Privacy Dynamics: Learning from the Wisdom of Groups

Lead Research Organisation: Imperial College London
Department Name: Dept of Computing

Abstract

We propose to study privacy management by investigating how individuals learn and benefit from their membership of social or functional groups, and how such learning can be automated and incorporated into modern mobile and ubiquitous technologies that increasingly pervade society. We will focus on the privacy concerns of individuals in the context of their use of pervasive technologies, such as Smartphones and personal sensors which share data in the Cloud.

We aim to contribute to research in three areas:

(1) software engineering of adaptive systems that guide their users to manage their privacy;

(2) development of machine learning techniques to alleviate the cognitive and physical load of eliciting and personalising users' privacy requirements; and

(3) empirical investigation of the privacy behaviour of, and in, groups, in the context of both collaboration and conflict.

The ability to control and maintain privacy is central to the preservation of identity. In recent years, social psychologists have made a core distinction between personal identity (which refers to what makes us unique, as individuals, compared to other individuals) and social identity (which refers to our sense of ourselves as members of a social group and the meaning that group has for us). In the latter case, our sense of who we are can be derived from our membership of social groups. Identity is not fixed, but is rather the outcome of a dynamic process. We can move from a personal to a social identity (and back again) depending on the context. We can move between different social identities (for example, as a male, a father, a worker, a football fan, English, British, etc). Identity matters because it provides a prism through which we perceive the world, experience events, decide how to act, and understand our relationships to other people. It tells who is and who is not of us, who is for us and who is against us. Understanding the identity process is therefore key to assessing the impact that privacy and security policies have on people's behaviours. This is essential in order to be able to deliver systems that can express and analyse users' privacy requirements and, at runtime, self-adapt and guide users as they move from context to context.

Broadly speaking, our proposed project asks the following two questions and attempts to answer them from both a social psychology and a computing perspective:

Can privacy be a distributed quality (across 'the group')?
If so, under what conditions might this be the case?

Can the group protect the privacy of the individual?
If so, how does the group manage the privacy-related behaviour of its members?

The research challenges for the project are to devise non-intrusive yet rigorous ways in which to study privacy, both using pervasive technologies (such as life-logging cameras and biometric sensors) and in order to deliver more effective privacy management. At the heart of the project is a hypothesis that individuals are able to better manage their privacy by adopting or learning from the 'wisdom of groups' - we use this term as an acknowledgement of the crowd sourcing movement, also adapted by others in the catchphrase 'wisdom of friends'. Our novelty is in extending this idea to exploit the wisdom of particular subsets of people - groups whose positions and knowledge are more nuanced than a crowd. Our technical challenge is to investigate what we call the privacy dynamics of individuals as they relate to their membership of social, professional or other groups, to develop computational (machine learning) techniques that support such dynamics, and then to deliver privacy management capabilities interactively, autonomously, and adaptively as individuals' contexts change.

Publications

10 25 50
 
Description The research conducted in this project made various research discoveries. A first major contribution is the development of new learning algorithms that are able to extract declarative knowledge from observations about human behaviours. the algorithms have increased expressivity than state-of-the-art systems and they are now de fact systems for learning declarative ASP programs. We have applied these systems to the context of privacy dynamics and have been able to show that it is indeed possible to learn privacy policies and privacy conflicts related to social identity conflicts. These results are currently been explored in the real setting of Facebook and other social networks. The second outcomes is the development of a probabilities learning approach as initially proposed. We are currently using this system for predicting level of sensitivity of pieces of information and use this in a framework for on-line learning of social benefits and privacy risk. These results also find immediate application in the context of social networks. A third contribution has been on the application of more conventional ML solutions to the classification of user behaviour on mobile phone with respect to their privacy awareness behaviour.
Exploitation Route The learning systems developed in this project are general purpose systems. They can be applied to a large variety of problems and they are not specific to the task of learning privacy behaviours. This is one of the advantage of knowledge-driven machine learning versus our conventional machine learning. We envisage that these systems will be used to tackle a variety of learning tasks where the objective is to extract knowledge that can be expressed in English from observations. Current planned pathway to impact is the integration of these algorithms into the IBM Watson cognitive system in order to augment its cognitive capability. Other related and promising area of impacts are in healthcare where learning of hypothesis (diagnosis) is a key task of doctors.
Sectors Communities and Social Services/Policy,Digital/Communication/Information Technologies (including Software),Energy,Healthcare,Transport,Other

URL http://www9.open.ac.uk/PrivacyDynamics/
 
Description The findings provided by this award are mainly in the area of knowledge-absed learning and various applications. One of the applications is in the area of legal reasoning. The outcomes of learning algorithms of legal arguments has been considered by legal experts in Japan. The work was indeed conducted in collaboration with researchers at the NII in Japan, who have joint expertise in computer science and law. We hope that this initial successful discussions with law experts will in the future lead to further impact of the outcomes in society
First Year Of Impact 2015
Sector Government, Democracy and Justice,Other
Impact Types Policy & public services

 
Description SAUSE
Amount £1,330,879 (GBP)
Funding ID EP/R013144/1 
Organisation Engineering and Physical Sciences Research Council (EPSRC) 
Sector Academic/University
Country United Kingdom
Start 04/2018 
End 03/2023
 
Description STRETCH
Amount £1,049,532 (GBP)
Funding ID EP/P01013X/1 
Organisation Engineering and Physical Sciences Research Council (EPSRC) 
Sector Academic/University
Country United Kingdom
Start 04/2017 
End 03/2020
 
Description Collaboration with University of York 
Organisation University of York
Department Centre for Reviews and Dissemination (CRD)
Country United Kingdom 
Sector Academic/University 
PI Contribution Application of probabilistic markov decision process to social networks.
Collaborator Contribution Use of online learning approach to social network
Impact No output yet.
Start Year 2016
 
Description Memorandum of understanding with NII 
Organisation National Institute of Informatics (NII)
Country Japan 
Sector Public 
PI Contribution Applied inductive learning algorithms develop during the EPSRC project to the context of legal reasoning. The collaboration is with Prof. Ken Satoh at the NII who is a computer scientist expert in knowledge representation of legal reasoning.
Collaborator Contribution professor Ken Satoh provided us with legal case studies to study and to use in our learning approach, as well an contributed to the writing up of a paper.
Impact Two papers: 1) Duangtida Athakravi, Dalal Alrajeh, Krysia Broda, Alessandra Russo, Ken Satoh: Inductive Learning Using Constraint-Driven Bias. ILP 2014: 16-32 2) Duangtida Athakravi, Ken Satoh, Mark Law, Krysia Broda, Alessandra Russo: Automated Inference of Rules with Exception from Past Legal Cases Using ASP. LPNMR 2015: 83-96. Legal reasoning Inductive learning
Start Year 2014
 
Description Research collaboration with Miriam Koschate-Reis, University of Exeter, UK 
Organisation University of Exeter
Department School of Psychology
Country United Kingdom 
Sector Academic/University 
PI Contribution Worked together on applying machine learning to two real social network datasets to perform social identity detection.
Collaborator Contribution They provided the social science input and the datasets.
Impact It is multidisciplinary as Miriam is a Lecturer in Lecturer in Social and Organisational Psychology. We are currently writing a paper and an EPSRC reseat proposal. No completed outputs yet.
Start Year 2014
 
Title ILASP 
Description Development of a state of the art learning system. Approach and method have been recognised as student IP by the college. 
IP Reference  
Protection Copyrighted (e.g. software)
Year Protection Granted 2017
Licensed Yes
Impact Formally licensed to college and myself on non-commercial basis.
 
Title Learning system 
Description State of the art ILASP system has been development during the life of this project and the executable is publicly available on https://www.doc.ic.ac.uk/~ml1909/ILASP/. 
Type Of Technology Software 
Year Produced 2016 
Impact Currently under discussion the possibility of an agreement with IBM for using this learning approach to develop cognitive system in health care.