Intelligent assistant to re-configure parameter-driven systems

Parameter-Driven Systems (PDS) are widely used in commerce for large-scale applications. Reusability is achieved with a PDS design by relocating implicit control structures in the software and the storage of explicit data in database files. This approach can accommodate various user requirements without tedious modification of the software. In order to specify appropriate parameters in a system, knowledge of both business activities and system behaviour are required. For large, complex software packages, this task becomes time consuming and requires specialist knowledge, yet the consistency and correctness still cannot be guaranteed. My research studied the types of knowledge required and agents involved in the PDS customisation. The work also identified the associated problems and constraints. A solution is proposed and implemented as an Intelligent Assistant prototype than a manual approach. Three areas of achievement have been highlighted: 1. The characteristics and problems of maintaining parameter instances in a PDS are defined. It is found that the verification is not complete with the technical/structural knowledge alone, but a context is necessary to provide semantic information and related business activities (thus the implemented parameters) so that mainline functions can relate with each other. 2. A knowledge-based modelling approach has been proposed and demonstrated via a practical implementation. A Specification Language was designed which can model various types of knowledge in a PDS and encapsulate relationships. The Knowledge-Based System (KBS) developed verifies parameters based on the interpreted model of a given context. 3. The performance of the Intelligent Assistant prototype was well received by the domain specialist from the participating organisation. The modelling and KBS approach developed in my research offers considerable promise in solving practical problems in the software industry.

Building reliable distributed systems

[No Abstract]

Preference networks : probabilistic models for recommendation systems

Recommender systems are important to help users select relevant and personalised information over massive amounts of data available. We propose an unified framework called Preference Network (PN) that jointly models various types of domain knowledge for the task of recommendation. The PN is a probabilistic model that systematically combines both content-based filtering and collaborative filtering into a single conditional Markov random field. Once estimated, it serves as a probabilistic database that supports various useful queries such as rating prediction and top-N recommendation. To handle the challenging problem of learning large networks of users and items, we employ a simple but effective pseudo-likelihood with regularisation. Experiments on the movie rating data demonstrate the merits of the PN.

The contribution of residential solar energy systems to the reduction of greenhouse gas emissions in the borough of Queenscliffe

The Borough of Queenscliffe has identified the importance of reducing its greenhouse gas emissions and aims to become a carbon zero municipality by 2020. For a house, suburb or town to become carbon neutral ideally it produces an equivalent amount of energy from renewable resources to that which it consumes. By increasing the number of solar systems, both photovoltaic (PV) and hot water, in the residential sector, greenhouse gas emissions will be reduced. The number of solar systems located in the Borough of Queenscliffe has been estimated and a database of these systems has been created, including the size and panel orientation. The energy generated by each solar system, in addition to the reductions in greenhouse gas emissions, has been calculated for an average year.

Preference Networks: probabilistic models for recommendation systems

Abstract
Recommender systems are important to help users select relevant and personalised information over massive amounts of data available. We propose an unified framework called Preference Network (PN) that jointly models various types of domain knowledge for the task of recommendation. The PN is a probabilistic model that systematically combines both content-based filtering and collaborative filtering into a single conditional
Markov random field. Once estimated, it serves as a probabilistic database that supports various useful queries such as rating prediction and top-N recommendation. To handle the challenging problem of learning large networks of users and items, we employ a simple but effective pseudo-likelihood with regularisation. Experiments on the movie rating data demonstrate the merits of the PN.

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

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Multinodal Load Forecasting in Power Electric Systems using a Neural Network with Radial Basis Function

In this paper we present the results of the use of a methodology for multinodal load forecasting through an artificial neural network-type Multilayer Perceptron, making use of radial basis functions as activation function and the Backpropagation algorithm, as an algorithm to train the network. This methodology allows you to make the prediction at various points in power system, considering different types of consumers (residential, commercial, industrial) of the electric grid, is applied to the problem short-term electric load forecasting (24 hours ahead). We use a database (Centralised Dataset - CDS) provided by the Electricity Commission de New Zealand to this work.

A comparative study of aquifer systems occurring at the Parana sedimentary basin, Brazil: major hydrochemical trends

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

An extended query-the-user tool for expert systems

An extended version of HIER, a query-the-user facility for expert systems is presented. HIER was developed to run over Prolog programs, and has been incorporated to systems that support the design of large and complex applications. The framework of the extended version is described,; as well as the major features of the implementation. An example is included to illustrate the use of the tool, involving the design of a specific database application.

Defects location in power systems components through a dedicated genetic algorithm

In this work the problem of defects location in power systems is formulated through a binary linear programming (BLP) model based on alarms historical database of control and protection devices from the system control center, sets theory of minimal coverage (AI) and protection philosophy adopted by the electric utility. In this model, circuit breaker operations are compared to their expected states in a strictly mathematical manner. For solving this BLP problem, which presents a great number of decision variables, a dedicated Genetic Algorithm (GA), is proposed. Control parameters of the GA, such as crossing over and mutation rates, population size, iterations number and population diversification, are calibrated in order to obtain efficiency and robustness. Results for a test system found in literature, are presented and discussed. © 2004 IEEE.

The plant energy-dissipating mitochondrial systems: Depicting the genomic structure and the expression profiles of the gene families of uncoupling protein and alternative oxidase in monocots and dicots

The simultaneous existence of alternative oxidases and uncoupling proteins in plants has raised the question as to why plants need two energy-dissipating systems with apparently similar physiological functions. A probably complete plant uncoupling protein gene family is described and the expression profiles of this family compared with the multigene family of alternative oxidases in Arabidopsis thaliana and sugarcane (Saccharum sp.) employed as dicot and monocot models, respectively. In total, six uncoupling protein genes, AtPUMP1-6, were recognized within the Arabidopsis genome and five (SsPUMP1-5) in a sugarcane EST database. The recombinant AtPUMP5 protein displayed similar biochemical properties as AtPUMP1. Sugarcane possessed four Arabidopsis AOx1-type orthologues (SsAOx1a-1d); no sugarcane orthologue corresponding to Arabidopsis AOx2-type genes was identified. Phylogenetic and expression analyses suggested that AtAOx1d does not belong to the AOx1-type family but forms a new (AOx3-type) family. Tissue-enriched expression profiling revealed that uncoupling protein genes were expressed more ubiquitously than the alternative oxidase genes. Distinct expression patterns among gene family members were observed between monocots and dicots and during chilling stress. These findings suggest that the members of each energy-dissipating system are subject to different cell or tissue/organ transcriptional regulation. As a result, plants may respond more flexibly to adverse biotic and abiotic conditions, in which oxidative stress is involved. © The Author [2006]. Published by Oxford University Press [on behalf of the Society for Experimental Biology]. All rights reserved.

Application of ontologies in peer-to-peer database with routing queries based on ant colony algorithm

A significant set of information stored in different databases around the world, can be shared through peer-topeer databases. With that, is obtained a large base of knowledge, without the need for large investments because they are used existing databases, as well as the infrastructure in place. However, the structural characteristics of peer-topeer, makes complex the process of finding such information. On the other side, these databases are often heterogeneous in their schemas, but semantically similar in their content. A good peer-to-peer databases systems should allow the user access information from databases scattered across the network and receive only the information really relate to your topic of interest. This paper proposes to use ontologies in peer-to-peer database queries to represent the semantics inherent to the data. The main contribution of this work is enable integration between heterogeneous databases, improve the performance of such queries and use the algorithm of optimization Ant Colony to solve the problem of locating information on peer-to-peer networks, which presents an improve of 18% in results. © 2011 IEEE.

Artificial Immune Systems Applied to the Analysis of Structural Integrity of a Building

This paper presents the application of artificial immune systems for analysis of the structural integrity of a building. Inspired by a biological process, it uses the negative selection algorithm to perform the identification and characterization of structural failure. This paper presents the application of artificial immune systems for analysis of the structural integrity of a building. Inspired by a biological process, it uses the negative selection algorithm to perform the identification and characterization of structural failure. This methodology can assist professionals in the inspection of mechanical and civil structures, to identify and characterize flaws, in order to perform preventative maintenance to ensure the integrity of the structure and decision-making. In order to evaluate the methodology was made modeling a two-story building and several situations were simulated (base-line condition and improper conditions), yielding a database of signs, which were used as input data for the negative selection algorithm. The results obtained by the present method efficiency, robustness and accuracy.

Economic Cost of Air Pollution in Cubatao - SP Based on Health Expenses Related to Diseases of the Respiratory and Circulatory Systems

The aim of this research was to evaluate economic costs of respiratory and circulatory diseases in the municipality of Cubatao, in the state of Sao Paulo, Brazil. Data on hospital admissions and on missed working days due to hospitalization (for age group 14 to 70 years old) from the database of Sistema Unico de Sa de (SUS - Brazilian National Health System) were used. Results: Based on these data, it was calculated that R$ 22.1 million were spent in the period 2000 to 2009 due to diseases of the respiratory and circulatory systems. Part of these expenses can be directly related to the emission of atmospheric pollutants in the city. In order to estimate the costs related to air pollution, data on Cubatao were compared to data from two other municipalities that are also located at the coast side (Guaruja and Peru be), but which have little industrial activity in comparison to Cubatao. It was verified that, in both, average per capita costs were lower when compared to Cubatao, but that this difference has been decreasing in recent years.

A rigorous approach to facilitate and guarantee the correctness of the genetic testing management in human genome information systems

Abstract Background Recent medical and biological technology advances have stimulated the development of new testing systems that have been providing huge, varied amounts of molecular and clinical data. Growing data volumes pose significant challenges for information processing systems in research centers. Additionally, the routines of genomics laboratory are typically characterized by high parallelism in testing and constant procedure changes. Results This paper describes a formal approach to address this challenge through the implementation of a genetic testing management system applied to human genome laboratory. We introduced the Human Genome Research Center Information System (CEGH) in Brazil, a system that is able to support constant changes in human genome testing and can provide patients updated results based on the most recent and validated genetic knowledge. Our approach uses a common repository for process planning to ensure reusability, specification, instantiation, monitoring, and execution of processes, which are defined using a relational database and rigorous control flow specifications based on process algebra (ACP). The main difference between our approach and related works is that we were able to join two important aspects: 1) process scalability achieved through relational database implementation, and 2) correctness of processes using process algebra. Furthermore, the software allows end users to define genetic testing without requiring any knowledge about business process notation or process algebra. Conclusions This paper presents the CEGH information system that is a Laboratory Information Management System (LIMS) based on a formal framework to support genetic testing management for Mendelian disorder studies. We have proved the feasibility and showed usability benefits of a rigorous approach that is able to specify, validate, and perform genetic testing using easy end user interfaces.