Abstract A novel probabilistic framework for modelling anomalous diffusion is
presented. The resulting process is Markovian, non-homogeneous, non-stationary, non-
ergodic, and state-dependent. fundamental law governing this driven by
two opposing forces: one proportional to the current state, representing intensity
of autocorrelation or contagion, another inversely elapsed
time, acting as a damping function. interplay between these forces...

Feature selection is a highly relevant task in any data-driven knowledge discovery project. The present research focuses on analysing the advantages and disadvantages of using mutual information (MI) data-based sensitivity analysis (DSA) for feature classification problems, by applying both to bank telemarketing case. A logistic regression model built tuned set features identified each two techniques as most influencing success contact, total 13 MI 9 DSA. latter performs better lower values...

A new software reliability model based on the empirical Bayes estimate is developed. The number of failures estimated up to a given time used in order probability failure appearance during next interval. Instead non homogeneous rate as it usually growth, depending linearly previous obtained from our model. mixed Poisson where mixing density function models growth. can be either simulate cumulative curve or between failures. Data similar project parameters project. Results simulations and...

A new software reliability model based on the Polya contagion stochastic process is proposed. We failure detection as a pure birth with rate that depends not just time but number of failures previously detected, it happens in obtained asymptotic limit urn for contagion. Since proposes an increasing (birth) rate, this suitable to be applied at beginning testing or when some code has been added project, situations where S-shaped usually used. The result applying our real data also shown.

Breast cancer in men is a rare disease, accounting for around 1% of all cancers males. Diagnosis often occurs advanced stages due to low awareness this disease To evaluate the clinical, epidemiological, and histopathological characteristics breast Aconcagua Valley its relationship with patient survival. A descriptive study was conducted at University Valparaíso, analyzing data from seven male patients diagnosed San Camilo Hospital Felipe between 2013 2018. Epidemiological, were analyzed...

A data mining of several Bugzilla datasets using Software Reliability models is presented. We analyzed reports from the Xfce, Firefox, Eclipse and Tomcat projects for a long period time thousand days. In all cases, an increasing failure rate have been found. Increasing rates are usually modeled by S-shaped model in literature. propose to use some where depends not just on time, but also number failures previously reported. These little modifications Yule Polya stochastic processes. The...

A new Software Reliability model based on a pure birth process is proposed. In our novel approach, the (failure) rate of considered to be non dependent time but linearly previous number births (failures), contrarily homogeneous processes, as it usually done in literature. We use empirical Bayes framework order get rate. Our approach allows either estimate mean failure MTTF or simulate stochastic process. analyze both, discrete and continuous case apply first real case.

Abstract Objectives To introduce a novel way of measuring the spreading speed an epidemic. Methods We propose to use mean time between infections (MTBI) metric obtained from recently introduced nonhomogeneous Markov stochastic model. Different types parameter calibration are performed. estimate MTBI using data different windows and whole stage history compare results. In order detect waves stages in input data, preprocessing filtering technique is applied. Results The results applying this...

In this paper, the chains of rare events model and its applications are analyzed. This was originally introduced in order to analyze which can be produced as simple, double, triple, etc. Every one is distributed according a Poisson law. A simple relation between parameters represent contagion phenomenon. particularly good grouped like accidents, telephone calls, death, birth, failures production, reliability, The best known used compound Poisson. we show generalization reliability queuing...

A new Software Reliability Model based on a Mixed Poisson process where the failure rate follows an Inverse Gaussian distribution is proposed. By using Empirical Bayes estimate of rate, our depends just number failures and total past time, it not necessary to know exact instants time were have occurred. We simulate detection with exponential waiting times randomly generated constantly updated estimate. Results simulations are compared two real projects.

The empirical Bayes estimator of the probability a successful event is deduced from mixed distributions. Specially, binomial and Poisson distributions are analyzed. Some mixing distributions, well known in Reliability, Queuing Theory other areas Engineering, considered. As it will be shown, family estimators with interesting characteristics obtained for different This includes those literature. another result, non linear behavior as function sample obtained.

A Multistage Software Reliability analysis is performed. We propose to use an increasing failure rate model at the very first stage of development and testing a Growth last before release, middle stages usually presents jumps generally due adding new code. remark then importance applied in order predict when reliability growth starts. An excessive fast increase could alarm Engineers as adjust processes. Two real projects with similar metrics are analyzed. found that logistic fits best stage,...

In order to study the properties of short-range in simple liquids, a ‘‘moving average filter’’ is applied positions and velocities obtained from molecular dynamics simulations L-J systems, reduce thermal motions. The fictitious systems that are allow better analysis, not only underlying structure, but also time-dependent low-frequency properties, such as diffusive From analysis correlation radial function g(r) recently proposed [S. Mazur, J. Chem. Phys. 97, 9276 (1992)] ‘‘neighborship...

Network traffic modeled as queues where the probability of a new entry depends on history arrival process is analyzed. Based this characteristic, we propose to use generalized Polya stochastic processes (GPPs). The theoretical background such type queue reviewed and heavy-tailed epochs property shown hold for rates consider. instability or congestion GPP/M/1 GPP/GPP/1 analyzed from simulations. results agree with process, show interesting similarities differences when recently proposed rate...

Abstract Computer simulation results for the angular dependence of spatial correlation functions in liquids are given. The were obtained using smoothing filters applied to molecular trajectories liquids. purpose this procedure is enhance local structure by attenuating thermal vibrations, and study low frequency motions. New positions velocities measured. These depend not only on distance but also angle between vector velocity reference particle. suggest that cooperative self-diffusion mainly...