tibhannover/confident_schema

The data schema for the ConfIDent project.

ConfIDent schema

This is a schema for the ConfIDent project that describes the necessary metadata requirements of academic events and event series.

Website

• https://TIBHannover.github.io/confident_schema

Repository Structure

• examples/ - example data
• project/ - project files (do not edit these)
• src/ - source files (edit these)
  • confident_schema
    • schema -- LinkML schema (edit this)
• datamodel -- Generated python datamodel
• tests - python tests

Developer Documentation

Use the `make` command to generate project artefacts:

• make all: make everything
• make deploy: deploys site

Credits

this project was made with linkml-project-cookiecutter

confident_schema's Issues

Add property for review process

The ORKG Team is in need of describing the type of review process a conference conducts.

They indent to link ORKG-descriptions of a paper to the conference it was submitted to. They recently partnered with two conferences where researchers in addition to their normal manuscribt could hand in a ORKG-description of their paper. The conferences descriptions in the ORKG currently look like this:

https://www.orkg.org/orkg/organizations/The_21st_International_Semantic_Web_Conference
https://www.orkg.org/orkg/organizations/SEMANTiCS_Conference

An apparently important part of these descriptions is the type of peer review.

We could offer a list of review types. I suggest to stick to these 7 different types for now: https://www.editage.com/insights/7-common-types-of-academic-peer-review

Use SMW conventions for naming slots and classes

Semantic Media Wiki's properties and classes always start with a capital letter, thus we decided to stick to the follwoing naming convention for classes and slots

• First letter is capitalaized no matter what
• No CamelCase or Pascal_Case but plain space
• no plurals for slot names except for container classes

add property for linking published review of conference

Some societies or organizers publish (quality) reviews of their conferences. These publications can describe a variety of different (quality) metrics or characteristics of the conference. For judging the conference and supporting researcher's participation decision, this can be a valuable ressource.

Example: https://doi.org/10.1016/j.tics.2021.01.007 mentions the number of attendees, fees and what formats and tools were used to support the virtual format of the conference. The publication refers to the neuromatch conference: http://www.bristol.ac.uk/neuroscience/events/2020/neuromatch-26oct.html.

I would not deem this a "conference output". Rather it is a publication that references the conference, similar to when a tweet, Wikipedia article or newsletter article would reference the conference. Maybe this property should be a kind of "sub-property" of "referenced-by" like "reviewed quality by" or something like that.

Event status questions

Question regarding: https://tibhannover.github.io/ConfIDent_schema/EventStatus/

What is the difference between a delayed and a postponed event status?

What does the "planned" Event Status mean? Is this used when no date has been specified yet? Would this even be usable in our framework, since Start date is a mandatory property?

Will the original Start and End date be archived when the status is switched to delayed or postponed? Maybe this is not necessary. I don't know who should benefit from this kind of information.

Adjust DBLP ID as Identifier for (collections of) Proceedings

Currently DBLP ID is modeled as External identifier for the Event Series or the Event. However, with regards to using the DBLP ID in this way Marcel Ackermann wrote the following:

"As for the events of a conference series, the modelling of that part is still an ongoing and by far not finished process at dblp. Historically, events had been "entered" as inhomogeneously and manually created HTML fragments on static web pages, and we are still in the process of reverse engineering them into truly identifiable entities within dblp.

Also, please don't semantically confuse the table of contents of a proceedings volume, e.g. given by resource URI

https://dblp.org/db/conf/jcdl/jcdl2002

with the actual conference event of JCDL 2002. They may be closely related, and often in a one-to-one relationship, but they are not the same. In general, the relation is many-to-many with plenty of crazy examples for crazy situations.

Additionally, conference event do not have a final ID in dblp, yet. In past collaborations we used intermediate resource URIs like

https://dblp.org/db/conf/jcdl/index#2002
https://dblp.org/streams/conf/jcdl#2002

to identify an event (and they still do work), but it remains undecided if that will be their final form. There is still a lot of work to do on our end before we can confidently consider conference events first-class citizens within dblp's data model."

Therefore, it might be better to consider the DBLP ID as an identifier for a Publication .

Add property for "programme"

at least to store information where a programme is available, e.g. URL etc.

add slot to relate missing contrinbutor roles to event (speaker, etc)

so far we only have slots for organizers and sponsors, but not for the other contributor types (attendee, moderator, speaker ...)

add property to Location to differentiate between international/national/regional?

the URIS schema does differentiate between these three and it could be useful for analyzing and filtering events.

Add distribution of countries of residence of contributors

Similar to the gender distribution of contributors the distribution of their countries of residence would help to evaluate in how far a conference meets its claim to be international and it can serve as an indicator in how far it is accessible to people from countries with less ressources.

Add Subtitle

Some conferences in the humanities have very long subtitles. This ensures that the titles entered are very long and, for example, obscure the date in the series view. Example the following two conferences:

https://www.confident-conference.org/index.php/Event:1546a94f-3702-4ff7-8c20-6ed778f81c66

Examples series:

https://www.confident-conference.org/index.php/Event_Series:7b30b7c0-14fd-49b6-996e-0b4c0561cfc7

Improve "Event Status" Enum by adding data properties to store previous dates

WRT the archival of postponed event dates, there are previously planned start date & previously planned end date data property in AEON which are not yet integrated in the MDS here.

This needs to be duscussed.

See also: #9

add OA property/ies for event output #3

We want to be able to find out quickly in ConfIDent if the proceedings/conference papers are open access.

see also StroemPhi/ConfIDent-schema#3

Adjust acceptance rate metric

We might have to differentiate between long and short paper submission and acceptance numbers (see Metric Types. For some conferences this seems to be a relevant distinction: https://github.com/lixin4ever/Conference-Acceptance-Rate

Add gender distribution of contributors

It could be interesting to capture the gender distribution of Organizers, Speakers and Participants.

This property and espacially the development of its values needs to be adviced by gender research experts.

Add event formats for career development opportunities

Some conferences offer dedicated support for early career researchers for example in the form of "career development workshops".

Add fees/ costs

We want to capture information about the fees of a conference. Options for modeling are: Specifying a range or providing multiple fields for student fee, regular fee, early bird, etc.. We will have to deal with the issue of different currencies.

Add social responsibility declarations

Organizers sometimes publish texts in which they declare their intention to meet the social responsibility they think they have.

These texts can address different issues of inequality. They are titled for example "gender equity statement", "diversity statement" or "code of conduct".

Publishing these kind of texts at least shows that the organizers have recognized an issue and presumably intend to meet their responsibility with regards to it.

Ideally we would want to add links to those texts to avoid the copyright issue. This again raises the problem that those links and consequently those texts are not permanent.

Add registration deadline

Adjust subfields of Subobject Publication

For our users browsing events and looking at event metadata, getting an overview of the publications coming directly out of the event or in relation to the event would be present a substantial benefit.

We could rethink not only to allow "collections of publications" like a whole proceeding to be added, but to allow for individual publications to be added. The advantage would be to be able to display author names on the event page and to give the user an impression of topics discussed at the event. Also some publications like posters or Videos are not always bundled in a referenceable collection. In this case we would have to significantly expand this subobject.

add property to differentiate between open vs closed audience

it needs to be possible to describe whether an event is free to come for whoever wants to participate or if it is a closed thing. I guess this can be implied by the fee specification, in that if the fee is specified as 0€ one could assume it to be free to attend for everyone. But off course it could be free of charge but you might need to be member of some kind of organization to participate.

Add property for "Recurrance Unit"?

We have modeled a property in the platform called "Recurrance Unit". Is this something that should be part of https://tibhannover.github.io/ConfIDent_schema/EventSeries/ too? Or will this be "calculated" based on the actual occurance of the related events of the series?

Adjust Alternative Name property by recording language of name

Maybe we should think about recording if an Alternative Name is written in a certain language. Later, when we have e.g. national conferences from Ukraine in the system, users should be able to switch between English or Ukrainian names in the display and search.

Add Field of Science and Technology (FOS) Classification by OECD to Academic Field

The Field of Science and Technology (FOS) Classification by OECD is quite simple compared to other classifications. It is a good candidate for our first classification schema to be used for the property Academic Field. It only has two levels of hierarchy and our curators, although not always being familiar with the specific field of a conference, will be able to apply it because of its simplicity.

Original document describing the FOS
OECD FOS Kategorien als Liste

Maybe this link also helps as an "External formater URI": https://bartoc.org/en/node/1042

DataCite is also using the FOS as a subject scheme: https://support.datacite.org/docs/field-descriptions-for-form#subjects. As a "subject scheme URI" they use the link to the pdf: http://www.oecd.org/science/inno/38235147.pdf

remodel the possible event-event and series-series relations

So far the relations between event were modeled with the related_to slot. However, this will not work as an intermediate relationship class is not what we want. We rather want a direct link between events or series and thus need more finegrained slots to reflect this.

make has_doi only single valued

currently there is https://tibhannover.github.io/ConfIDent_schema/has_publication/ to relate a publication to an event or series. The range of this is a Publication which uses the "has_doi" slot. Considering that a publication should only have one DOI it seems better to reflect this restriction in the schema accordingly.

Add ROR ID

See in how far we can provide the ROR ID of an org already with the external ID pattern, or if a new class will be needed for it.

add number of participants as a metric

Probably working with ranges such as:

Until 20
20-50
50-100
100-200
200-500
500-1000
More than 1000

Values of Location for country, region and city properties

The RWTH Team has suggested to stick to ISO norms for the values of Country and Region. They implemented the location notation like this for example: CA/BC/Vancouver

Codes for countries: https://en.wikipedia.org/wiki/ISO_3166-1_alpha-2#Officially_assigned_code_elements
Codes for regions: https://en.wikipedia.org/wiki/ISO_3166-1_alpha-2#Officially_assigned_code_elements

This seems like a good approach to avoid ambiguities with location descriptions. They also have managed to automatically fill in geo-coordinates based on this notation. They are using geograpy3.

See for an overview of already used locations: https://confident.dbis.rwth-aachen.de/or/index.php?title=Category:Location