Paulo Roberto Massa Cereda

An Approach to Treat the User’s Preferences About Personal Data

The concern with users’ privacy in the Internet is an argued subject in some areas of knowledge. Mechanisms and actions exist to promote guarantees of user’s privacy; the majority of them, however, present limitations in the privacy control. This paper describes a procedure for the user’s privacy preferences treatment in web systems based on a contextualization model. It offers a personalization choice to the user guaranteeing that his/her privacy preferences are respected. The procedure is applied in a case study and the observed results are presented, evidencing that the privacy personalization improves the user trust in web transactions.

A middleware architecture for adaptive devices

Abstract: The intuitive notion of an adaptive layer enclosing an underlying rule-driven device, thus making it adaptive, does not specify how coupling between the underlying rule-driven device and the adaptive mechanism should happen from an architectural point of view. In this paper, as a means to address component heterogeneity and exposure, we present a middleware architecture for adaptive devices, acting as a support layer between the underlying rule-driven device and the adaptive mechanism. The proposed middleware spontaneously provides aggregation and composition services to both components, making them interoperable and working as a single cohesive unit.

Adaptive Data Mining: Preliminary Studies

This paper presents preliminary studies on a proposal of an adaptive model for subject identification in large data sets. Additionally, an initial proposal of the adaptive data mining concept is introduced as a viable computational solution for problems originated from large data sets.

Approach to Cold-Start Problem in Recommender Systems in the Context of Web-Based Education

In this paper we present an approach to treatment of the Cold-Start Problem in Recommendation System for Environment Education Web. Our approach is based on the concept of Coupled-Learning and Bootstrapping. Based on an initial set of data we apply algorithms traditional machine learning to cooperate with each other, forming various views on its outputs and allowing the data set to be classified incrementally. Thus, it is possible to increase the initial volume of data and to improve the performance of a recommender more instances for analysis. The vast majority of the efforts attack the cold start problem with variations of the CBF algorithm. In our approach, we use the incremental semi-supervised learning based on pairs in order to increase the initial training set in order to allow the generation of more recommendations.

Access Control Model Formalism using Adaptive Automaton

Privacy is an important aspect when modeling computer systems which need to establish relationships among users and their information. To face this issue, many privacy protection mechanisms have been considered. Generic mechanisms deal with the information security of a certain user restricting a system in a way to have only relationships among subjects and objects which comply with this system rules. This paper presents an access control model formalism using adaptive automaton on a auditable privacy system. The adaptive automaton is suitable for this purpose, because of the self-modifications capacity and complex languages recognition. The model using the the proposed formalism is a new approach to privacy protection mechanisms, and is generic enough to be used not just in privacy, but in another classes of problems.

Modeling Pedestrian Dynamics with Adaptive Cellular Automata

Abstract This work presents a proposal for modeling pedestrian dynamics by means of Adaptive Cellular Automata, a dynamically adjustable approach that uses an underlying cellular automata and a set of adaptive functions intended to reconfigure the automata internal structure and behavior according to observable events and rules, making them adaptable to environment changes.

Syntactic analysis of natural language sentences based on rewriting systems and adaptivity

Abstract The intricate, dependent structures found in natural language pose as a challenge for computational processing. Existing approaches resort to either probabilistic models or case-oriented syntactic mappings, leading to unsatisfactory or excessively convoluted grammatical rules. As a means to reduce complexity and offer an incremental, hierarchical approach to the phenomenon of context sensitivity, this paper presents a rule-based rewriting system using adaptive technology for syntactic analysis of sentences in natural language. We provide a detailed description of a sentence with dependent constructs being decomposed into a syntactic tree through successive reductions as a proof of concept.

Instrumenting a Context-free Language Recognizer.

Instrumentation plays a crucial role when building language recognizers, as collected data provide basis for achieving better performance and model improvements, thus offering a balance between time and space, as demanded by practical applications. This paper presents a simple yet functional semiautomatic approach for generating a instrumentation-aware context-free language recognizer, enhanced with hooks, from a grammar written using the Wirth syntax notation. The entire process is aided by a set of command line tools, freely available for download. We also introduce the concept of an instrumentation layer enclosing the underlying recognizer, acting as observer for each computational step and collecting data for later use.

Towards performance-focused implementations of adaptive devices

Abstract: This paper presents instrumentation metrics for adaptive rule-driven devices as a means to obtain performance-focused imple- mentations, from the underlying non-adaptive rule-driven device to the adaptive mechanism, as well as discussions regarding the adaptive behaviour and its corresponding operations, from theoretical and practical points of view.

Resource Recommendation Using Adaptive Automaton

Recommender systems identify preferences of a certain user, as well as the rest of a community, related to the available resources, aiming at providing personalized resources to the users. This paper presents a recommender system using an adaptive automaton to analyze the existing relationships among resources and users, and determine the possible personalized recommendations. An experiment is presented in order to verify and analyze the generated recommendation on the developed system.