ODDOL

Inspiration

With the growing availability of open data sources, it is time that online learning makes the most of these sources allowing students to easily experiment, verify, improve and interact with them and build quick and shareable dashboards with other students.

Problem

During the COVID-19 crisis, a lot of data has been released around the world daily. Most of these are not properly curated and therefore not easily findable. Most of our current online learning approaches make use of some pre-fabricated data. They may help explain some topics, but the students and online learners learn very little about the real social and economic impacts from these prepared datasets. Therefore, online learning must be driven by open datasets.

Secondly, even if some open datasets are used by a small group of researchers, their usage may not be known to other online learners. It is essential to know the purpose, usage, and the results obtained by the data analysis. Not only will this help the online instructors to prepare and improve their courses with interesting real-life use cases but also help the students verify as well as improve already published research works. Therefore they need to be curated considering the choice of licenses, purpose, use, etc.

Finally, there must be a possibility to contribute and curate new datasets obtained from the data analysis results as well as new experimental results.

Solution: ODDOL

ODDOL revolves around four principles for making the best use of open data.

Search
Analyze
Describe
Contribute

It helps the users search the datasets based on the license, help them analyze their existing uses and purposes, use them for their projects, describe their use as well as help them contribute in the form of curation of new experiment datasets and results.

Tools used

Wikidata
HTML, Javascript, Nodejs, CSS
SPARQL
Mediawiki API
Material design
Shape Expressions (ShEx)
Python

What you have done during the weekend

Architecture, Models and demo for an open dataset use: Design
Building open models on Wikidata for the following related to data sources
- dataset: https://www.wikidata.org/wiki/EntitySchema:E207
- SPARQL endpoint: https://www.wikidata.org/wiki/EntitySchema:E208
- API endpoint: https://www.wikidata.org/wiki/EntitySchema:E209

The solution’s impact to the crisis

ODDOL ensures that the curation of open data by the community members achieves the following

lessons from others' experiences
maximum data reuse
verifiability of results
improvements and possible corrections

The necessities in order to continue the project

Automated annotation linking datasets and results
Periodic open dataset usage reports
Analysis of results based on algorithms
Ability to edit and describe a dataset and write a report directly on the platform
Integration with other applications like Zenodo, Figshare, etc.
Easy access on mobile devices

The value of your solution(s) after the crisis

The current crisis has shown the importance of open data and the need to understand the decisions taken by the policymakers around the world. This transparent nature of open-data and decisions driven by open data will continue to play a major role even after the crisis. Thus community-driven approach to use, curate, publish and contribute open data needs to be supported.