Wikidata:Events/Data Quality Days 2022

<< 2021
Data Quality Days, 8-10 July 2022



Welcome!
The Data Quality Days is an online event designed for the Wikidata community and focusing on processes around data quality. During 3 days of presentations, workshops and facilitated conversations, we will discuss how we are currently identifying and fixing incorrect data on Wikidata, how we could improve these processes to increase data quality, and what concrete measures we could put in place together, with policies, tools or documentation.
Participation
The sessions are open to everyone involved in Wikidata and interested in the topic of data quality. The sessions took place on the online platform Jitsi. Some sessions have been recorded in video and uploaded to this Youtube playlist, and collaborative notes are accessible from the schedule. Slides from speakers are added to the Commons category Data Quality Days 2022 presentations. If you attended the event, feel free to add your name on the participants list! Please note that the entire initiative, including the presentations and discussions on various channels, is covered by the Code of Conduct for technical spaces. Please contribute in making the discussion spaces safe and welcoming for everyone.
Discussions and help
If you have questions or suggestions about the event, feel free to use the talk page. Announcements about the event are talking place on the same talk page, and on Wikidata Telegram group. If you want to reach out to the organizers directly, feel free to contact Lea Lacroix (WMDE) (lea.lacroix@wikimedia.de or @Auregann on Telegram).

Sessions

edit

Friday, July 8th

edit
Time & links Session & speakers
16:00 UTC

80min

Replay

Notes

Slides

Opening session and presentation of the program

During the opening session, we will give you an introduction to the topic of data quality on Wikidata, as well as a glimpse on all the exciting sessions that will take place during the event. On top of that, we will present the various discussions topics that could be covered during the event, and we will offer you to vote for your favorite. Don't miss the opening!

  • Introduction (10min)
  • Data quality: what are we talking about? (20min)
  • Program teasing: each speaker can come and present their upcoming session in 2min (30min)
  • Presentation of the discussion topics and how to vote on them (20min)

with Léa Lacroix, Lydia Pintscher and Manuel Merz

17:30 UTC

80min

Notes

Discover data quality and patrolling tools

This session will allow you to dive into data quality by discovering various projects and tools. Each topic will be presented in 10min, questions will take place in the chat.

19:00 UTC

60min

Social event

Let's have fun together around Wikidata! Join us to participate in a pub quiz, "guess the item", and play other games together.

Saturday, July 9th

edit
Time & links Session & speakers
09:00 UTC

50min

Notes

Structured conversation #1: Round-tripping data

There are excellent gold-standard sources out there that Wikidata can use, but even those make mistakes. The same is true if you use data from Wikidata for your own project. Therefore, syncs from and to Wikidata should ideally go in both directions (so called “data round-tripping”). Unfortunately, it is currently not as simple as it should be to set this up sustainably.

Goals of the session:

  • Collect existing hurdles for setting up round-tripping
  • Additional collection (likely no focus in the discussion)
    • Share examples of where this works particularly well already (we could use these for sharing best practices)
    • Collect examples where building up new syncs with external sources would be of great benefit to data quality
  • Discuss how we can help users who want to set-up sustainable round-tripping
  • Find allies to improve the status-quo

Facilitation: Manuel Merz and Lydia Pintscher

10:00 UTC

60min

Notes

Build queries for data quality and maintenance

with Harmonia Amanda and Lucas Werkmeister

Break
12:00 UTC

50min

Replay

Presentation

Notes

How do we deal with concurrent uses of different properties? The example of modeling data for humans

The same information can sometimes be stored in different statements, thus making it very difficult to be found through a single query; I will show some examples, mainly from instance of (P31)human (Q5) items, and I will survey the existing ways of dealing with the problem and possible future solutions.

with Epìdosis - presentation

13:00 UTC

50min

Notes

Bug Triage Hour: Data Quality edition

During the Bug Triage Hour, we look at Phabricator tickets together and see how we could improve their description and tags. Come with your favorite tickets related to data quality!

with Lydia Pintscher and Manuel Merz

14:00 UTC

50min

Notes

Slides

Discussion: Matching new Wikipedia articles to Wikidata items

New Wikipedia articles are continually being created, but they frequently aren't matched up with existing Wikidata items, leading to missing interwiki links, and duplicates on Wikidata. This session will discuss the status quo - including existing matching mechanisms, mass new item creations, and merging work - and identify ways to improve this work in the future.

with Mike Peel

15:00 UTC

50min

Notes

Structured conversation #2: Rules and anarchy

Some policies and guidelines on Wikidata are not fully enforced in practice (e.g. Wikidata:Bots, Wikidata:Notability). Ignoring important rules can have negative consequences for the Community and data quality.

Goals of the session:

  • Collect examples for policies and guidelines that are currently ignored with negative consequences
  • Discuss possible solutions
  • Find allies to improve the status-quo

Facilitation: Manuel Merz and Lydia Pintscher

16:00 UTC

50min

Replay

Notes

Using Scholia in curation workflows

We will use Scholia to engage in various curation workflows that address several kinds of data issues, from inconsistencies to incompleteness, lack of references and lack of updates.

with Daniel Mietchen

17:00 UTC

50min

Notes

Slides

Dealing with cross-wiki spam on Wikidata

Presentation of the most important tools and how to use them (RfD, nominating local pages for deletion, useful gadgets etc.), followed by questions and discussions.

with Bencemac

18:00 UTC

60min

Replay

Livestream with Ainali & Abbe98 - special Data Quality

Sunday, July 10th

edit
Time & links Session & speakers Session & speakers
08:00 UTC

50min

Notes

Structured conversation #3: Why isn’t there more guidance on this?

Doing the right thing on Wikidata is often harder than necessary. Some defaults and processes are annoying to the best of us and for new editors a lack of guidance can even be an unsolvable problem.  

Goals of the session:

  • Collect examples where missing guidance and bad defaults cause editors problems which eventually leads to lower quality data
  • Discuss possible solutions
  • Find allies to improve the status-quo

Facilitation: Manuel Merz and Lydia Pintscher

09:00 UTC

50min

Replay

Notes

Slides

Using Lexemes in Abstract Wikipedia: how can we improve the data?

Abstract Wikipedia's ultimate goal is to automatically create articles in various languages using Natural Language Generation techniques. The success of any such approach will depend to an important extent on the quality of lexicographical and grammatical data in Wikidata. While data in Wikidata was designed to be machine-readable, it is not necessarily in the best shape to assist Abstract Wikipedia's mission yet. In the presentation we will examine the relevant data extant in Wikidata, and how it can be improved.

with Ariel Gutman

10:00 UTC

60min

2 parallel rooms

Workshop: improving documentation and ontology for Lexemes

with Ariel Gutman

Notes

Follow-up from the conversation about round-tripping data
  • Create a landing page about round-tripping data
  • Create a property proposal for a property that records how to contact data sources for error corrections

Notes

Break
12:00 UTC

50min

Replay

Notes

Slides

Telegram group

Introduction to Entity Schemas and Shape Expressions

with Kat Thornton, Andra Waagmeester, Eric Prud'hommeaux

13:00 UTC

50min

2 parallel rooms

Schema editing session

with Kat Thornton, Andra Waagmeester, Eric Prud'hommeaux

Notes Telegram group

Constraint-a-thon

Let's set property constraints together (bring your favorite property ID!)

with Mike Peel

Notes Slides

14:00 UTC

40min

Notes

Discover data quality and patrolling tools

This session will allow you to dive into data quality by discovering various projects and tools. Each topic will be presented in 10min, questions will take place in the chat.

14:40 UTC

20min

Notes

Slides

Closing session

Wrap-up of the event, documentation and next steps

with Léa Lacroix, Lydia Pintscher and Manuel Merz

Participants

edit

If you plan to join one or several events, or to work on projects related to data reuse during this period, you can sign up here!
Feel free to indicate your username and the specific topics you're interested in around data quality.

  1. Lea Lacroix (WMDE) - Interested in Entity Schemas, processes for data reusers to give back "cleaned" data, improve clarity around ontology
  2. AGutman-WMF - Representation, consistency and quality of lexical data for use in a possible Natural Language Generation pipeline of Abstract Wikipedia
  3. Epìdosis - mainly authority control (improving the connection of Wikidata to the authority files of the most important VIAF members) and merging duplicates
  4. DrThneed - tips on maintaining data quality for new Wikiproject owners - what sort of problems should I be looking for?
  5. Alhassan Mohammed Awal (talk) 07:48, 9 June 2022 (UTC)[reply]
  6. MisterSynergy — vandalism and patrolling
  7. Aisha Khatun — Learning how to ensure quality data extracted from Wikidata for various use cases.
  8. -Mohammed Sadat (WMDE) (talk) 15:30, 14 June 2022 (UTC)[reply]
  9. Jahl de Vautban duplicates prevention and dialog between Wikidata and other data providers we link to.
  10. Sotho Tal Ker - mainly authority control topics (external data provider quality, conflations/duplicates), fighting vandalism
  11. Semantisch - Daniil Dobriy @ Vienna University of Economics and Business: ontology reuse in WD, linking WD to LOD
  12. salgo60 - learn if someone is looking into Sustainable Development Goals (Q7649586) see test using an API and if that is something for WD
  13. Akbarali (talk) 17:04, 20 June 2022 (UTC) data extraction and ensure quality.[reply]
  14. Feliciss - Data quality in scholarly articles.
  15. James Moore200 (talk) 08:27, 27 June 2022 (UTC) tools on data quality.[reply]
  16. Favourdare123 (talk) 08:56, 27 June 2022 (UTC)[reply]
  17. Profayobami (talk) 09:21, 27 June 2022 (UTC)[reply]
  18. Musahfm (talk) 10:03, 27 June 2022 (UTC)[reply]
  19. Sannita - Lexemes and entity schemas!
  20. Kingaustin07 (talk) 10:51, 27 June 2022 (UTC)[reply]
  21. maiarocg Sources and references in Wikidata, how the community approaches and sees them, and how to ensure their quality!
  22. Praze02 (talk) 11:31, 27 June 2022 (UTC) user:Praze02 (utc)[reply]
  23. Dnshitobu (talk) 15:31, 27 June 2022 (UTC)[reply]
  24. Krasimir Angelov (talk) - Lexical resources and NLG for Abstract Wikipedia in GF
  25. Wolfgang8741 (talk) 21:07, 29 June 2022 (UTC) - Workflow transparency and monitoring for data corruption, highlight instance of tools changing statements values cross wiki, but keeping external references that don't match - Unfortunately I am unable to attend on these dates.[reply]
  26. Erfurth (talk) 05:56, 1 July 2022 (UTC) curates with Wikidata:WikiProject Scholia][reply]
  27. Rima H (WMID) - how to model intangible cultural data in Wikidata Rima H (WMID) (talk) 08:07, 1 July 2022 (UTC)[reply]
  28. Pigsonthewing - when constraints go bad
  29. PKM (talk) 10:48, 5 July 2022 (UTC)[reply]
  30. Ainali (talk) Will do a livestream on manual quality work. 16:11, 6 July 2022 (UTC)[reply]
  31. Removena - I am interested in federated queries to fill gaps. Currently working on a comparison between Wikidata and Factgrid Removena (talk) 18:31, 6 July 2022 (UTC)[reply]
  32. Rdrg109 - I simply love Wikidata Rdrg109 (talk) 23:27, 6 July 2022 (UTC)[reply]
  33. Lately I am interested about lexicographical data. --KuboF Hromoslav (talk) 18:34, 7 July 2022 (UTC)[reply]
  34. Lucas Werkmeister (WMDE) (talk)
  35. VIGNERON (talk) 16:01, 8 July 2022 (UTC)[reply]
  36. Ambrosia10 (talk) 16:08, 8 July 2022 (UTC)[reply]
  37. Andrea Westerinen (talk) 16:11, 8 July 2022 (UTC)[reply]
  38. Diego (WMF) (talk) - Currently working on automatic assessing edit and item quality.
  39. Yupik (talk) 16:24, 8 July 2022 (UTC)[reply]
  40. Jelabra (talk) 18:26, 8 July 2022 (UTC) - Working on tools with Entity schemas and ShEx[reply]
  41. Lydia Pintscher (WMDE) (talk) 17:15, 8 July 2022 (UTC)[reply]
  42. Manuel (WMDE) (talk) 06:49, 9 July 2022 (UTC)[reply]
  43. Inaccuracies, inconsistencies, incompleteness, mismatches with external databases, documentation for all entity types and workflows. Focus on scholarly content. Daniel Mietchen (talk)
  44. 圍繞著如何「管理龐大的資料庫」,「讓資料庫能發揮最大的效用」,「中央樞紐的管制功能與地區樞紐的配合度」古代遺留的智慧如果沒人學習並且傳承給下一代,它屬於死物。任何人花費時間去學習把這些智慧實踐在日常生活當中,間接的影響了其他人的生活也因此提升品質。卻因為擁有權力的資產階級為了鞏固自身的利益,對於底層階級施加了各式各樣的枷鎖,社會的金字塔結構就一直存在。我的興趣只有接收資訊化為自己的知識,知識是一股沒人可以奪走的無形力量,金字塔結構的枷鎖只有善用知識的人群團結一致才有打破的機會。我想觀察這個理念能不能得到更多的共鳴!J.zht (talk) 15:04, 9 July 2022 (UTC)[reply]
  45. TiagoLubiana (talk) 16:02, 9 July 2022 (UTC)[reply]
  46. Jheald (talk) 16:09, 9 July 2022 (UTC)[reply]
  47. --Harmonia Amanda (talk) 09:51, 10 July 2022 (UTC) (I have been here, just forgot to sign in)[reply]
  48. Fexpr - Lexicographical data and learning more tools to improve data quality on Wikidata. Fexpr (talk) 11:57, 10 July 2022 (UTC)[reply]

See also

edit