Difference: PublicationsCInBTC (11 vs. 12)

Revision 122009-12-11 - EricaHori

Line: 1 to 1
 
META TOPICPARENT name="WebHome"

Publications

Line: 29 to 29
 

Juliana Mafra, Breno Miranda, Juliano Iyoda and Augusto Sampaio.. Test Case Selector: Uma Ferramenta para Seleção de Testes.

Changed:
<
<
In: Brazilian Workshop on Systematic and Automated Software Testing. August, 2009, Gramado, RS, Brazil

>
>
In: Brazilian Workshop on Systematic and Automated Software Testing. August, 2009, Gramado, RS, Brazil [PDF] 

 

Emanuela Gadelha, Patricia Machado and Francisco Neto.

Line: 1049 to 1049
 
META FILEATTACHMENT attachment="Estimating_Test_Execution_Effort_Based_on_Test_Specifications.pdf" attr="h" comment="This thesis was developed by Eduardo Aranha, and advised by Paulo Borba. This research aims to propose test execution effort estimation models that regard the test size and execution complexity measured from the test specification. For that, we identified risk factors related to test projects. Also, we aim to calibrate and evaluate the proposed estimation models through empirical studies on the mobile application domain." date="1242762753" name="Estimating_Test_Execution_Effort_Based_on_Test_Specifications.pdf" path="Estimating_Test_Execution_Effort_Based_on_Test_Specifications.pdf" size="5236135" stream="Estimating_Test_Execution_Effort_Based_on_Test_Specifications.pdf" tmpFilename="/var/tmp/CGItemp53776" user="EricaHori" version="1"
META FILEATTACHMENT attachment="dissertacao_GlauciaPeres.pdf" attr="h" comment="This dissertation was developed by Glaucia Peres, advised by Alexandre Mota and co-advised by Marcelo D'Amorim. Despite the technological advances in languages and tools to support systems development, programmers still deliver software with errors. Several techniques have been proposed to this end – to improve software reliability. Testing is one of them. In fact, software testing is the dominant approach in industry to assure software quality. One way of testing a system is to run it until an incorrect behaviour happens – crash – and so expose a defect. But testing is not cheap. In a typical commercial development organization, the cost of testing, debugging, and verification activities can easily range from 50%25to 75%25of the total development cost [43]. Automation of software testing then becomes a very important mean to reduce this cost. The objective of this work is to propose a black-box testing technique – the Atoms Framework (AF) – to help the detection of crashes. AF is based on automated test scenarios, i.e. sequences of executable steps that have been manually written. We also conducted a set of experiments on cellular phones to check the effectiveness of AF with respect to its capability to find crashes." date="1242828488" name="dissertacao_GlauciaPeres.pdf" path="dissertacao_GlauciaPeres.pdf" size="978796" stream="dissertacao_GlauciaPeres.pdf" tmpFilename="/var/tmp/CGItemp53598" user="EricaHori" version="1"
META FILEATTACHMENT attachment="Dissertacao-Leonardo-Lima-final.pdf" attr="h" comment="This dissertation was developed by Leonardo Lima, and advised by Flávia Barros. This document presents the Class-Test, a tool designed to assist professionals in creating long test suites. In general, each test suite must contain a number of test cases of each type (e.g., Negative tests, Boundary tests, Interaction tests, etc.). This number is determined by test engineers/test designers of the company. One of the main problems faced by testers to create these suites is the time spent on manual classification of the tests pre-selected to compose large suites (around 1,000 tests, for example). The Class-Test is a tool for automatic classification of cases of test, which aims to reduce the effort and time spent in the manual classification of tests. The tool was built based on machine-learning techniques, particularly in the area of text classification. Three classifiers were built using a corpus composed by 879 tests cases, with 191 Boundary test cases, 338 Negative test cases, and 350 Interaction test cases. Each classifier is specialized in only one of these three types of tests. It was necessary to create three classifiers because some test cases can be linked to more than one class at the same time. Two case studies were accomplished. The first case study aimed to evaluate, among the four learning algorithms selected, which one would show better accuracy for the corpus in question (the algorithm SVM - Support Vector Machine showed better performance in this study). The second case study aimed to compare the accuracy of automatic classification versus the classification manual. This work is part of the Test Research Project of CIn-BTC, which is being developed in partnership between CIn-UFPE and Motorola. The general purpose of this project is to automate the generation, selection and evaluation of test cases for mobile applications." date="1242828666" name="Dissertacao-Leonardo-Lima-final.pdf" path="Dissertacao-Leonardo-Lima-final.pdf" size="484997" stream="Dissertacao-Leonardo-Lima-final.pdf" tmpFilename="/var/tmp/CGItemp53319" user="EricaHori" version="1"
Added:
>
>
META FILEATTACHMENT attachment="Test_Case_Selector_Uma_Ferramenta_para_Seleo_de_Testes.pdf" attr="h" comment="" date="1260553759" name="Test_Case_Selector_Uma_Ferramenta_para_Seleo_de_Testes.pdf" path="Test Case Selector Uma Ferramenta para Seleção de Testes.pdf" size="370958" stream="Test Case Selector Uma Ferramenta para Seleção de Testes.pdf" tmpFilename="/var/tmp/CGItemp36457" user="EricaHori" version="1"
 
This site is powered by the TWiki collaboration platformCopyright © 2008-2019 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback

mersin escort bayan adana escort bayan izmit escort ankara escort bursa escort