Say 1000
words versus 20 citations.
David
At 10:48 AM 4/25/2018, Kevin Boyack wrote:
>Hi David,
>
>I think you state ideals here that do not reflect the current reality.
>Language is not more objective than citations; it is just as ambiguous.
>Authors choose the words they use just as much as they do the papers they
>cite. Specific language is an ideal that is not met often enough,
>particularly when we consider that over half of the scientific literature
>written in English is authored by non-native English speakers.
>
>Having said that, I do think that full text is a potential gold mine
>(despite the fact that the FUSE program didn’t find the silver bullet
>with tens of millions of dollars), and that we will ultimately use it to
>learn things that we can’t learn from citation analysis. I view it as
>complementary and very valuable, but not “far superior†to citation
>analysis.
>
>Cheers!
>Kevin
>
>
>
>From: SIGMETRICS <sigmetrics-bounces at asist.org> On Behalf Of David Wojick
>Sent: Wednesday, April 25, 2018 5:14 AM
>To: Éric Archambault <eric.archambault at science-metrix.com>
>Cc: sigmetrics at mail.asis.org
>Subject: Re: [Sigmetrics] 1findr: research discovery & analytics platform
>
>Yes, Eric, I think that language analysis is potentially far superior to
>citation analysis, now that full text is available. Citations were great
>when they were all we had. In addition to no lag, language is more
>objective than citations. Authors can often choose who to cite or not, but
>specific language is necessary in order to state their results.
>
>By the same token, language use is probably a better indicator of the
>extent and evolution of research areas than subjectively defined and
>applied discipline categories. New ideas almost always require new
>language, new phrases at least, sometimes even new words. As I like to put
>it, the science frontier is a language frontier. The words we use have the
>meanings they do because we are trying to say what is true. Much follows
>from this.
>
>David
>
>On Apr 25, 2018, at 6:49 AM, Éric
>Archambault<<mailto:eric.archambault at science-metrix.com>eric.archambault at science-metrix.com>
>wrote:
>Good points David.
>
>For recommenders, using bibliographic coupling is among the most powerful
>tools and articles need not use the same vocabulary. There has been much
>emphasis placed on co-citation in the bibliometric/scientometric community
>but bibliographic coupling is extremely powerful and way more convenient
>(stable and no lag contrary to co-citation analysis which needs to way for
>citations to materialize and then is always evolving as the citation graph
>builds up).
>
>Again, I don’t mean to say you’re not right to emphasize the need for
>full-text searching, but in many cases there are workaround. One case
>where full-text is more particularly useful is corpus building for e.g.
>text/data mining and literature related discovery (LRD) studies.
>
>Éric
>
>Eric Archambault, PhD
>CEO | Chef de la direction
>C. 1.514.518.0823
><mailto:eric.archambault at science-metrix.com>eric.archambault at science-metrix.com
><http://www.science-metrix.com/>science-metrix.com &
><http://www.science-metrix.com/>1science.com
>
>From: David Wojick
><<mailto:dwojick at craigellachie.us>dwojick at craigellachie.us>
>Sent: April-24-18 5:41 PM
>To: Éric Archambault
><<mailto:eric.archambault at science-metrix.com>eric.archambault at science-metrix.com>
>Cc: <mailto:sigmetrics at mail.asis.org>sigmetrics at mail.asis.org
>Subject: Re: [Sigmetrics] 1findr: research discovery & analytics platform
>
>I agree up to a point, Eric. Metadata (especially including an abstract)
>is usually sufficient for what we might call a standard search. This is
>one where what we are looking for is the central topic of the article.
>
>But there are many other sorts of search, where the thing sought is
>relatively secondary to the article and here only full text search works.
>Examples might include those climate change articles that rely on a
>specific model, or nuclear physics that uses the Monte Carlo method. A
>great many questions of this form can arise, in research and in science
>metrics.
>
>
>
>Then too there is the powerful "more like this" (MLT) function which
>requires full text. This finds closely related research that does not use
>the same language. An example is author disambiguation versus name
>identity. Google Scholar's version of MLT is very useful.
>
>
>
>In fact I developed an algorithm for DOE OSTI that uses "more like this"
>technology to find all and only those articles closely related to a given
>topic, ranked by closeness. When you get full text will be happy to show
>it to you.
>
>
>
>But the fact that your system does not do everything now is not a
>criticism, merely a direction for possible progress.
>
>
>
>Best of luck,
>
>
>
>David
>
>On Apr 24, 2018, at 4:18 PM, Éric
>Archambault<<mailto:eric.archambault at science-metrix.com>eric.archambault at science-metrix.com>
>wrote:
>David,
>
>Thanks for your encouraging comments. You are right, we don’t do full
>text indexing search – yet. We want to get there though as a
>bibliometrician I have always been a tad skeptical about the need to go
>much beyond high quality metadata. When you can’t find a paper and you
>have title, journal, abstract, references/citations, chances are the paper
>won’t be all that sharp for most of the mainstream applications.
>Probably the term is not that key if it can’t be found anywhere in the
>metadata. I’m not saying there are absolutely no cases for searching in
>the metadata but most of the people want sharp results, and though we are
>all impressed by zillions of results, we rarely if ever use the long tail.
>This only became stronger with Google that made us lazy, naïve, and not
>curious enough. 1findr is not perfect as it is, but it presents a nice
>compromise being sharp and being extensive enough. But duly noted we may
>miss a few diamonds, and have a shorter tail in our results.
>
>Over time, we hope to have more publishers helping us built a high quality
>full-text. We’ve started with Karger who likes to think outside the box.
>We’ve started experimenting with the Frontiers corpus as well. This is
>still small scale but we are careful and reflective about our development.
>Once we’ll have determined the investments in technical complexity and
>index size is worth our while to improve the user experience, we’ll
>start deploying full-text indexing on a larger but progressive scale, at
>least for those publishers who want their material to be discoverable to
>the maximum extent.
>
>
>Éric
>
>
>Eric Archambault, PhD
>CEO | Chef de la direction
>1335, Mont-Royal E
>Montréal QC Canada H2J 1Y6
>
>T. 1.514.495.6505 x.111
>C. 1.514.518.0823
><mailto:eric.archambault at science-metrix.com>eric.archambault at science-metrix.com
><http://www.science-metrix.com/>science-metrix.com &
><http://www.science-metrix.com/>1science.com
><image003.png> <image004.png>
>
>From: SIGMETRICS
><<mailto:sigmetrics-bounces at asist.org>sigmetrics-bounces at asist.org> On
>Behalf Of David Wojick
>Sent: April 24, 2018 2:16 PM
>To: <mailto:sigmetrics at mail.asis.org>sigmetrics at mail.asis.org
>Subject: Re: [Sigmetrics] 1findr: research discovery & analytics platform
>
>It appears not to be doing full text search, which is a significant
>limitation. I did a search on "chaotic" for 2018 and got 527 hits. Almost
>all had the term in the title and almost all of the remainder had it in
>the abstract. Normally with full text, those with the term only in the
>text are many times more than those with it in title, often orders of
>magnitude more.
>
>But the scope is impressive, as is the ability to filter for OA.
>
>David
>
>David Wojick, Ph.D.
>Formerly Senior Consultant for Innovation
>DOE OSTI <https://www.osti.gov/>https://www.osti.gov/
>
>
>At 08:00 AM 4/24/2018, you wrote:
>
>
>
>Content-Language: en-US
>Content-Type: multipart/related;
> type="multipart/alternative";
> boundary="----=_NextPart_001_00EE_01D3DBBD.BC977220"
>
>Greetings everyone,
>
>Today, 1science announced the official launch of 1findr, its platform for
>research discovery and analytics. Indexing 90 million articlesÂof which
>27 million are available in OAÂit represents the largest curated
>collection worldwide of scholarly research. The platform aims to include
>all articles published in peer-reviewed journals, in all fields of
>research, in all languages and from every country.
>
>Here are a few resources if you’re interested in learning more:
>
>• Access 1findr platform: <http://www.1findr.com>www.1findr.com
>• p; Visit the 1findr website:
><http://www.1science.com/1findr>www.1science.com/1findr
>• Send in your questions:
><mailto:1findr at 1science.com>1findr at 1science.com
>• See the press release:
><http://www.1science.com/1findr-public-launch>www.1science.com/1findr-public-launch
>
>
>Sincerely,
>
>Grégoire
>
>Grégoire Côté
>President | Président
>Science-Metrix
>1335, Mont-Royal E
>Montréal, QC H2J 1Y6
>Canada
>
><https://www.linkedin.com/company/science-metrix-inc><image001.png><https://twitter.com/ScienceMetrix><image002.png>
>
>T. 1.514.495.6505 x115
>T. 1.800.994.4761
>F. 1.514.495.6523
><mailto:gregoire.cote at science-metrix.com>gregoire.cote at science-metrix.com
>www.science-metrix.com
>
>
>
>
>Content-Type: image/png;
> name="image001.png"
>Content-Description: image001.png
>Content-Disposition: inline;
> creation-date=Tue, 24 Apr 2018 12:00:30 GMT;
> modification-date=Tue, 24 Apr 2018 12:00:30 GMT;
> filename="image001.png";
> size=1068
>Content-ID:
><<mailto:image001.png at 01D3DB57.02A76980>image001.png at 01D3DB57.02A76980>
>
>Content-Type: image/png;
> name="image002.png"
>Content-Description: image002.png
>Content-Disposition: inline;
> creation-date=Tue, 24 Apr 2018 12:00:30 GMT;
> modification-date=Tue, 24 Apr 2018 12:00:30 GMT;
> filename="image002.png";
> size=1109
>Content-ID:
><<mailto:image002.png at 01D3DB57.02A76980>image002.png at 01D3DB57.02A76980>
>
>
>_______________________________________________
>SIGMETRICS mailing list
><mailto:SIGMETRICS at mail.asis.org>SIGMETRICS at mail.asis.org
>http://mail.asis.org/mailman/listinfo/sigmetrics
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.asis.org/pipermail/sigmetrics/attachments/20180425/c2232203/attachment.html>
More information about the SIGMETRICS
mailing list