Royal Library of Belgium

Sven Lieber

Hi everyone,

this last summer-break edition ties in with the previous edition. [Last
week](https://fair-data-digest.org/archive/12) I've wrote over research
to make **publication metadata** from the free publishing service
CEUR-WS **available in a FAIR fashion**, such that the data can be
explored. Today I focus on a smaller initiative to make metadata of **50
papers** available, as well as research to make metadata about
astonishing **249 million publications** available! Both are necessary.

::: {.subscription-widget-wrap attrs="{\"url\":\"https://fairdata.substack.com/subscribe?\",\"text\":\"Subscribe\",\"language\":\"en\"}" component-name="SubscribeWidgetToDOM"}
::: {.subscription-widget .show-subscribe}
::: {.preamble}
Thanks for reading FAIR Data Digest! Subscribe for free to receive new
posts and support my work.
:::

::: {.fake-input-wrapper}
::: {.fake-input}
:::

::: {.fake-button}
:::
:::
:::
:::

🔍 Usually there are many players in a data ecosystem. Anyone of those
players can contribute towards FAIR data. Recently, the organizers of
the upcoming SEMANTiCS conference
([Q59916981](https://www.wikidata.org/wiki/Q59916981)) wrote a blog post
about how they embrace **Open Science best practices**: with a
semi-automatic pipeline and manual curation. I briefly talk about their
recent blog post and link it for you.

🧪 *What's the trend of publications around a certain topic for an
institution?* Obtaining such strategically important information is
cumbersome. Information must be curated or alternatively must be
retrieved via paid services from publishing houses. But those data silos
might be difficult to access or may just be discontinued, such as the
Microsoft Academic Graph
([Q62056662](https://www.wikidata.org/wiki/Q62056662))
([source](https://www.nature.com/nature-index/news/microsoft-academic-graph-discontinued-whats-next)).
In today's Science Spotlight I present a paper about an operative **open
data platform** that makes **metadata of 249 million publications
available under the open license CC0 (thus anyone can freely use their
data)**. I will summarize their incredible work and show that nobody's
perfect.

<div>

------------------------------------------------------------------------

</div>

# 🔍 Insights: Open Science best practices at SEMANTiCS conference {#insights-open-science-best-practices-at-semantics-conference}

<div>

------------------------------------------------------------------------

</div>

Let's start small. There are many players in the academic world ranging
from conferences organized by researchers to publishers, academic
institutions or funding bodies. Anyone of those players can contribute
to a FAIR data ecosystem.

Recently the organizers of the SEMANTiCS conference announced how they
*eat their own dog food* and embrace Open Science best practices. They
created and published FAIR data about their 50 accepted papers,
involving automatic processing and manual curation.

This allowed them to create **insightful statistics**. For example,
anyone can see that 26 out of the 50 accepted papers are resource
papers, meaning that they present a *resource* such as a dataset or a
tool that was deemed relevant by the peer-reviewing. Thanks to the
statistics, interpretations can be made. According to the organizers, it
does **not seem to be widespread practice to share persistent
identifiers for code** - which becomes apparent thanks to the
statistics.

Other Semantic-Web related conferences such as ESWC
([Q17012957](https://www.wikidata.org/wiki/Q17012957)) also publish FAIR
metadata
([source](https://2022.eswc-conferences.org/conference-metadata/)). If
more conferences would do this, we would get high quality metadata **at
the source** which surely would improve the data quality of downstream
applications and data aggregation.

So far they use the data as found in the conference management system
EasyChair ([Q1278323](https://www.wikidata.org/wiki/Q1278323)).
Unfortunately, there are no author identifiers which makes downstream
tasks more difficult due to ambiguous names. See also the following item
in the newsletter to see why this could become a problem. Curious? Check
out their short blog post to learn about their semi-automatic workflow
with manual curation.

[Read the SEMANTiCS blog
post](https://semantics-org.github.io/page/news?page=2023-08-10){.button
.primary}

<div>

------------------------------------------------------------------------

</div>

# 🧪 Science Spotlight: Metadata of 249 million publications in SemOpenAlex {#science-spotlight-metadata-of-249-million-publications-in-semopenalex}

<div>

------------------------------------------------------------------------

</div>

Staying on top of the research game is difficult, because millions of
articles are published every year. Whether you look for a simple
information or more analytical data, finding it might be relevant for
strategic decisions. More than that, all the data-driven services we are
used to nowadays, rely on high quality data. **Believe it or not, since
recently you can access metadata about 249 million publications under
the open CC0 license.**

Since the de-continuation of the Microsoft Academic Graph, the
[OpenAlex](https://openalex.org/) initiative
([Q107507571](https://www.wikidata.org/wiki/Q107507571)) makes large
amounts of metadata about scientific publications freely available.
However, as a reader of this newsletter you probably know that *having
data* and *having FAIR data* makes a huge difference! (Spoiler: the
latter is better).

A paper that will be presented at the upcoming *International Semantic
Web Conference 2023*
([Q119153957](https://www.wikidata.org/wiki/Q119153957)), or in short
ISWC, presents SemOpenAlex! A web platform and data dump that is created
monthly on top of the OpenAlex data. It makes their data truly FAIR by
following the Linked Open Data principles.

## A publicly available resource with many use cases

As outlined by the authors, the presented SemOpenAlex Knowledge Graph
that can be accessed and queried by anyone has several use cases.

- Scholarly Big Data analytics

- Large-Scale scientific impact quantification

- Scholarly search and recommender systems

- Semantic scientific publishing

- Research project management and modeling

- Groundwork for scientific publishing in the future

- Knowledge-guided language models

- Benchmarking

## Human-in-the-loop seems necessary

I looked for myself in their database and found 3 entries (see image
below). The first match is the most comprehensive record covering almost
all of my publications. The two others are duplicates where each is
linked to only one publication. Funnily, it is the same publication
linked to both duplicates, yet also that publication is a duplicate.
**And even more ironically, the duplicate is a publication about Linked
Open Data validity!**

::: {.captioned-image-container}
<figure>
<a
href="https://substackcdn.com/image/fetch/f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fd506e3b1-4459-49a5-9138-ecff709dc6bf_336x422.png"
class="image-link is-viewable-img image2" target="_blank"
data-component-name="Image2ToDOM"></a>
<div class="image2-inset">
<img
src="https://substackcdn.com/image/fetch/w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fd506e3b1-4459-49a5-9138-ecff709dc6bf_336x422.png"
class="sizing-normal"
data-attrs="{&quot;src&quot;:&quot;https://substack-post-media.s3.amazonaws.com/public/images/d506e3b1-4459-49a5-9138-ecff709dc6bf_336x422.png&quot;,&quot;srcNoWatermark&quot;:null,&quot;fullscreen&quot;:null,&quot;imageSize&quot;:null,&quot;height&quot;:422,&quot;width&quot;:336,&quot;resizeWidth&quot;:null,&quot;bytes&quot;:20297,&quot;alt&quot;:null,&quot;title&quot;:null,&quot;type&quot;:&quot;image/png&quot;,&quot;href&quot;:null,&quot;belowTheFold&quot;:true,&quot;topImage&quot;:false,&quot;internalRedirect&quot;:null}"
srcset="https://substackcdn.com/image/fetch/w_424,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fd506e3b1-4459-49a5-9138-ecff709dc6bf_336x422.png 424w, https://substackcdn.com/image/fetch/w_848,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fd506e3b1-4459-49a5-9138-ecff709dc6bf_336x422.png 848w, https://substackcdn.com/image/fetch/w_1272,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fd506e3b1-4459-49a5-9138-ecff709dc6bf_336x422.png 1272w, https://substackcdn.com/image/fetch/w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fd506e3b1-4459-49a5-9138-ecff709dc6bf_336x422.png 1456w"
sizes="100vw" loading="lazy" width="336" height="422" />
<div class="image-link-expand">
<img
src="data:image/svg+xml;base64,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"
class="lucide lucide-maximize2" />
</div>
</div>
<figcaption>Finding myself three times in SemOpenAlex: automatic
processing is not perfect, manual curation is often
necessary</figcaption>
</figure>
:::

I would love to signal the duplicates such that they can be merged and
that the quality can be improved. Unfortunately in the current setting
of the platform this is not possible. I guess this is because they just
use data from OpenAlex, so I would have to figure out how to correct my
data there at the source. Anyway, OpenAlex is also a data aggregator
which makes it more difficult. This can probably be avoided when **using
unique author identifiers at all time** (an issue also apparent at the
SEMANTiCS data outlined in the previous newsletter item).

Are you intrigued to learn more about SemOpenAlex? Check out their
openly available paper [DOI:
10.48550/arXiv.2308.03671](https://doi.org/10.48550/arXiv.2308.03671) or
dive into their data yourself by using their web application!

[Explore their data yourself](https://semopenalex.org/){.button
.primary}

<div>

------------------------------------------------------------------------

</div>

That's it for this week of the FAIR Data Digest. I hope you found the
content interesting. Don't forget to share or subscribe. See you next
week!

Sven

::: {.subscription-widget-wrap attrs="{\"url\":\"https://fairdata.substack.com/subscribe?\",\"text\":\"Subscribe\",\"language\":\"en\"}" component-name="SubscribeWidgetToDOM"}
::: {.subscription-widget .show-subscribe}
::: {.preamble}
Thanks for reading FAIR Data Digest! Subscribe for free to receive new
posts and support my work.
:::

::: {.fake-input-wrapper}
::: {.fake-input}
:::

::: {.fake-button}
:::
:::
:::
:::


Hi everyone,  this last summer-break edition ties in with the previous edition. Last week I’ve wrote over research to make
<strong>
 publication metadata
</strong>
from the free publishing service CEUR-WS
<strong>
 available in a FAIR fashion
</strong>
, such that the data can be explored.

Messaggi di Rogue Scholar

FAIR Data Digest #13