Wikidata:WikiProject Scholia/Robustifying/Usage
Usage of Scholia is whatever user activities which the Scholia team can measure of detect regarding the use of Scholia. The single most conventional metric to report is number of Scholia profiles which the tool presents. Other conventional usage metrics report what sort of audience seeks what sort of profile. As Scholia also accepts user generated content, another important class of metrics is reporting what sorts of individuals and organizations contribute content for Scholia to present, and what kind of content this is, and the extent of the reuse.
Usage is an important subset of overall Scholia growth. Most Scholia end users seek a benefit for themselves only, and not the online public commons. However, contributions of user generated content may contribute to general growth of the project beyond the user only getting profiles from Scholia. Grow also includes contributions which are not usage, such as contributing tool development, corpora of data, or administrative support of any kind.
Key resources[edit]
- Scholia usage tickets in GitHub
- Scholia pageview statistics in paws-public.wmflabs.org
- Monthly pulse on GitHub
Usage goals for 2020[edit]
We already publish some basic metrics for usage statistics of Scholia-related Wikidata properties (cf. Table 2 of the proposal). We lack more granular insights, like the impact of Scholia on a particular field or institution. Paving the way towards routine availability of such metrics is thus an aim of this project, so that we support our contributors and partners with the media metrics that institutions use to demonstrate the value they get from engagement. What is clear is that the traffic to Scholia pages is increasing (cf. Fig. 2 of the proposal).—Scholia team, Robustifying Scholia, 2019
Milestones[edit]
- changes in Overall Scholia/ WikiCite stats
- Compare https://scholia.toolforge.org/ November with February
- author name string (P2093) ⇒ author (P50) conversion
- LargeDatasetBot joins the collaboration
- https://www.wikidata.org/wiki/User:LargeDatasetBot
- Imports the missing ~12 million PubMed articles
- Blocked in mid-November 2019: https://www.wikidata.org/w/index.php?diff=1052224707&diffmode=source
Update for 2019[edit]
From February 2019 to November 2019, users have accessed a median of 8000 Scholia profiles/day, with a max of about 21,000/day, a few days offline with 0, and consistency near the median otherwise. The growth from before February has slowed. One likely contributing factor is general suspension in Wikidata of importing large datasets. Consistent use is helpful, and Scholia is not in a period of general user recruitment at this time.
The ideal content curation cycle for Scholia content in Wikidata is bulk data import, curation, then presentation as a corpus to profile with Scholia. Wikidata has a backlog with bulk data import, so at this time Scholia usage development focuses on curation and presentation. An example of curation is the author name string (P2093) ⇒ author (P50) conversion, which is author disambiguation. An hour of curation in this workflow can lead a user to an interesting presentation and profile. Scholia's data foundation is in academic citations, but when matched with new data corpora for chemicals, clinical trials, topic tagging, or institutional affiliations, Scholia is able to present exponentially more populated profiles.
About[edit]
A general goal of Scholia is to make insights from research media more accessible. When users get Scholia profiles, ideally they use them in their work, further enrich the content by applying their expertise to fill content gaps related to their Scholia use, and that they should do so with the support of their institution. Common paths to this point include researchers annotating publications or combining datasets in their field, or of someone in communications or marketing contributing already existing profile data to Wikidata as a distribution strategy.
The most important metrics in this project are the ones which would persuade institutional partners, like universities, to engage in contributing and curating their own data into the public data commons. The universe of possible measurements includes base citation data, like PubMed and CrossRef; enriched data, like author disambiguation applied to the base data; usage data of all of these, divided into use by machines versus presentation to human eyes; and counts of engagement, including number of people and organizations which participate in the project.
Conventional use[edit]
Conventional Scholia use is the generation of profiles that are browsed by individual human users for purposes of research, education and administration. Useful metrics in this category include basic information about that use
- Count of profiles generated
- Classification of profile use
- profiles of individuals, organizations, topics, etc
- field of use, such as medicine, humanities, meta, etc.
- Count of unique users
- Classification of users
- Individuals
- Organizations
- Bots
- Cost of providing the tool
- Computation
- Other inputs
Educational use[edit]
For educational use beyond education about the subjects profiled by Scholia, see the more detailed listing at Wikidata:Wikidata curricula/Activities/Explore Scholia as well as Wikidata:Wikidata curricula/Activities for educational activities around Wikidata more broadly.
Other use[edit]
Advanced use of Scholia is any activity in which the user contributes to the tool in a cycle of improvement.
- Development of corpora
- Contribution to code
- Presentations