DSpace Collection:
http://hdl.handle.net/11624/161
2024-03-20T23:35:22ZIdentificação e contagem de insetos através de técnicas de processamento de imagens.
http://hdl.handle.net/11624/2203
Title: Identificação e contagem de insetos através de técnicas de processamento de imagens.
Authors: Wartchow, Fernando
Abstract: This final work presents a study related to image processing techniques with the support of an open source library called OpenCV for the development of an application to run an automatic insect counting from traps. The insect counting is a procedure to find and allow infestation control in the tobacco companies’ warehouse. Traps that contain a sexual pheromone that attract and glue insects on it are used. The goal is improve the counting process, making it faster and more assertive to determine actions and practices to be taken to treat the infested area. The outcome analyses indicate that is possible to use image processing techniques to perform an automatic insect counting. According the test results, the application achieved 3% variance between automatic and manual counting in a sample of 722 insects.2016-01-01T00:00:00ZAmbiente de recomendação de índices para bancos de dados MySQL.
http://hdl.handle.net/11624/2158
Title: Ambiente de recomendação de índices para bancos de dados MySQL.
Authors: Weiland, Eduardo
Abstract: Index advisor tools are used to assist in the definition of indexes that should be created in a relational database in order to obtain better performance for running queries. Several databases already offer tools for this purpose, such as Microsoft SQL Server, Oracle Database, IBM DB2 and PostgreSQL. This paper proposes the development of an index advisor environment for MySQL databases. The environment analyzes a workload consisting of several SQL queries. This workload is loaded into the tool from an XML file in a predefined format. Each query is interpreted and a set of candidate indexes is generated. The indexes are then created in a
database configured by the user, which should already contain all the tables and data required. Candidate indexes are evaluated by EXPLAIN statements, calculating the cost of all queries using every possible index. The output at the of the execution is a set of recommended indexes that offers the lowest total cost to the workload. The resulting set of indexes showed a considerable performance gain compared to the absence of indexes in the database. However, the result was not conclusive when compared to the existing indexes in the database that was used to run the tests, as all the generated solution were already used by the system.2016-01-01T00:00:00ZUso de técnicas de aprendizado de máquina no auxílio em previsão de resultados de partidas de futebol.
http://hdl.handle.net/11624/2157
Title: Uso de técnicas de aprendizado de máquina no auxílio em previsão de resultados de partidas de futebol.
Authors: Schmidt, Henrique Luis
Abstract: Football is one of the most popular sports in the world, the betting market in this sport is a multibillionaire industry and helping the prediction of a football match is not something easy, because there are many factors that influence it. For all these reasons, the prediction of football matches is an important research problem and the machine learning techniques usage is indicated. According to (MITCHELL, 1997), machine learning aghoritms are specially useful when there is a big database and hidden regularities can be found. In this context, this work goal is to develop a system that can help predicting the result of a football match, having as
an output the match result with greater chances to occur. The methodoly defined to this work involved a bibliographic research to found related works, the analysis of these works to define the techniques, the attributes and the system validation, that was made comparing the real results with the predicted ones. To develop this system, statistical data from matches in the Premier League and hability data from players in the computer game Fifa were used. The techniques Random Forest, Support Vector Machine and Artificial Neural Networks were developed and the best result was 58.77% of accuracy in the match predictions.2017-01-01T00:00:00ZDetecção e análise proativa de anomalias no tráfego de rede.
http://hdl.handle.net/11624/2156
Title: Detecção e análise proativa de anomalias no tráfego de rede.
Authors: Leal, Tiago Silva
Abstract: The computers networks are fundamental in the day-to-day of companies. In particular, the local networks (LANs) represent a vital part of the industry, which over time has generated operational dependence, positive results depend on the correct functioning. This situation, coupled with the use of networks at non-estimated levels, where the pattern of traffic varies greatly, has made it difficult to diagnose and maintain operational networks in situations that deviate from normal behavior patterns. To assist in the protection and availability it is necessary to analyze the traffic to detect possible anomalies. Currently, several solutions based on detection and analysis of network traffic anomalies are found in the literature, but not focused on local networks. Given the above facts, the present work of course completion aims to propose a tool for detection and analysis proactive of anomalies in local network traffic. Therefore, it will be using the detection methodology based on knowledge, together with traffic analysis to provide a database of signatures concerning anomalies known in local networks. The signature base will be a key part of the proposed tool for identifying anomalies in network traffic. This work also presents the goal of using the tool in real network infrastructures, considering small, medium and large companies in the region.2017-01-01T00:00:00Z