Supporting the learning of rapid application development in a database environment

Intelligent Tutoring Systems (ITSs) are computer systems designed to provide individualised help to students, learning in a problem solving context. The difference between an ITS and a Computer Assisted Instruction (CAI) system is that an ITS has a Student Model which allows it to provide a better educational environment. The Student Model contains information on what the student knows, and does not know, about the domain being learnt, as well as other personal characteristics such as preferred learning style. This research has resulted in the design and development of a new ITS: Personal Access Tutor (PAT). PAT is an ITS that helps students to learn Rapid Application Development in a database environment. More specifically, PAT focuses on helping students to learn how to create forms and reports in Microsoft Access. To provide an augmented learning environment, PAT’s architecture is different to most other ITSs. Instead of having a simulation, PAT uses a widelyused database development environment (Microsoft Access). This enables the students to ask for help, while developing real applications using real database software. As part of this research, I designed and created the knowledge base required for PAT. This contains four models: the domain, student, tutoring and exercises models. The Instructional Expert I created for PAT provides individualised help to the students to help them correctly finish each exercise, and also proposes the next exercise that a student should work on. PAT was evaluated by students enrolled in the Databases subject at QUT, and by staff members involved in teaching the subject. The results of the evaluation were positive and are discussed in the thesis.

Typical load profiles in the smart grid context – a clustering methods comparison

The present research paper presents five different clustering methods to identify typical load profiles of medium voltage (MV) electricity consumers. These methods are intended to be used in a smart grid environment to extract useful knowledge about customer’s behaviour. The obtained knowledge can be used to support a decision tool, not only for utilities but also for consumers. Load profiles can be used by the utilities to identify the aspects that cause system load peaks and enable the development of specific contracts with their customers. The framework presented throughout the paper consists in several steps, namely the pre-processing data phase, clustering algorithms application and the evaluation of the quality of the partition, which is supported by cluster validity indices. The process ends with the analysis of the discovered knowledge. To validate the proposed framework, a case study with a real database of 208 MV consumers is used.

A data-mining based methodology for wind forecasting

In many countries the use of renewable energy is increasing due to the introduction of new energy and environmental policies. Thus, the focus on the efficient integration of renewable energy into electric power systems is becoming extremely important. Several European countries have already achieved high penetration of wind based electricity generation and are gradually evolving towards intensive use of this generation technology. The introduction of wind based generation in power systems poses new challenges for the power system operators. This is mainly due to the variability and uncertainty in weather conditions and, consequently, in the wind based generation. In order to deal with this uncertainty and to improve the power system efficiency, adequate wind forecasting tools must be used. This paper proposes a data-mining-based methodology for very short-term wind forecasting, which is suitable to deal with large real databases. The paper includes a case study based on a real database regarding the last three years of wind speed, and results for wind speed forecasting at 5 minutes intervals.

A Data Mining Framework for Electric Load Profiling

This paper consists in the characterization of medium voltage (MV) electric power consumers based on a data clustering approach. It is intended to identify typical load profiles by selecting the best partition of a power consumption database among a pool of data partitions produced by several clustering algorithms. The best partition is selected using several cluster validity indices. These methods are intended to be used in a smart grid environment to extract useful knowledge about customers’ behavior. The data-mining-based methodology presented throughout the paper consists in several steps, namely the pre-processing data phase, clustering algorithms application and the evaluation of the quality of the partitions. To validate our approach, a case study with a real database of 1.022 MV consumers was used.

A Data-mining-based Methodology to support MV Electricity Customers' Characterization

This paper presents an electricity medium voltage (MV) customer characterization framework supportedby knowledge discovery in database (KDD). The main idea is to identify typical load profiles (TLP) of MVconsumers and to develop a rule set for the automatic classification of new consumers. To achieve ourgoal a methodology is proposed consisting of several steps: data pre-processing; application of severalclustering algorithms to segment the daily load profiles; selection of the best partition, corresponding tothe best consumers’ segmentation, based on the assessments of several clustering validity indices; andfinally, a classification model is built based on the resulting clusters. To validate the proposed framework,a case study which includes a real database of MV consumers is performed.

Load Profiling Tool to Support Smart Grid Operation Scenarios

This paper presents the characterization of high voltage (HV) electric power consumers based on a data clustering approach. The typical load profiles (TLP) are obtained selecting the best partition of a power consumption database among a pool of data partitions produced by several clustering algorithms. The choice of the best partition is supported using several cluster validity indices. The proposed data-mining (DM) based methodology, that includes all steps presented in the process of knowledge discovery in databases (KDD), presents an automatic data treatment application in order to preprocess the initial database in an automatic way, allowing time saving and better accuracy during this phase. These methods are intended to be used in a smart grid environment to extract useful knowledge about customers’ consumption behavior. To validate our approach, a case study with a real database of 185 HV consumers was used.

Short-term Wind Speed Forecasting using Support Vector Machines

Wind speed forecasting has been becoming an important field of research to support the electricity industry mainly due to the increasing use of distributed energy sources, largely based on renewable sources. This type of electricity generation is highly dependent on the weather conditions variability, particularly the variability of the wind speed. Therefore, accurate wind power forecasting models are required to the operation and planning of wind plants and power systems. A Support Vector Machines (SVM) model for short-term wind speed is proposed and its performance is evaluated and compared with several artificial neural network (ANN) based approaches. A case study based on a real database regarding 3 years for predicting wind speed at 5 minutes intervals is presented.

Detection and Identification of Abnormalities in Customer Consumptions in Power Distribution Systems

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Caracterização geométrica e parametrização de depósitos transacionais recentes e sua aplicação na modelagem de reservatórios petrolíferos

In this beginning of the XXI century, the Geology moves for new ways that demand a capacity to work with different information and new tools. It is within this context that the analog characterization has important in the prediction and understanding the lateral changes in the geometry and facies distribution. In the present work was developed a methodology for integration the geological and geophysical data in transitional recent deposits, the modeling of petroliferous reservoirs, the volume calculation and the uncertainties associate with this volume. For this purpose it was carried planialtimetric and geophysics (Ground Penetrating Radar) surveys in three areas of the Parnaíba River. With this information, it was possible to visualize the overlap of different estuary channels and make the delimitation of the channel geometry (width and thickness). For three-dimensional visualization and modeling were used two of the main reservoirs modeling software. These studies were performed with the collected parameters and the data of two reservoirs. The first was created with the Potiguar Basin wells data existents in the literature and corresponding to Açu IV unit. In the second case was used a real database of the Northern Sea. In the procedures of reservoirs modeling different workflows were created and generated five study cases with their volume calculation. Afterwards an analysis was realized to quantify the uncertainties in the geological modeling and their influence in the volume. This analysis was oriented to test the generating see and the analogous data use in the model construction

3D geovisualisation techniques applied in spatial data mining

The increase in the number of spatial data collected has motivated the development of geovisualisation techniques, aiming to provide an important resource to support the extraction of knowledge and decision making. One of these techniques are 3D graphs, which provides a dynamic and flexible increase of the results analysis obtained by the spatial data mining algorithms, principally when there are incidences of georeferenced objects in a same local. This work presented as an original contribution the potentialisation of visual resources in a computational environment of spatial data mining and, afterwards, the efficiency of these techniques is demonstrated with the use of a real database. The application has shown to be very interesting in interpreting obtained results, such as patterns that occurred in a same locality and to provide support for activities which could be done as from the visualisation of results. © 2013 Springer-Verlag.

Estimação do número de internações por doenças respiratórias utilizando lógica fuzzy tipo-2 intervalar

In this work was developed a fuzzy computational model type-2 predictive interval, using the software of the type-2 fuzzy MATLAB toolbox, the final idea is to estimate the number of hospitalizations of patients with respiratory diseases. The interest in the creation of this model is to assist in decision makeshift hospital environment, where there are no medical or professional equipment available to provide the care that the population need. It began working with the study of fuzzy logic, the fuzzy inference system and fuzzy toolbox. Through a real database provided by the Departamento de Informática do Sistema Único de Saúde (DATASUS) and Companhia de Tecnologia de Saneamento Básico (CETESB), was possible to start the model. The analyzed database is composed of the number of patients admitted with respiratory diseases a day for the public hospital in São José dos Campos, during the year 2009 and by factors such as PM10, SO2, wind and humidity. These factors were analyzed as input variables and, through these, is possible to get the number of admissions a day, which is the output variable of the model. For data analysis we used the fuzzy control method type-2 Mamdani. In the following steps the performance developed in this work was compared with the performance of the same model using fuzzy logic type-1. Finally, the validity of the models was estimated by the ROC curve

Uma nova arquitetura para Internet das Coisas com análise e reconhecimento de padrões e processamento com Big Data.

A Internet das Coisas é um novo paradigma de comunicação que estende o mundo virtual (Internet) para o mundo real com a interface e interação entre objetos. Ela possuirá um grande número de dispositivos heteregôneos interconectados, que deverá gerar um grande volume de dados. Um dos importantes desafios para seu desenvolvimento é se guardar e processar esse grande volume de dados em aceitáveis intervalos de tempo. Esta pesquisa endereça esse desafio, com a introdução de serviços de análise e reconhecimento de padrões nas camadas inferiores do modelo de para Internet das Coisas, que procura reduzir o processamento nas camadas superiores. Na pesquisa foram analisados os modelos de referência para Internet das Coisas e plataformas para desenvolvimento de aplicações nesse contexto. A nova arquitetura de implementada estende o LinkSmart Middeware pela introdução de um módulo para reconhecimento de padrões, implementa algoritmos para estimação de valores, detecção de outliers e descoberta de grupos nos dados brutos, oriundos de origens de dados. O novo módulo foi integrado à plataforma para Big Data Hadoop e usa as implementações algorítmicas do framework Mahout. Este trabalho destaca a importância da comunicação cross layer integrada à essa nova arquitetura. Nos experimentos desenvolvidos na pesquisa foram utilizadas bases de dados reais, provenientes do projeto Smart Santander, de modo a validar da nova arquitetura de IoT integrada aos serviços de análise e reconhecimento de padrões e a comunicação cross-layer.

Modelos em provisão de sinistros

Part of the work of an insurance company is to keep claims reserves, which is known as the technical reserves, in order to mitigate the risk inherent in their activities and comply with the legal obligations. There are several methods for estimate the claims reserves, deterministics and stochastics methods. One of the most used method is the deterministic method Chain Ladder, of simple application. However, the deterministics methods produce only point estimates, for which the stochastics methods have become increasingly popular because they are capable of producing interval estimates, measuring the variability inherent in the technical reserves. In this study the deterministics methods (Grossing Up, Link Ratio and Chain Ladder) and stochastics (Thomas Mack and Bootstrap associated with Overdispersed Poisson model) will be applied to estimate the claims reserves derived from automobile material damage occurred until December 2012. The data used in this research is based on a real database provided by AXA Portugal. The comparison of results obtained by different methods is hereby presented.

Secure Concurrency Control in Firm Real-Time Database Systems

Many real-time database applications arise in electronic financial services, safety-critical installations and military systems where enforcing security is crucial to the success of the enterprise. For real-time database systems supporting applications with firm deadlines, we investigate here the performance implications, in terms of killed transactions, of guaranteeing multilevel secrecy. In particular, we focus on the concurrency control (CC) aspects of this issue. Our main contributions are the following: First, we identify which among the previously proposed real-time CC protocols are capable of providing covert-channel-free security. Second, using a detailed simulation model, we profile the real-time performance of a representative set of these secure CC protocols for a variety of security-classified workloads and system configurations. Our experiments show that a prioritized optimistic CC protocol, OPT-WAIT, provides the best overall performance. Third, we propose and evaluate a novel "dual-CC" approach that allows the real-time database system to simultaneously use different CC mechanisms for guaranteeing security and for improving real-time performance. By appropriately choosing these different mechanisms, concurrency control protocols that provide even better performance than OPT-WAIT are designed. Finally, we propose and evaluate GUARD, an adaptive admission-control policy designed to provide fairness with respect to the distribution of killed transactions across security levels. Our experiments show that GUARD efficiently provides close to ideal fairness for real-time applications that can tolerate covert channel bandwidths of upto one bit per second.