What is TDM?  

There are various definitions of Text and Data Mining (TDM) which cover both the technicalities and utilities of the practice. The UK Intellectual Property Office (IPO) usefully define TDM as: ‘The use of automated analytical techniques to analyse text and data for patterns, trends and other useful information’. Even within TDM, there are different definitions for both text and data mining. Text mining is more commonly seen as the computational process of discovering and extracting knowledge from unstructured data. Data mining, on the other hand, is the computational process of discovering and extracting knowledge from structured data. There has been a surge of interest in the use of TDM in academia across all disciplines ranging from the sciences to the humanities. Yet undertaking TDM has also entailed a whole host of legal and political issues, which have nearly threatened to hinder the practice. These issues have largely centred around copyright, intellectual property rights, licenses and download limits. 

(more…)

All applications for the UK Research and Innovation Future Leaders Fellowships require a data management plan. As the umbrella organisation of the UK’s major Research Councils, there is an expectation that all UKRI-funded research is ‘to be made available to the research community in a timely and responsible manner unless there are exceptional reasons why this cannot happen’. Researchers are also advised to consult the Research Council common principles on data policy as this provides the overarching framework for the individual UK research councils.

Researchers should use the template and guidance provided by the URKI. Plans can be up to three-page long but can be as little as a quarter of a page of A4 for less complex research projects. The data management plan must also demonstrate how ‘the applicant will meet, or already meets their responsibilities for research data quality, sharing and security’.

You can also find a template for the UKRI data management plan on DMP Online.

Further Resources

• UKRI Data Policy
• UKRI guidance on best practice in the management of research data

This guide will aim to provide some useful advice for students on finding data to re-use during their research projects.

Data Resources Online

You may want to start by using UCL Explore to search for research studies based on secondary datasets. From there, you can consult the Registry of Research Data Repositories (re3data.org) – a global repository of research data. It is also best kept in mind that different datasets require different permissions. If you are planning to use safeguarded or controlled access data, you will need to abide by additional conditions for accessing it. For example, this may include specific forms of citation, depositor permission to registration and authentication of the users of the data. There are plenty of online courses to help you navigate your way through the use of digital datasets. The University of Edinburgh run an online research data management training programme called MANTRA to help researchers learn how to manage their digital data. Furthermore, the UK Data Service provides a range of dataset and topic guides along with video tutorials on how to use data.

(more…)

The rapid growth of social media has inevitably led to a wealth of data into all aspects of our everyday lives. At the same time, this has been accompanied by concerns about how we use data harvested from social media sources. There is already a growing literature out there on the ethical implications of using social media and so this FAQ will aim to briefly summarise some of key arguments that researchers will need to consider.

(more…)

The Freedom of Information Act was created in 2000 to increase the transparency of the public sector and their activities. Under the Act, research data can be requested although copyright and IP rights to the data remain with the original researcher. When a request for information is made, there are normally 20 working days to respond. In the UK, most universities are defined as public authorities and are thus legally obliged to respond to FOI requests. There are, however, some key exemptions to the Act in the UK including:

1. Personal data about living individuals cannot be requested, unless it is about you.
2. Information that is accessible by other means.
3. Information intended for future publication.
4. Information that is subject to a confidentiality agreement.
5. Information whose release would prejudice legitimate commercial interests.
6. In Scotland, information that is part of a finite research programme for which there is a publication schedule and clear intent to publish.
7. If the public interest in withholding the information is greater than the public interest disclosing.

Source: Corti et al 2014

There have been a number of high profile cases in which FOI requests have been made against research data being collected by the universities. Data cannot be withheld indefinitely so you should always work under the presumption that if any information is not released the first time round, it can always be requested again. Researchers are therefore advised to detail the ways in which they plan to release their data in a data management plan in order to avoid any unanticipated FOI requests. It is therefore paramount that you plan for FOI and the implications of the Act when creating a data management plan. Writing a plan will also enable you to think more carefully through issues surrounding consent, sharing and ethics. Yet researchers should always remember that the FOI Act can often act as a useful means of data collection in its own right.

At UCL, further guidance on FOI can be found on the Research Integrity pages. If in doubt, researchers are also advised to get in contact with UCL’s specialist Legal Services team.

Further Information

• JISC guide to Freedom of Information and Research Data: http://www.lancaster.ac.uk/users/research/documents/foiresearchdata.pdf

A format is essentially the form your data will take once you collect and archive it. Researchers are strongly advised to think very carefully about the final format their data will take so that it can be preserved for future use.

There are two main two main categories of files – proprietary and non-proprietary formats. Proprietary formats are more limited as they only work with the software provided by the creator of that data. On the other hand, non-proprietary formats can be used by anyone, are usually free of charge and therefore have more utility for future researchers. Plus open formats provide instant and easy access to data. In most cases, you should aim for your data to take the following formats:

1. Non-proprietary
2. Unencrypted
3. Uncompressed
4. Open, documented standard
5. Commonly used by your research community
6. Use common character encodings – ASCII, Unicode, UTF-8

There will always be cases where you will inevitably need to change the format of your data during the course of your research. This is why it is important that you provide further details about the format your data will take in your DMP and any features that may be lost once you convert it for archiving. Open formats may not support all of the original functionality of proprietary formats so you must take steps to hold on to both your raw and converted data sets. Some funders may also have specific requirements surrounding the final form your data should take so be sure to check their policies before committing to any set format.

Further links

• Scientific Data Formats: http://justsolve.archiveteam.org/index.php/Scientific_Data_formats
• LOC: http://www.loc.gov/preservation/resources/rfs/
• Digital Preservation Coalition: https://www.dpconline.org/handbook/technical-solutions-and-tools/file-formats-and-standards

Research software is an increasingly important research output across all disciplines. Alongside this, there is an increasing recognition of the role of software in supporting the reproducibility of papers.  Whilst there is still work to be done on ensuring proper credit for research software there are already steps that researchers can take to make their software more sustainable whether this software consists of a short script used to generate graphs and other visualisations in a paper or a large software project with many collaborators:

• We have guidance available outlining some approaches to preserving and sharing research software.
• UCL benefits from a Research IT Services team who support researchers using software in their research
• Regular drop-ins offer the opportunity to get support from teams from the Library and Research IT Services on issues related to research software alongside data management, open science or policy questions.

The FAIR data principles aim to provide a framework to ensure that research data can be effectively reused. The principles are outlined below alongside recommendations for practically achieving these principles.

Why FAIR data? 

The FAIR Data Principles were developed by a FORCE11 group and originally published in Nature Scientific Data in 2016. The authors argue that ‘Good data management is not a goal in itself, but rather is the key conduit leading to knowledge discovery and innovation, and to subsequent data and knowledge integration and reuse by the community after the data publication process.’

What are the fair data principles?

(more…)

The new General Data Protection Regulation (GDPR) comes into effect in May 2018. It will replace the current Directive and apply to all EU member states without the need for national legislation. The implementation will require comprehensive changes to the way in which organisations, like UCL, collect, use and transfer personal data.

Please see the UCL Data Protection Office’s guidance on the impact of the GDPR on how researchers will seek consent, on privacy notices, data breaches and more.

Following the agreement between the European Commission and United States in 2016, the ‘EU-US Privacy Shield’ is now in force and is therefore the main means of allowing personal data to be transferred to the US.

The EU-US Privacy Shield replaces the invalidated Safe Harbour agreement whilst providing additional obligations to protect personal data, as well as establishing annual monitoring and reporting.

Any new agreement to transfer personal data (including transient transfer) can only be done if the US recipient (this includes universities) has signed up to the Privacy Shield Framework.

Researchers planning on transferring data to the US to a recipient that has not signed up to the Privacy Shield Framework, or who are already working under an existing Safe Harbour agreement, should contact the UCL Data Protection Office (data-protection@ucl.ac.uk).

Further information about the EU-US Privacy Shield can be found on the UCL Data Protection webpages.