Marlon Mejias - Academia.edu (original) (raw)
Uploads
Papers by Marlon Mejias
iiis.org
In this paper, we evaluate the performance of a mixture of software reliability growth models (Su... more In this paper, we evaluate the performance of a mixture of software reliability growth models (Super models) against the original base models. The base models chosen are four well known nonhomogeneous Poisson process (NHPP) growth models proposed in literature. We use the method of maximum likelihood for fitting these models to data.
elrond.informatik.tu-freiberg.de
presents a novel approach to feeding the CS pipeline, by creating a program that includes cultura... more presents a novel approach to feeding the CS pipeline, by creating a program that includes culturally-relevant pedagogy and experiences to teach middle-school students computer science fundamentals, such as algorithms, problem solving, programming, and computational thinking. Through this unique program, which includes parental engagement, students are able to understand the broad range of opportunities in CS, develop basic competencies in CS fundamentals, and maintain a positive perception of CS.
Systems Engineering, 2008 …, 2008
... Thomas A. Mazzuchi George Washington University Washington DC USA mazzu@gwu.eduMarlon Mejias ... more ... Thomas A. Mazzuchi George Washington University Washington DC USA mazzu@gwu.eduMarlon Mejias George Washington University Washington DC USA mmejias@gwu.edu Abstract ...Mills et al. [20] suggest that by using a more formal approach (using mathematical 439 ...
work
This is the continuation of research addressing the performance of a subset of a specific class o... more This is the continuation of research addressing the performance of a subset of a specific class of software reliability models -the nonhomogeneous Poisson Process (NHPP) models. The object of this paper is to determine the reliability of software during the early stages of the system engineering period using software reliability growth models (SRGMs). Actual software failure data are analyzed with the early stage of system test being represented as the 30% percentile of the total test time of each dataset. The study takes three well known NHPP models and determines their ability to estimate reliability during the defined early stages of the system testing period. The predictive quality of each of the models on test is examined. The method of parameter estimation for the models is the Maximum Likelihood Method. Results of the study also address the reliability growth of the failure data for the datasets on test.
iiis.org
In this paper, we evaluate the performance of a mixture of software reliability growth models (Su... more In this paper, we evaluate the performance of a mixture of software reliability growth models (Super models) against the original base models. The base models chosen are four well known nonhomogeneous Poisson process (NHPP) growth models proposed in literature. We use the method of maximum likelihood for fitting these models to data.
elrond.informatik.tu-freiberg.de
presents a novel approach to feeding the CS pipeline, by creating a program that includes cultura... more presents a novel approach to feeding the CS pipeline, by creating a program that includes culturally-relevant pedagogy and experiences to teach middle-school students computer science fundamentals, such as algorithms, problem solving, programming, and computational thinking. Through this unique program, which includes parental engagement, students are able to understand the broad range of opportunities in CS, develop basic competencies in CS fundamentals, and maintain a positive perception of CS.
Systems Engineering, 2008 …, 2008
... Thomas A. Mazzuchi George Washington University Washington DC USA mazzu@gwu.eduMarlon Mejias ... more ... Thomas A. Mazzuchi George Washington University Washington DC USA mazzu@gwu.eduMarlon Mejias George Washington University Washington DC USA mmejias@gwu.edu Abstract ...Mills et al. [20] suggest that by using a more formal approach (using mathematical 439 ...
work
This is the continuation of research addressing the performance of a subset of a specific class o... more This is the continuation of research addressing the performance of a subset of a specific class of software reliability models -the nonhomogeneous Poisson Process (NHPP) models. The object of this paper is to determine the reliability of software during the early stages of the system engineering period using software reliability growth models (SRGMs). Actual software failure data are analyzed with the early stage of system test being represented as the 30% percentile of the total test time of each dataset. The study takes three well known NHPP models and determines their ability to estimate reliability during the defined early stages of the system testing period. The predictive quality of each of the models on test is examined. The method of parameter estimation for the models is the Maximum Likelihood Method. Results of the study also address the reliability growth of the failure data for the datasets on test.