Towards a professional code of ethics for data mining

Click here to sign up to the 'Developing best practice in data analytics' event

Click here to sign up to the ‘Developing best practice in data analytics’ event

The Society of Data Miners is pleased to announce ‘Developing best practice in data analytics‘; the first in a series of meetings for consultation and discussion on a proposed code of ethics for the data mining profession.

Data mining and data analysis have begun to play an enormous role in social and economic life, and that role is destined to increase dramatically in the coming years. The growth of methods and technologies and the proliferation of applications have not, however, been accompanied by appropriate reflection on the ethical responsibilities of practitioners actually engaged in analysing and often re-analysing data, sometimes data of a highly confidential nature. Is it possible to rely on the consent of the data subject? Can that consent be granted meaningfully? Can consent extend beyond the immediate purposes for which the data has been collected? What should, or should not, be encouraged in the day-to-day practice of data analysis?

These are questions in urgent need of answers; the Society of Data Miners will conduct a series of meetings to obtain a broad knowledge-base to draft a code addressing the ethical issues which face the data analytics profession.

The first of these meetings will take the form of a panel discussion, involving senior experts on issues of data ethics:

Hetan Shah is Executive Director of the Royal Statistical Society, a charity with over 8,000 members worldwide with a vision to put data at the heart of understanding and decision-making. He will talk about the wide range of ethical issues which emerge in the data science landscape, and argue that multiple mechanisms are needed to deal with this, including a Council for Data Ethics.

Mariarosaria Taddeo is a Researcher at the Oxford Internet Institute, University of Oxford, and Faculty Fellow at the Alan Turing Institute. Her recent work focuses on data ethics, the ethical analysis of cyber security practices, and information conflicts. Her area of expertise is Information and Computer Ethics, and she has also worked on issues concerning Philosophy of Information, Epistemology, and Philosophy of AI.

Mariarosaria will focus on the relevance of Data Ethics to support the development and employment of data science for the good of society. Data science provides huge opportunities to improve private and public life, but this potentially highly positive impact is coupled to significant ethical challenges. Focal points of discussion will be (i) the ethics of data: how to protect privacy, foster trust, and respect transparency while harnessing the value of data; (ii) the ethics of algorithms: focusing on issues such as accountability and responsibility as well as ethical auditing for algorithms to ensure that ethical consequences can be identified and averted; (iii) the ethics of practises: what are the responsibilities of data scientists and practitioners in this area.

Richard Beaumont is Product Manager at OneTrust, providers of privacy management software to help businesses meet legal obligations for privacy and data protection around the world, supporting the principles of Privacy by Design, and enabling forward thinking brands to turn privacy into a competitive advantage. Prior to this he was responsible for the Optanon cookie compliance platform, purchased this year by OneTrust, helping clients in a field where privacy compliance decisions clash directly with the desire of businesses to know as much as possible about their digital customers.

Richard will discuss questions such as: Just because you can, does it mean you should? When it comes to privacy, do you go with the letter of the law, or embrace its spirit? Sail as close to the wind as you can, or play it safe? Is transparency a risk to be managed? Are privacy policies an opportunity for obfuscation? Or are you looking to be clear and open, while trying to persuade about the benefits of data usage.

Hugh Lawson-Tancred has worked on both the practical and theoretical side of data ethics. He combines extensive experience in applied natural language processing with researching the philosophical background to the rapid advance in data science and AI, for the optimal integration of these technologies into human society.

Hugh will primarily be concentrating on the issue of deanonymisation and the ethical cost of collecting, holding and analysing large datasets. Hugh will also summarise the work of the Society so far on drafting a code of ethics, and the proposed series of consultative meetings.

Tom Khabaza is the Chairman of the Society of Data Miners, and a data analytics practitioner of 25 years standing. Tom will chair the panel discussion, hoping to elicit a draft list of issues that the Society should address in its code of ethics.

To attend this exciting event on Monday 5 December 2016 at London’s Digital Catapult Centre, please sign up to attend here. For more news on personal data and trust, make sure to follow us on Twitter @PDTNetwork