Stanley Robson de Medeiros Oliveira

Protecting sensitive knowledge by data sanitization

We address the problem of protecting some sensitive knowledge in transactional databases. The challenge is on protecting actionable knowledge for strategic decisions, but at the same time not losing the great benefit of association rule mining. To accomplish that, we introduce a new, efficient one-scan algorithm that meets privacy protection and accuracy in association rule mining, without putting at risk the effectiveness of the data mining per se.

Algorithms for balancing privacy and knowledge discovery in association rule mining

The discovery of association rules from large databases has proven beneficial for companies since such rules can be very effective in revealing actionable knowledge that leads to strategic decisions. In tandem with this benefit, association rule mining can also pose a threat to privacy protection. The main problem is that from non-sensitive information or unclassified data, one is able to infer sensitive information, including personal information, facts, or even patterns that are not supposed to be disclosed. This scenario reveals a pressing need for techniques that ensure privacy protection, while facilitating proper information accuracy and mining. In this paper, we introduce new algorithms for balancing privacy and knowledge discovery in association rule mining. We show that our algorithms require only two scans, regardless of the database size and the number of restrictive association rules that must be protected. Our performance study compares the effectiveness and scalability of the proposed algorithms and analyzes the fraction of association rules, which are preserved after sanitizing a database. We also report the main results of our performance evaluation and discuss some open research issues.

Achieving Privacy Preservation when Sharing Data for Clustering

In this paper, we address the problem of protecting the underlying attribute values when sharing data for clustering. The challenge is how to meet privacy requirements and guarantee valid clustering results as well. To achieve this dual goal, we propose a novel spatial data transformation method called Rotation-Based Transformation (RBT). The major features of our data transformation are: a) it is independent of any clustering algorithm, b) it has a sound mathematical foundation; c) it is efficient and accurate; and d) it does not rely on intractability hypotheses from algebra and does not require CPU-intensive operations. We show analytically that although the data are transformed to achieve privacy, we can also get accurate clustering results by the safeguard of the global distances between data points.

Secure association rule sharing

The sharing of association rules is often beneficial in industry, but requires privacy safeguards. One may decide to disclose only part of the knowledge and conceal strategic patterns which we call restrictive rules. These restrictive rules must be protected before sharing since they are paramount for strategic decisions and need to remain private. To address this challenging problem, we propose a unified framework for protecting sensitive knowledge before sharing. This framework encompasses: (a) an algorithm that sanitizes restrictive rules, while blocking some inference channels. We validate our algorithm against real and synthetic datasets; (b) a set of metrics to evaluate attacks against sensitive knowledge and the impact of the sanitization. We also introduce a taxonomy of sanitizing algorithms and a taxonomy of attacks against sensitive knowledge.

A privacy-preserving clustering approach toward secure and effective data analysis for business collaboration

The sharing of data has been proven beneficial in data mining applications. However, privacy regulations and other privacy concerns may prevent data owners from sharing information for data analysis. To resolve this challenging problem, data owners must design a solution that meets privacy requirements and guarantees valid data clustering results. To achieve this dual goal, we introduce a new method for privacy-preserving clustering called Dimensionality Reduction-Based Transformation (DRBT). This method relies on the intuition behind random projection to protect the underlying attribute values subjected to cluster analysis. The major features of this method are: (a) it is independent of distance-based clustering algorithms; (b) it has a sound mathematical foundation; and (c) it does not require CPU-intensive operations. We show analytically and empirically that transforming a data set using DRBT, a data owner can achieve privacy preservation and get accurate clustering with a little overhead of communication cost.

A unified framework for protecting sensitive association rules in business collaboration

The sharing of association rules has been proven beneficial in business collaboration, but requires privacy safeguards. One may decide to disclose only part of the knowledge and conceal strategic patterns called sensitive rules. The challenge here is how to protect the sensitive rules without losing the benefit of mining. To address this problem, we propose a unified framework that combines: a set of algorithms to protect sensitive knowledge; retrieval facilities to speed up the process of knowledge protecting; and a set of metrics to evaluate the effectiveness of the proposed algorithms in terms of information loss and private information disclosure.

Mineração de dados e estimativa da mortalidade alta de frangos quando expostos a onda de calor

Heat waves usually result in losses of animal production since they are exposed to thermal stress inducing an increase in mortality and consequent economical losses. Animal science and meteorological databases from the last years contain enough data in the poultry production business to allow the modeling of mortality losses due to heat wave incidence. This research analyzes a database of broiler production associated to climatic data, using data mining techniques such as attribute selection and data classification (decision tree) to model the impact of heat wave incidence on broiler mortality. The temperature and humidity index (THI) was used for screening environmental data. The data mining techniques allowed the development of three comprehensible models for estimating specifically high mortality during broiler production. Two models yielded a classification accuracy of 89.3% by using Principal Component Analysis (PCA) and Wrapper feature selection approaches. Both models obtained a class precision of 0.83 for classifying high mortality. When the feature selection was made by the domain experts, the model accuracy reached 85.7%, while the class precision of high mortality was 0.76. Meteorological data and the calculated THI from meteorological stations were helpful to select the range of harmful environmental conditions for broilers 29 and 42 days old. The data mining techniques were useful for building animal production models.

Mineração de dados para inferência de relações solo-paisagem em mapeamentos digitais de solo

The objective of this work was to develop a methodology for digital soil mapping at a 1:100,000 scale by applying data mining techniques to preexisting relief descriptors and data from pedological and geological maps. A digital database was created from topographic and thematic maps, and allowed the generation of a digital elevation model (DEM) of the Dois Corregos (SP, Brazil) sheet (1:50,000 scale). The slope gradient, slope profile, contour profile, basin contributing area, and diagonal distance to drainage geomorphometric parameters were extracted from the DEM. The matrix which associated this georeferred data was analyzed by means of decision trees within the Weka machine-learning environment, and a model for soil mapping unit prediction was generated. The overall model accuracy increased from 54 to 61% when soil classes with no chances of being predicted were excluded. The association of data mining techniques with geographical information systems produced digital soil maps feasible to be used in studies requiring less detail than those made with the original reference soil maps.

Efficiency of distinct data mining algorithms for classifying stress level in piglets from their vocalization

Entre os desafios da suinocultura no atual mercado competitivo, destaca-se a rastreabilidade do produto, que garante, entre muitos pontos, a questao do bem-estar animal. A vocalizacao e uma ferramenta util para identificar situacoes de estresse em suinos e pode ser usada em registros de bem-estar, em processos de rastreabilidade. Este trabalho teve o objetivo de identificar estresse em leitoes atraves da vocalizacao, classificando esse estresse em tres niveis: sem estresse, estresse moderado e estresse agudo. Foi realizado um experimento em granja comercial da cidade de Holambra-SP, onde se gravou a vocalizacao de vinte leitoes durante o procedimento de castracao, separados em dois grupos: sem anestesia e com anestesia local a base de Lidocaina. Para a captura dos sinais acusticos, foi utilizado um microfone unidirecional conectado a um gravador digital, em que os sinais foram digitalizados a uma frequencia de 44.100 Hz. Para analises dos sinais sonoros, foi usado o software Praat®, e diferentes algoritmos de mineracao dos dados foram aplicados no software Weka®. A selecao de atributos melhorou a acuracia do modelo, sendo que o melhor metodo de selecao de atributos usado foi o Wrapper, enquanto os melhores algoritmos de classificacao foram o k- NN e o Naive Bayes. De acordo com os resultados, foi possivel classificar o nivel de estresse em suinos atraves de sua vocalizacao.

Privacy-Preserving Clustering to Uphold Business Collaboration: A Dimensionality Reduction Based Transformation Approach

While the sharing of data is known to be beneficial in data mining applications and widely acknowledged as advantageous in business, this information sharing can become controversial and thwarted by privacy regulations and other privacy concerns. Data clustering for instance could be more accurate if more information is available, hence the data sharing. Any solution needs to balance the clustering requirements and the privacy issues. Rather than simply hindering data owners from sharing information for data analysis, a solution could be designed to meet privacy requirements and guarantee valid data clustering results. To achieve this dual goal, this article introduces a method for privacy-preserving clustering called dimensionality reduction-based transformation (DRBT). This method relies on the intuition behind random projection to protect the underlying attribute values subjected to cluster analysis. It is shown analytically and empirically that transforming a dataset using DRBT, a data owner can achieve privacy preservation and get accurate clustering with little overhead of communication cost. Such a method presents the following advantages: it is independent of distance-based clustering algorithms, it has a sound mathematical foundation, and it does not require CPU-intensive operations.