Trust Management XII

Smartwatches, which contain an accelerometer and gyroscope, have recently been used to implement gait/activity-based biometrics. However, many research questions have not been addressed in the prior work such as the training and test data was collected in the same day from a limited dataset, using unrealistic activities (e.g., punch) and/or the authors did not carry out any particular study to identify the most discriminative features. This paper aims to highlight the impact of these factors on the biometric performance. The acceleration and gyroscope data of the gait and game activity was captured from 60 users over multiple days, which resulted in a totally of 24 h of the user’s movement. Segment-based approach was used to divide the time-series acceleration and gyroscope data. When the cross-day evaluation was applied, the best obtained EER was 0.69%, and 4.54% for the walking and game activities respectively. The EERs were significantly reduced into 0.05% and 2.35% for the above activities by introducing the majority voting schema. These results were obtained by utilizing a novel feature selection process in which the system minimizing the number of features and maximizing the discriminative information. The results have shown that smartwatch-based activity recognition has significant potential to recognize individuals in a continuous and user friendly approach.

Privacy policies go largely unread as they are not standardized, often written in jargon, and frequently long. Several attempts have been made to simplify and improve readability with varying degrees of success. This paper looks at keyword extraction, comparing human extraction to natural language algorithms as a first step in building a taxonomy for creating an ontology (a key tool in improving access and usability of privacy policies).

In this paper, we present two alternatives to using costly domain experts are used to perform keyword extraction: trained participants (non-domain experts) read and extracted keywords from online privacy policies; and second, supervised and unsupervised learning algorithms extracted keywords. Results show that supervised learning algorithm outperform unsupervised learning algorithms over a large corpus of 631 policies, and that trained participants outperform the algorithms, but at a much higher cost.

Trust is a fundamental element of any social network. However, despite numerous studies on trust, few have conducted studies across disciplines to provide a complete picture of the different dimensions of trustworthiness, such as integrity, competence and benevolence. In this paper, we focus on two of these dimensions, competence and benevolence. We propose techniques to evaluate the competence of the trustee in specific situations and infer the benevolence of the tustee towards the trustor when the trust evaluation is made. Moreover, we evaluate both competence and benevolence on the perceived trustworthiness of the trustee, taking into consideration the development of the relationship between the trustor and the trustee over time. We identified different stages in this relationship development and use them to evaluate trustworthiness of trustee in the absence of evidence that could be used to evaluate trustworthiness. Finally, we set an experimental scenario implemented as an agent-based model to evaluate our approach. The results obtained from these experiments show that the proposed techniques can improve the reliability of the estimation of the trustworthiness of the agents.

The information society is building on data and the software required to collect and analyse these data, which means that the trustworthiness of these data and software systems is crucially important for the development of society as a whole. Efforts to establish the trustworthiness of software typically include parameters, such as security, reliability, maintainability, correctness and robustness.

In this paper we explore ways to determine the trustworthiness of software, in particular code where some of the constituent components are externally sourced, e.g. through crowd sourcing and open software systems. We examine different quality parameters that we believe define key quality indicators for trustworthy software and define CodeTrust, which is a content based trust metric for software.

We present the design and evaluation of a research prototype that implements the proposed metric, and show the results of preliminary evaluations of CodeTrust using well known open source software projects.

The evaluation of geospatial data quality and trustworthiness presents a major challenge to geospatial data users when making a dataset selection decision. Part of the problem arises from the inconsistent and patchy nature of data quality information, which makes intercomparison very difficult. Over recent years, the production and availability of geospatial data has significantly increased, facilitated by the recent explosion of Web-based catalogues, portals, standards and services, and by initiatives such as INSPIRE and GEOSS. Despite this significant growth in availability of geospatial data and the fact that geospatial datasets can, in many respects, be considered commercial products that are available for purchase online, consumer trust has to date received relatively little attention in the GIS domain.

In this paper, we discuss how concepts of trust, trust models, and trust indicators (largely derived from B2C e-Commerce) apply to the GIS domain and to geospatial data selection and use. Our research aim is to support data users in more efficient and effective geospatial dataset selection on the basis of quality, trustworthiness and fitness for purpose. To achieve this, we propose a GEO label – a decision support mechanism that visually summarises availability of key geospatial data informational aspects. We also present a Web service that was developed to support generation of dynamic GEO label representations for datasets by combining producer metadata (from standard catalogues or other published locations) with structured user feedback.

Social networks that use geolocalization enable receiving data from users in order to provide information based on their collective experience. Specifically, this article is interested in the social network Waze, a real-time navigation application for drivers. This application uses methods for identifying users that are open and free, where people are able to hide their identity by using a pseudonym. In this context, malicious behaviors can emerge, endangering the quality of the reports on which the application is based. We propose a method to detect malicious behavior on Waze, which crawls information from the application, aggregates it and models the data relationships in graphs. Using this model the data is analyzed according to the size of the graph: for large interaction graphs, we use a Sybil detection technique, while for small graphs we propose the use of a threshold-based mechanism to detect targeted behaviors. The results show that it is complex to use the large-scale Sybil attack detection techniques due to parameter tuning. However, good success rates can be achieved to tag users as honest and malicious if there are a small number of interactions between these groups of users. On the other hand, for small graphs, a straightforward analysis can be performed, since the graphs are sparse and the users have a limited number of connections between them, making clear the presence of outliers.

Computational trust is the digital counterpart of the human notion of trust as applied in social systems. Its main purpose is to improve the reliability of interactions in online communities and of knowledge transfer in information management systems. Trust models are typically composed of two parts: a trust computing part and a trust manipulation part. The former serves the purpose of gathering relevant information and then use it to compute initial trust values; the latter takes the initial trust values as granted and manipulates them for specific purposes, like, e.g., aggregation and propagation of trust, which are at the base of a notion of reputation. While trust manipulation is widely studied, very little attention is paid to the trust computing part. In this paper, we propose a formal language with which we can reason about knowledge, trust and their interaction. Specifically, in this setting it is possible to put into direct dependence possessed knowledge with values estimating trust, distrust, and uncertainty, which can then be used to feed any trust manipulation component of computational trust models.

Information has been an essential element in the development of collaborative and cooperative models. From decision making to the attainment of varying goals, people have been relatively adept at making judgments about the trustworthiness of information, based on knowledge and understanding of a normative model of information. However, recent events, for example in elections and referenda, have stretched the ability of people to be able to measure the veracity and trustworthiness of information online. The result has been an erosion of trust in information online, its source, its value and the ability to objectively determine the trustworthiness of a piece of information, a situation made more complex by social networks, since social media have made the spread of (potentially untrustworthy) information easier and faster. We believe that this exacerbated the need for assisting humans in their judgment of the trustworthiness of information. We have begun working on a social cognitive construct: a trust model for information. In this paper we outline the problems and the beginnings of our trust model and highlight future work.

In this work in progress paper, we present a description of a new view of privacy in public, examining how it is possible to ascertain the privacy levels of individuals in context and in groups, and different ways of visualising these Public Privacy levels. We examine how awareness of one’s Public Privacy may have an impact on behaviour and privacy protection in general, and propose future work to examine the concept in more detail.

As the size, complexity and ubiquity of socio-technical systems increases, there is a concomitant expectation that humans will have to establish and maintain long-lasting ‘relationships’ with many types of digital artefact: for example with humanoid robots, driverless cars or software agents running on ‘smart’ devices. Rather than being limited to one-off interactions, these relationships will continue over longer time frames, correspondingly increasing the likelihood of errors occurring from numerous causes. When digital errors occur, often complete human mistrust and distrust is the outcome. The situation is exacerbated when the computer can make no act of reparation and no avenue of forgiveness is open to the human. In the pursuit of designing long-lasting socio-technical systems that are fit-for purpose, this position paper reviews past work in relevant social concepts and, based on the sociological theory of social constructivism, proposes a new approach to the joint human-computer construction of a “shared reality”.