Reproducible research | Competition Data Observatory

dataset: Create Interoperable FAIR Datasets

Thu, 11 Aug 2022 00:00:00 +0000

Interoperable, FAIR datasets

The primary aim of dataset is create well-referenced, well-described, interoperable datasets from data.frames, tibbles or data.tables that translate well into the W3C DataSet definition within the Data Cube Vocabulary in a reproducible manner. The data cube model in itself is is originated in the Statistical Data and Metadata eXchange, and it is almost fully harmonzied with the Resource Description Framework (RDF), the standard model for data interchange on the web[^1].

A mapping of R objects into these models has numerous advantages:

Makes data importing easier and less error-prone;
Leaves plenty of room for documentation automation, resulting in far better reusability and reproducability;
The publication of results from R following the FAIR principles is far easier, making the work of the R user more findable, more accessible, more interoperable and more reusable by other users;
Makes the placement into relational databases, semantic web applications, archives, repositories possible without time-consuming and costly data wrangling (See From dataset To RDF).

Our package functions work with any structured R objects (data.fame, data.table, tibble, or well-structured lists like json), however, the best functionality is achieved by the (See The dataset S3 Class), which is inherited from data.frame().

Contact

For contact information, contributors, see the package homepage.

Code of Conduct

Please note that the dataset project is released with a Contributor Code of Conduct. By contributing to this project, you agree to abide by its terms.

Click the Cite button above to demo the feature to enable visitors to import publication metadata into their reference management software.

Credibility is Enhanced Through Cross Links Between Different Data from Different Domains

Tue, 08 Jun 2021 18:50:00 +0000

As a consultant, what type of data do you usually work with?

I work at the intersection between strategy, finance and organisation. My usual dataset is quite broad - and sometimes unstructured. Oftentimes, the most decisive data are ones that cross domains: economic data coupled with environmental measurements, sociodemographic characteristics linked with online analytics.

If you were able to pick, what would be the ultimate dataset, or datasets that you would like to see in the Green Deal Data Observatory? And the Economy Data Observatory?

If I may venture that far, the interesting point is where these two data observatories meet. But high on my wishlist would be anything related to geospatial dispersion of environmental and climate data: land erosion, aerosols, solar incidence. From an economic perspective, my interest would go especially to - again - dispersion across regions or other geographical domains of, say, number of new enterprises, disposable income, tax incidence…

See our case study on connecting local tax revenues, climate awareness poll data and drought data in Belgium.

Why did you decide to join the challenge and why do you think that this would be a game changer for policymakers and for business leaders?

There is, both from an ecological and a societal point of view, an urgent need for open-access, real-time, trustworthy data to base decisions on. Ever since Kydland & Prescott’s analyses of “rules rather than discretion” and even earlier analyses of investment under uncertainty, the dynamic rules for optimal decision-making (including investment) require fast-response reliable data.

Do you have a favorite, or most used open governmental or open science data source? What do you think about it? Could it be improved?

Let me give one example: the AMECO annual macro-economic database is great for long-term historical analyses but its components ought to be real-time available. As an anecdote, as a fund manager in emerging markets we needed to anticipate macro-economic evolutions and in particular the manner in which capital markets anticipate these evolutions by adjusting foreign exchange rates or positioning themselves along yield curves. To some extent, we needed to predict what AMECO would tell us one year later by means of any real-time trustworthy assessments of the financial or economic situation. The latter data is what we would ideally have in an observatory.

To some extent, we needed to predict what AMECO would tell us one year later by means of any real-time trustworthy assessments of the financial or economic situation. The latter data is what we would ideally have in an observatory.

Is there a piece of information that recently surprised you? What was it?

I am currently working on water-related issues and came across a result reported in Nature Energy earlier this year that in more than one in ten hydropower stations, the extra warming from the dark surface of the water reservoir was enough to outbalance its “green” electricity generation potential, leading to no net climate benefits.

The researchers found that almost half of the reservoirs they surveyed took just four years to reach a net climate benefit. Unfortunately, they also found that 19% of those surveyed took more than 40 years to do so, and approximately 12% of them took 80 years—the average lifetime of a hydroelectric plant. Calculating the albedo-climate penalty of hydropower dammed reservoirs

Again: spatial distribution matters…

Photo: Kees Streefkerk, Unplash License

From your experience, what do you think the greatest problem with open data in 2021 will be?

Trust. In a society where “value” and even “truth” is determined more by the amount of (web) links to a particular “fact” than by its intrinsic characteristics, we need to be able to trust data — open data because it’s open and “closed” data because it’s closed.

What can our automated data observatories do to make open data more credible in the European economic policy and climate change or mitigation community and be more accepted as verified information?

If I may refer to the previous answer: credibility is enhanced through cross-links between different data from different domains that “does not disprove” one another or that is internally consistent. If, say, data on taxable income goes in one direction and taxes in another, it is the reasoned reconciliation of the - alleged or real - inconsistency that will validate the comprehensive data set. So I am a great believer in broad, real-time observatories where not only the data capture, but the data reconciliation is automated, sometimes by means of a simple comparative statics analysis, in other cases maybe through quite elaborate artificial intelligence.

Join us

Join our open collaboration Economy Data Observatory team as a data curator, developer or business developer. More interested in environmental impact analysis? Try our Green Deal Data Observatory team! Or your interest lies more in data governance, trustworthy AI and other digital market problems? Check out our Digital Music Observatory team!

Our Datasets Should be Retrieved Cleaned and Assessed in Order to Deliver Efficient Relevant and Credible Information

Mon, 07 Jun 2021 20:00:00 +0000

As a consultant, what type of data do you usually use in your work at ECORYS?

We work with a great variety of data – both from qualitative and quantitative sources – that we retrieve from publicly available sources or get through our clients. Since we are a public policy consultancy, most of the datasets are related to government reports, policies, statistics or surveys that we analyse and assess within a specific timeframe. Oftentimes, we gather open data like non-textual or numeric, such as maps and satellite images; so-called “raw data,” like weather, geospatial and environmental data; or data such as that generated in research like genomes, medical data, mathematical and scientific formulas.

If you were able to pick, what would be the ultimate dataset, or datasets that you would like to see in the Green Deal Data Observatory?

I would like to see more data on the consequences and impact of increasing drought and urban heat in our cities in the Green Deal Data Observatory. Because of the complexity of rapidly developing metropolitan regions and the uncertainty associated with climate change, we need to explore more climate change adaptation and mitigation activities, or disaster risk reduction, not only climate change itself.

See our drought case study on how we combine very different data in our observatory

We need more reliable datasets on the effect of global warming on urban resilience and more indicators to inform stakeholders on disaster risk reduction. The Green Deal Observatory could build indexes for public and private entities once we would have all the relevant data at hand. With this project, we could explore many possibilities to actually utilise open data for a common and societal good, working towards a great social cause.

Why did you decide to join the challenge and why do you think that this would be a game changer for policymakers and for business?

As a consultant for many socially relevant projects, everyday I see the importance of high quality and diverse datasets. I joined the challenge to contribute to significant causes enabled through the Green Deal Data Observatory and Economy Data Observatory. We can all benefit from the usage of open data, which is, in my opinion, a prerequisite for open government partnerships.

I believe that through our work and through open data collaborations, we show a good example for a cultural change in the relationship between citizens and the state, which can contribute to more transparency, more participation and more intensive cooperation.

The access and analysis of open data for the general public would make political action more transparent and more comprehensible. This can lead to greater accountability and a sense of duty on the part of public officials to the general public, which in turn can lead to greater acceptance of government action and strengthen the public’s trust in their government and administration.

Is there a number that recently surprised you? What was it?

Climate change is increasing people’s exposure to heat. Extreme temperature events have been documented to be rising in frequency, duration, and magnitude over the world. The number of persons exposed to heatwaves grew by roughly 125 million between 2000 and 2016.

Sydney by Marek Piwnicki Unplash License

From your experience, what do you think the greatest problem with open data in 2021 will be?

I see two great problems with the use of open data. The first one is the low level of exploitation. The other is the lack of transparency in data processing.

The use of open data should be transparent and meet high quality standards. If we want to enable communities to use it for solving local problems, we must do two things. First, data must be made easy to use (or actionable), and second, we have to increase public awareness and offer training for use. Furthermore, governments should release data in usable formats that follow open data guidelines. Currently, there is very little effort made at the community level to encourage the reuse of public data for the public good.

What can our automated data observatories do to make open data more credible in the European economic policy community and be more accepted as verified information?

Almost nothing is being done to help communities build the capability to analyze and implement open data without relying on technology.

Our API contains rich processing and descriptive metadata besides our high-quality indicators.

This is a critical task that the our fledlging data Observatories, the Digital Music Observatory, Green Deal Data Observatory and Economy Data Observatory, may be able to help with. Facilitating private-public partnerships is one step to encourage the data community to work with valuable open data. However, transparency and a high level quality assurance step must be given. In a joint collaboration with data curators, developers, technical specialists and academics, the datasets should be retrieved, cleaned and assessed in order to deliver efficient, relevant and credible information. The constant monitoring and regulation as well as compliance with data security guidelines are indispensable.

Join us

Comparing Data to Oil is a Cliché: Crude Oil Has to Go Through a Number of Steps and Pipes Before it Becomes Useful

Mon, 07 Jun 2021 10:00:00 +0000

As a developer at rOpenGov, and as an economic sociologist, what type of data do you usually use in your work?

Generally speaking, people’s access to (or inequalities in accessing) different types of resources and their ability in transforming these resources to other types of resources is what interests me. The data I usually work with is the kind of data that is actually nicely covered by existing rOpenGov tools: data about population demographics and administrative units from Statistics Finland, statistical information on welfare and health from Sotkanet and also data from Eurostat. Aside from these a lot of information is of course data from surveys and texts scraped from the internet.

We are placing the growing number of rOpenGov tools in a modern application with a user-friendly service and a modern data API.

In your ideal data world, what would be the ultimate dataset, or datasets that you would like to see in the Music Data Observatory?

Late spring and early summer time is, at least for me, defined by the Eurovision Song Contest. Every year watching the contest makes me ponder the state of the music industry in my home country Finland as well as in Europe. Was the song produced by homegrown talent or was it imported? Was it better received by the professional jury or the public? How well does the domestic appeal of an artist translate to the international stage? Many interesting phenomena are difficult to quantify in a meaningful way and writing a catchy song with international appeal is probably more an art than a science. Nevertheless that should not deter us from trying as music, too, is bound by certain rules and regularities that can be researched.

Music, too, is bound by certain rules and regularities that can be researched. Our Digital Music Observatory and its Listen Local experimental App does this exactly, and we would love to create Eurovision musicology datasets. Photo: Eurovision Song Contest 2021 press photo by Jordy Brada

Why did you decide to join the EU Datathon challenge team and why do you think that this would be a game changer for researchers and policymakers?

The challenge has, in my opinion, great potential in leading by example when it comes to open data access and reproducible research. Comparing data to oil is a common phrase but fitting in the sense that crude oil has to go through a number of steps and pipes before it becomes useful. Most users and especially policymakers appreciate ease-of-use of the finished product, but the quality of the product and the process must also be guaranteed somehow. Openness and peer-review practices are the best guarantors in the field of data, just as industrial standards and regulations are in the oil industry.

We provide many layers of fully transparent quality control about the data we are placing in our data APIs and provide for our end-users.

Join us

Creating Algorithmic Tools to Interpret and Communicate Open Data Efficiently

Fri, 04 Jun 2021 10:00:00 +0000

As a developer at rOpenGov, what type of data do you usually use in your work?

As an academic data scientist whose research focuses on the development of general-purpose algorithmic methods, I work with a range of applications from life sciences to humanities. Population studies play a big role in our research, and often the information that we can draw from public sources - geospatial, demographic, environmental - provides invaluable support. We typically use open data in combination with sensitive research data but some of the research questions can be readily addressed based on open data from statistical authorities such as Statistics Finland or Eurostat.

In your ideal data world, what would be the ultimate dataset, or datasets that you would like to see in the Music Data Observatory?

One line of our research analyses the historical trends and spread of knowledge production, in particular book printing based on large-scale metadata collections. It would be interesting to extend this research to music, to understand the contemporary trends as well as the broader historical developments. Gaining access to a large systematic collection of music and composition data from different countries across long periods of time would make this possible.

Why did you decide to join the challenge and why do you think that this would be a game changer for researchers and policymakers?

Joining the challenge was a natural development based on our overall activities in this area; the rOpenGov project has been around for a decade now, since the early days of the broader open data movement. This has also created an active international developer network and we felt well equipped for picking up the challenge. The game changer for researchers is that the project highlights the importance of data quality, even when dealing with official statistics, and provides new methods to solve these issues efficiently through the open collaboration model. For policymakers, this provides access to new high-quality curated data and case studies that can support evidence-based decision-making.

Do you have a favorite, or most used open governmental or open science data source? What do you think about it? Could it be improved?

Regarding open government data, one of my favorites is not a single data source but a data representation standard. The px format is widely used by statistical authorities in various countries, and this has allowed us to create R tools that allow the retrieval and analysis of official statistics from many countries across Europe, spanning dozens of statistical institutions. Standardization of open data formats allows us to build robust algorithmic tools for downstream data analysis and visualization. Open government data is still too often shared in obscure, non-standard or closed-source file formats and this is creating significant bottlenecks for the development of scalable and interoperable AI and machine learning methods that can harness the full potential of open data.

Regarding open government data, one of my favorites is not a single data source but a data representation standard, the Px format.

From your perspective, what do you see being the greatest problem with open data in 2021?

Although there are a variety of open data sources available (and the numbers continue to increase), the availability of open algorithmic tools to interpret and communicate open data efficiently is lagging behind. One of the greatest challenges for open data in 2021 is to demonstrate how we can maximize the potential of open data by designing smart tools for open data analytics.

What can our automated data observatories do to make open data more credible in the European economic policy community and be accepted as verified information?

The role of the professional network backing up the project, and the possibility of getting critical feedback and later adoption by the academic communities will support the efforts. Transparency of the data harmonization operations is the key to credibility, and will be further supported by concrete benchmarks that highlight the critical differences in drawing conclusions based on original sources versus the harmonized high-quality data sets.

We need to get critical feedback and later adoption by the academic communities.

How we can ensure the long-term sustainability of the efforts?

The extent of open data space is such that no single individual or institution can address all the emerging needs in this area. The open developer networks play a huge role in the development of algorithmic methods, and strong communities have developed around specific open data analytical environments such as R, Python, and Julia. These communities support networked collaboration and provide services such as software peer review. The long-term sustainability will depend on the support that such developer communities can receive, both from individual contributors as well as from institutions and governments.

Join us

EU Datathon 2021

Tue, 16 Mar 2021 00:00:00 +0000

Reprex, a Dutch start-up enterprise formed to utilize open source software and open data, is looking for partners in an agile, open collaboration to win at least one of the three EU Datathon Prizes. We are looking for policy partners, academic partners and a consultancy partner. Our project is based on agile, open collaboration with three types of contributors.

With our competing prototypes we want to show that we have a research automation technology that can find open data, process it and validate it into high-quality business, policy or scientific indicators, and release it with daily refreshments in a modern API.

We are looking for institutions to challenge us with their data problems, and sponsors to increase our capacity. Over then next 5 months, we need to find a sustainable business model for a high-quality and open alternative to other public data programs.

The EU Datathon 2021 Challenge

To take part, you should propose the development of an application that links and uses open datasets. - our data curator team
Your application … is also expected to find suitable new approaches and solutions to help Europe achieve important goals set by the European Commission through the use of open data.” - this application is developed by our technology contributors
Your application should showcase opportunities for concrete business models or social enterprises. - our service development team is working to make this happen!
We use open source software and open data. The applications are hosted on the cloud resources of Reprex, an early-stage technology startup currently building a viable, open-source, open-data business model to create reproducible research products.
We are working together with experts in the domain as curators (check out our guidelines if you want to join: Data Curators: Get Inspired!).
Our development team works on an open collaboration basis. Our indicator R packages, and our services are developed together with rOpenGov.

Mission statement

We want to win an EU Datathon prize by processing the vast, already-available governmental and scientific open data made usable for policy-makers, scientific researchers, and business researcher end-users.

“To take part, you should propose the development of an application that links and uses open datasets. Your application should showcase opportunities for concrete business models or social enterprises. It is also expected to find suitable new approaches and solutions to help Europe achieve important goals set by the European Commission through the use of open data.”

We aim to win at least one first prize in the EU Datathon 2021. We are contesting all three challenges, which are related to the EU’s official strategic policies for the coming decade.

Challenge 1: A European Grean Deel

Our Green Deal Data Observatory connects socio-economic and environmental data to help understanding and combating climate change.

Challenge 1: A European Green Deal, with a particular focus on the The European Climate Pact, the Organic Action Plan, and the New European Bauhaus, i.e., mitigation strategies.

Climate change and environmental degradation are an existential threat to Europe and the world. To overcome these challenges, the European Union created the European Green Deal strategic plan, which aims to make the EU’s economy sustainable by turning climate and environmental challenges into opportunities and making the transition just and inclusive for all.

Our Green Deal Data Observatory is a modern reimagination of existing ‘data observatories’; currently, there are over 70 permanent international data collection and dissemination points. One of our objectives is to understand why the dozens of the EU’s observatories do not use open data and reproducible research. We want to show that open governmental data, open science, and reproducible research can lead to a higher quality and faster data ecosystem that fosters growth for policy, business, and academic data users.

We provide high quality, tidy data through a modern API which enables data flows between public and proprietary databases. We believe that introducing Open Policy Analysis standards with open data, open-source software, and research automation, can help the Green Deal policymaking process. Our collaboration is open for individuals, citizens scientists, research institutes, NGOS, and companies.

Challenge 2: An economy that works for people

Our Economy Data Observatory will focus on competition, small and medium sized enterprizes and robotization.

Challenge 2: An economy that works for people, with a particular focus on the Single market strategy, and particular attention to the strategy’s goals of 1. Modernising our standards system, 2. Consolidating Europe’s intellectual property framework, and 3. Enabling the balanced development of the collaborative economy strategic goals.

Big data and automation create new inequalities and injustices and have the potential to create a jobless growth economy. Our Economy Data Observatory is a fully automated, open source, open data observatory that produces new indicators from open data sources and experimental big data sources, with authoritative copies and a modern API.

Our observatory monitors the European economy to protect consumers and small companies from unfair competition, both from data and knowledge monopolization and robotization. We take a critical Small and Medium-Sized Enterprises (SME)-, intellectual property, and competition policy point of view of automation, robotization, and the AI revolution on the service-oriented European social market economy.

We would like to create early-warning, risk, economic effect, and impact indicators that can be used in scientific, business, and policy contexts for professionals who are working on re-setting the European economy after a devastating pandemic in the age of AI. We are particularly interested in designing indicators that can be early warnings for killer acquisitions, algorithmic and offline discrimination against consumers based on nationality or place of residence, and signs of undermining key economic and competition policy goals. Our goal is to help small and medium-sized enterprises and start-ups to grow, and to furnish data that encourages the financial sector to provide loans and equity funds for their growth.

Challenge 3: A Europe fit for the digital age

Our Digital Music Observatory is not only a demo of the European Music Observatory, but a testing ground for data governance, Digital Servcies Act, and trustworthy AI problems.

Challenge 3: A Europe fit for the digital age, with a particular focus Artificial Intelligence, the European Data Strategy, the Digital Services Act, Digital Skills and Connectivity.

The Digital Music Observatory (DMO) is a fully automated, open source, open data observatory that creates public datasets to provide a comprehensive view of the European music industry. It provides high-quality and timely indicators in all four pillars of the planned official European Music Observatory as a modern, open source and largely open data-based, automated, API-supported alternative solution for this planned observatory. The insight and methodologies we are refining in the DMO are applicable and transferable to about 60 other data observatories funded by the EU which do not currently employ governmental or scientific open data.

Music is one of the most data-driven service industries where most sales are currently executed by AI-driven autonomous systems that influence market shares and intellectual property remuneration. We provide a template that enables making these AI-driven systems accountable and trustworthy, with the goal of re-balancing the legitimate interests of creators, distributors, and consumers. Within Europe, this new balance will be an important use case of the European Data Strategy and the Digital Services Act.

The DMO is a fully functional service that can serve as a testing ground of the European Data Strategy. It can showcase the ways in which the music industry is affected by the problems that the Digital Services Act and European Trustworthy AI initiatives attempt to regulate. It is being built in open collaboration with national music stakeholders, NGOs, academic institutions, and industry groups.

Our Product/Market Fit was validated in the world’s 2nd ranked university-backed incubator program, the Yes!Delft AI Validation Lab. We are currently developing this project with the help of the JUMP European Music Market Accelerator program.

Problem Statement

The EU has an 18-year-old open data regime and it makes public taxpayer-funded data in the values of tens of billions of euros per year; the Eurostat program alone handles 20,000 international data products, including at least 5,000 pan-European environmental indicators.

As open science principles gain increased acceptance, scientific researchers are making hundreds of thousands of valuable datasets public and available for replication every year.

The EU, the OECD, and UN institutions run around 100 data collection programs, so-called ‘data observatories’ that more or less avoid touching this data, and buy proprietary data instead. Annually, each observatory spends between 50 thousand and 3 million EUR on collecting untidy and proprietary data of inconsistent quality, while never even considering open data.

Our automated data observatories are modern reimaginations of the existing observatories that do not use open data and research automation.

The problem with the current EU data strategy is that while it produces enormous quantities of valuable open data, in the absence of common basic data science and documentation principles, it seems often cheaper to create new data than to put the existing open data into shape.

This is an absolute waste of resources and efforts. With a few R packages and our deep understanding of advanced data science techniques, we can create valuable datasets from unprocessed open data. In most domains, we are able to repurpose data originally created for other purposes at a historical cost of several billions of euros, converting these unused data assets into valuable datasets that can replace tens of millions’ worth of proprietary data.

What we want to achieve with this project – and we believe such an accomplishment would merit one of the first prizes - is to add value to a significant portion of pre-existing EU open data (for example, available on data.europa.eu/data) by re-processing and integrating them into a modern, tidy database with an API access, and to find a business model that emphasises a triangular use of data in 1. business, 2. science and 3. policy-making. Our mission is to modernize the concept of data observatories.

Our solution

We are empowering data curators with reproducible research solutions to create high-quality, rigorously tested original datasets from low quality, not validated, not tidy open data. We help them to design meaningful business, policy or scientific indicators and provide them with a software and API to keep the data up-to-date. We help them deposit a copy of the authoritative, uncompromised dataset onto Zenodo, the EU’s data repository, with a DOI or new DOI version.
We create a research workflow that periodically (daily, weekly, monthly, quarterly or annually) collects, corrects and re-processes the data. We use peer-reviewed statistical software and unit-tests to make sure that the data is sound.

Panning out gold from muddy open sources - with automation technology.

We add value with correcting open (and proprietary!) data problems that make open data hard to use, and proprietary, in-house data hard to re-use.
- regions corrects inconsistent geographical coding. Eurostat has no mandate to correct geographical coding, and member states do not historically adjust their data. With many thousands of parish, county, region, province, state boundary changes within states, regional and metropolitian area datasets are not usable without our software.
- iotables puts extremely complex national accounts data into actually useful environmental and economic impact indicators. Instead of working with each country separately, our standardized system can calculate direct and indirect effects, as well as multipliers for every European country that works in the European statistical framework (EU member states, EEA, UK, member candidates.)
- retroharmonize connects cross-sectional surveys with non-European countries, puts pan-European surveys into time series, and corrects regional subsamples. We are creating new indicators from Eurobarometer, Afrobarometer, Arab Barometer, and standardized CAP surveys, as well as other harmonized surveys. We help design surveys that can utilize data from already existing, openly available surveys.
We place the authoritative copy to a data repository (Zenodo or Dataverse), automatically document the data, and make it available in a modern API for SQL queries or CSV downloads.
We present the data with commentary and blog posts from our curators (see: Is Drought Risk Uninsurable? - solidarity and climate change in Belgium) and contributors on a semi-automatically refreshed, open source web portal.
We are perfecting the agile open collaboration model in a triangular setting, where corporate users, scientific researchers, public and non-governmental policy makers, and even citizen scientists can work around a single data ecoystem.
We are validating a business model that allows the commercial, scientific, and policy use of re-processed, high quality data products made from open and shared data.