Loredana Parasiliti Provenza

State-of-the-art in privacy preserving data mining

We provide here an overview of the new and rapidly emerging research area of privacy preserving data mining. We also propose a classification hierarchy that sets the basis for analyzing the work which has been performed in this context. A detailed review of the work accomplished in this area is also given, along with the coordinates of each work to the classification hierarchy. A brief evaluation is performed, and some initial conclusions are made.

A Framework for Evaluating Privacy Preserving Data Mining Algorithms

Recently, a new class of data mining methods, known as privacy preserving data mining (PPDM) algorithms, has been developed by the research community working on security and knowledge discovery. The aim of these algorithms is the extraction of relevant knowledge from large amount of data, while protecting at the same time sensitive information. Several data mining techniques, incorporating privacy protection mechanisms, have been developed that allow one to hide sensitive itemsets or patterns, before the data mining process is executed. Privacy preserving classification methods, instead, prevent a miner from building a classifier which is able to predict sensitive data. Additionally, privacy preserving clustering techniques have been recently proposed, which distort sensitive numerical attributes, while preserving general features for clustering analysis. A crucial issue is to determine which ones among these privacy-preserving techniques better protect sensitive information. However, this is not the only criteria with respect to which these algorithms can be evaluated. It is also important to assess the quality of the data resulting from the modifications applied by each algorithm, as well as the performance of the algorithms. There is thus the need of identifying a comprehensive set of criteria with respect to which to assess the existing PPDM algorithms and determine which algorithm meets specific requirements.In this paper, we present a first evaluation framework for estimating and comparing different kinds of PPDM algorithms. Then, we apply our criteria to a specific set of algorithms and discuss the evaluation results we obtain. Finally, some considerations about future work and promising directions in the context of privacy preservation in data mining are discussed.

End users as unwitting software developers

The widespread use of personal software systems and the boom of the so-called Web 2.0 is erasing the distinctions between those who create software products and those who use or consume them. End users are increasingly involved in the design and development of the tools they use. Unfortunately, there is a high incidence of errors in applications developed by end users. In this paper, a view on end-user development is outlined, which identifies the communication gap between end users and professional software developers as one main source of errors. The spectrum of users that lie between pure end users and professional developers is examined. In particular, the focus is on a particular type of end users that are very active in shaping software tools to their needs without being aware that they are programming: in short, they are unwitting programmers. Their characteristics and their need of appropriate development techniques and environments are analyzed. Finally, the meta-design participatory approach we have developed is briefly described to show how it fills the communication gap and well supports the activities of unwitting programmers.

A meta-design approach to the development of e-government services

This paper describes a meta-design approach to the development of online services for citizens of a government agency. The goal is to transfer the development of government-to-citizen services from professional software developers to administrative employees, without forcing employees to acquire any programming skills. The approach encompasses two main phases. The first phase analyzes the different perspectives of the stakeholders involved in service creation and usage - employees, citizens, software developers and human-computer interaction specialists - in order to derive a meta-model of e-government services. The latter applies the meta-model to design and develop an end-user development environment that properly supports employees in creating an instance of the service meta-model, which is then automatically interpreted to generate the service pages for citizens. A pilot application of the proposed approach is illustrated with reference to a specific class of e-government services offered by the Brescia Municipality, even though the approach is general enough to be applied to different kinds of e-government services and application domains. The results of the evaluation with a group of municipality employees provide initial feedback from the government field and show how to proceed along this research direction.

DESIGNING CUSTOMIZED AND TAILORABLE VISUAL INTERACTIVE SYSTEMS

This paper presents a novel participatory approach to the design of customized and tailorable visual interactive systems; it includes end users as domain experts in the design team. A design method is described which leads to two different visual specifications, one suitable for end users and the other suitable for software engineers. It is also shown how this second specification is directly mapped to the implementation architecture, based on XML technology. The discussion is supported by the description of an example in the mechanical engineering domain.

End-user development of e-government services through meta-modeling

This paper describes an approach to the end-user development of online services for citizens of a government agency. With reference to a typical government-to-citizen service, the paper discusses how such services are currently structured and provided to citizens, and how their implementation can be transferred from software professionals to administrative personnel, who do not generally possess any programming expertise. The analysis of e-government services is carried out according to different perspectives pertaining to the citizen, the employee, the software engineer and the human-computer interaction expert. This analysis leads to define an abstract service model (a meta-model) and constitutes the first phase of the end-user development approach here proposed. The meta-model can then be used to design an environment for service creation suitable to the competencies and background of the target end-user developers. This design activity constitutes the second phase of the proposed approach.

Advanced visual systems supporting unwitting EUD

The ever increasing use of interactive software systems and the evolution of the World Wide Web into the so-called Web 2.0 determines the rise of new roles for users, who evolve from information consumers to information producers. The distinction between users and designers becomes fuzzy. Users are increasingly involved in the design and development of the tools they use, thus users and developers are not anymore two mutually exclusive groups of people. In this paper types of users that are between pure end users and software developers are analyzed. Some users take a very active role in shaping software tools to their needs, but they do it without being aware of programming, they are unwitting programmers who need appropriate development techniques and environments. A meta-design participatory approach for supporting unwitting end-user development through advanced visual systems is briefly discussed.

A universal design resource for rich Internet applications based on design patterns

Rich Internet applications have removed most of the constraints of Web 1.0 while giving users more responsiveness and advanced browsing and interaction experiences. These new horizons, however, raise many challenges for people with disabilities or using limited hardware and software technologies, whose risk to be excluded from the benefits deriving from advanced web applications. To address this problem, WCAG 2.0 guidelines have been released as the newest World Wide Web Consortium recommendation for accessible web content, and WAI-ARIA is a candidate recommendation which provides reference specifications for accessible rich Internet applications. However, both specifications contain a huge amount of information that often discourages most web designers from dealing with accessibility issues. Moreover, guidelines are suitable and usually adopted to judge a design solution a posteriori, but they do not suggest how to face a design problem constructively. This paper proposes a design pattern language for accessibility. The language can be regarded as a universal design resource for helping web designers create accessible rich Internet applications compliant with the most recent standards. Knowledge representation through design patterns reflects the problem-solving approach usually followed by software and web designers, while pattern organization in a structured language aims to guide web designers throughout the design process. The language has been implemented as an accessible rich Internet application itself, thus allowing designers with disabilities to participate in web design. In order to evaluate the design pattern language, a three-step process was carried out including: (1) a heuristic analysis with a group of human–computer interaction experts, (2) a survey study with a group of web designers, and (3) a validation on the field with two designers who have been requested to apply the language in real design cases.

A multimodal interactive system to create and explore graph structures

This work introduces a multimodal interactive system to create, edit and explore graph structures through direct manipulation operations. The system being designed is based on audio-haptic interaction, supported by visual feedback. A star life design and development cycle was undertaken. The main design choices and early implementation and evaluation results are illustrated.

Signature and Access Control Policies for XML Documents

Information push is today an approach widely used for information dissemination in distributed systems. Under information push, a Web data source periodically (or whenever some relevant event arises) broadcasts data to clients, without the need of an explicit request. In order to make information push usable in a large variety of application domains, it is however important that the authenticity and privacy requirements of both the receiver subjects and information owners be satisfied. Although the problem of confidentiality has been widely investigated, no comparable amount of work has been done for authenticity. In this paper, we propose a model to specify signature policies, specifically conceived for XML data. The model allows the specification of credential-based signature policies, supporting both single and joint signatures. Additionally, we provide an architecture for supporting the generation of selectively encrypted and authenticated XML document, ensuring at the same time the satisfaction of both access control and signature policies.