5,346 research outputs found
Analisis perbandingan uji autokorelasi Durbin-Watson dan Breusch-Godfrey
No full textINDONESIA:
Metode Durbin-Watson dan metode Breusch-Godfrey merupakan dua metode yang digunakan untuk menguji autokorelasi yang merupakan gangguan pada fungsi yang berupa korelasi di antara variabel error.
Tujuan dari penelitian ini adalah untuk membandingkan
uji autokorelasi dengan metode Durbin-Watson dan metode
Breusch-Godfrey. Metode penelitian dalam skripsi ini adalah
metode penelitian pustaka (library research), langkah langkah yang dilakukan dalam penelitian ini adalah sebagai berikut: menganalisis metode Durbin-Watson dan metode BreuschGodfrey. Metode Durbin-Watson dilakukan dengan menentukan error dari model regresi, menghitung nilai d Durbin-Watson, setelah nilai hitung statistik d diketahui, kemudian dibandingkan dengan batas atas atau upper bound (dU) dan batas bawah atau lower bound (dL) yang tertera dalam tabel Durbin-Watson, kemudian mengambil keputusan. Metode Breusch-Godfrey dilakukan dengan menentukan error dari model regresi, kemudian meregresikan variabel error menggunakan autoregressive model orde ρ kemudian mengambil keputusan.
Berdasarkan hasil pembahasan pada penelitian ini, diperoleh hasil perbandingan uji autokorelasi dengan metode Durbin-Watson dan Breusch-Godfrey pada data model regresi linier berganda menunjukkan bahwa ketelitian menguji autokorelasi dengan metode Breusch-Godfrey lebih mendekati dalam menguji adanya autokorelasi dari pada metode DurbinWatson
ENGLISH :
Durbin-Watson and Breusch-Godfrey methods are two methods used to test the autocorrelation of the disturbances in the
form of the correlation function between the error variables.
The purpose of this study was to compare the test of
autocorrelation with the Durbin-Watson and Breusch-Godfrey
methods. The methods of research that used in this thesis is the library research, the undertaken steps in this study are as follows: analyza Durbin-Watson and Breusch-Godfrey methods. Durbin-Watson method is done by determining the error of the regression model, calculate the value of Durbin-Watson d, after statistically calculated value d is known, then compare it with the upper bound (dU) and the lower bound (dL) listed in the table DurbinWatson, then take a decision. Breusch-Godfrey's method is done by determining the error of the regression model, then regress error variables using autoregressive model of order ρ, then make a decision.
Based on the discussion in this study, the comparison of
test results obtained with the autocorrelation of Durbin-Watson and Breusch-Godfrey methods on the data model of multiple linear regression showed that the accuracy of autocorrelation using Breusch-Godfrey's method is closer to the test for autocorrelation than Durbin-Watson’s metho
Utilising Deep Learning Models for the Surface Registration Problem in HoloNav
No full textSurface Registration is a registration problem that handles the registration of two similar surfaces. In most research that utilises Deep Learning (DL) models to handle surface registration two theories are investigated; the first being whether surfaces sampled from the same origin can be registered together, and the second theory being whether the models can register Point Clouds with low overlapping data for utilisation in Simultaneous Localisation and Mapping (SLAM) applications. However, the surface registration to be utilised in the HoloNav Augmented Reality (AR) navigation system will utilise Point Clouds sampled from different origins with a high overlap ratio. This research, therefore, aims to determine the viability of DL methods for surface registration in HoloNav data. To determine the viability, rotation and translation errors in the match were used, with the aforementioned metrics later being evaluated manually with the utilisation of a visualiser. The results indicate that the models can generalise on the navigator data for an initial Euler angle difference of 45 degrees, but due to the difference in sampling density on the utilised point clouds can not provide accurate matches. Therefore, the utilisation of DL models can be considered to be viable if the navigator data has a sampling density similar to the pre-operative model.https://github.com/alpcicimen/holonav-dl-registration The link to the github repository containing the utilised dataset, scripts, as well as the modified DL models RPMNet and PREDATOR.CSE3000 Research ProjectComputer Science and Engineerin
The Scent of a Smell: An Extensive Comparison between Textual and Structural Smells
No full textCode smells are symptoms of poor design or implementation choices that have a negative effect on several aspects of software maintenance and evolution, such as program comprehension or change- and fault-proneness. This is why researchers have spent a lot of effort on devising methods that help developers to automatically detect them in source code. Almost all the techniques presented in literature are based on the analysis of structural properties extracted from source code, although alternative sources of information (e.g., textual analysis) for code smell detection have also been recently investigated. Nevertheless, some studies have indicated that code smells detected by existing tools based on the analysis of structural properties are generally ignored (and thus not refactored) by the developers. In this paper, we aim at understanding whether code smells detected using textual analysis are perceived and refactored by developers in the same or different way than code smells detected through structural analysis. To this aim, we set up two different experiments. We have first carried out a software repository mining study to analyze how developers act on textually or structurally detected code smells. Subsequently, we have conducted a user study with industrial developers and quality experts in order to qualitatively analyze how they perceive code smells identified using the two different sources of information. Results indicate that textually detected code smells are easier to identify and for this reason they are considered easier to refactor with respect to code smells detected using structural properties. On the other hand, the latter are often perceived as more severe, but more difficult to exactly identify and remove.Accepted Author ManuscriptSoftware Engineerin
Fasting blood glucose between 100 and 109 mg/dL versus prediabetes according to glycosylated hemoglobin
No full textIntroducción: algunos estudios han señalado que valores de
glucemia en ayunas entre 100 y 109 mg/dL se asocian con
frecuencias elevadas de prediabetes cuando el criterio de cla-
sificación son los valores de HbA1c. La Sociedad Argentina de
Diabetes (SAD) sostiene a 110 mg/dL como valor a partir del
cual se clasifica a un paciente como portador de glucemia en
ayunas alterada; la frecuencia de individuos posiblemente cla-
sificados en forma incorrecta, según este criterio, aún no se
conoce en la población argentina.
Objetivos: establecer la frecuencia con que se presenta pre-
diabetes según HbA1c en una población sin diagnóstico de
diabetes mellitus (DM) con glucemias en ayunas entre 100 y
109 mg/dL; correlacionar las dos variables y cuantificar la pro-
babilidad de que esto ocurra respecto de otros con glucemias
en ayunas <100 mg/dL.
Materiales y métodos: se incluyeron 1.002 muestras de igual
número de sujetos desde 45 laboratorios de análisis clínicos de la
Asociación de Laboratorios de Alta Complejidad (ALAC), con pro-
cesamiento local de glucemia y centralizado de HbA1c por high
performance liquid chromatography (HPLC). Análisis estadístico:
chi cuadrado, odds ratio, coeficiente de correlación y determina-
ción de Pearson, y correlación serial de Durbin-Watson.
Resultados: frecuencia de HbA1c ≥5,7% en la población estu-
diada con glucemias de ayunas entre 100 y 109 mg/dL=29,7%;
test de chi cuadrado: p<0,001; odds ratio de tener HbA1c ≥5,7%
entre la población con glucemias en ayunas de 100 a 109 mg/dL
vs aquella con valores <100 mg/dL=4,328 (IC 95% 2,922-6,411);
r=0,852, r2
= 0,727, Durbin-Watson=1,152.
Conclusiones: la prediabetes diagnosticada por HbA1c resultó
cuatro veces más frecuente en la población estudiada con glu-
cemias en ayunas entre 100 y 109 mg/dL, que en aquella con
valores por debajo de 100 mg/dL.Introduction: some studies have shown that fasting blood glu-
cose values between 100 and 109 mg/dL are associated with
high rates of prediabetes when the classification criteria are
HbA1c values. The Argentine Diabetes Society still maintains
110 mg/dL as the value from which a patient is classified as
having impaired fasting blood glucose; the frequency of indivi-
duals possibly incorrectly classified, according to this criterion,
is not yet known in any Argentine population.
Objectives: to establish the frequency in a population without
a diagnosis of diabetes mellitus with fasting blood glucose le-
vels between 100 and 109 mg/dL in which prediabetes occurs
according to HbA1c, to correlate both variables and to quantify
the probability that this predicts with respect to others with
fasting blood glucose levels <100 mg/dL.
Materials and methods: 1.002 samples from the same number
of subjects from 45 clinical laboratories belonging to ALAC, with
local processing of blood glucose and centralized processing of
HbA1c by high performance liquid chromatography (HPLC).
Statistical analysis: chi square, odds ratio, Pearson correlation
coefficient, coefficient of determination and Durbin-Watson
serial correlation.
Results: frequency of HbA1c ≥5.7% in the studied population
with fasting blood glucose levels between 100 and 109 mg/
dL = 29.7%, chi square test: p<0.001; odds ratio of having
HbA1c ≥5.7% between the population with fasting blood glu-
cose levels of 100 to 109 mg/dL vs that one with values <100
mg/dL=4.328 (95% CI 2.922-6.411); r=0.852, r2
=0.727, Durbin-
Watson=1.152.
Conclusions: prediabetes diagnosed by HbA1c was four ti-
mes more frequent in the studied population with fasting glu-
cose values between 100 and 109 mg/dL than in that one with
values below 100 mg/dL
Quality interoperability within digital libraries: the DL.org perspective
No full textQuality is the most dynamic aspect of DLs, and becomes even more complex with respect to interoperability. This paper formalizes the research motivations and hypotheses on quality interoperability conducted by the Quality Working Group within the EU-funded project DL.org (<a href="http://www.dlorg.eu">http://www.dlorg.eu/</a>). After providing a multi-level interoperability framework – adopted by DL.org - the authors illustrate key-research points and
approaches on the way to the interoperability of DLs quality, grounding them in the DELOS Reference Model. By applying the DELOS Reference Model Quality Concept Map to their interoperability motivating scenario, the authors subsequently present the two main research outcomes of their investigation - the Quality Core Model and the Quality Interoperability Survey
March dl: Adding Adaptive Heuristics and a New Branching Strategy
No full textWe introduce the march dl satisability (SAT) solver, a successor of march eq. The latter was awarded state-of-the-art in two categories during the Sat 2004 competition. The focus lies on presenting those features that are new in march dl. Besides a description, each of these features is illustrated with some experimental results. By extending the pre-processor, using adaptive heuristics, and by using a new branching strategy, march dl is able to solve nearly all benchmarks faster than its predecessor. Moreover, various instances which were beyond the reach of march eq, can now be solved - relatively easily - due to these new features.Software TechnologyElectrical Engineering, Mathematics and Computer Scienc
Power
No full textOn poster: "An exciting dramatization of the development and use of electric energy in the U.S.A. Directed by Brett Warren. Settings by Howard Bay. Music by Lee Wainer. Entire production under supervision of Morris Watson.
Crash Reproduction Using Helper Objectives
No full textEvolutionary-based crash reproduction techniques aid developers in their debugging practices by generating a test case that reproduces a crash given its stack trace. In these techniques, the search process is typically guided by a single search objective called Crash Distance. Previous studies have shown that current approaches could only reproduce a limited number of crashes due to a lack of diversity in the population during the search. In this study, we address this issue by applying Multi-Objectivization using Helper-Objectives (MO-HO) on crash reproduction. In particular, we add two helper-objectives to the Crash Distance to improve the diversity of the generated test cases and consequently enhance the guidance of the search process. We assessed MO-HO against the single-objective crash reproduction. Our results show that MO-HO can reproduce two additional crashes that were not previously reproducible by the single-objective approach.Virtual/online event due to COVID-19 Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.Software EngineeringSoftware Technolog
SPARQL-DL queries for antipattern detection
No full textOntology antipatterns are structures that reflect ontology modelling problems, they lead to inconsistencies, bad reasoning performance or bad formalisation of domain knowledge. Antipatterns normally appear in ontologies developed by those who are not experts in ontology engineering. Based on our experience in ontology design, we have created a catalogue of such antipatterns in the past, and in this paper we describe how we can use SPARQL-DL to detect them. We conduct some experiments to detect them in a large OWL ontology corpus obtained from the Watson ontology search portal. Our results show that each antipattern needs a specialised detection method
- …
