Poly-bagging predictors for classification modelling for credit scoring

Credit scoring modelling comprises one of the leading formal tools for supporting the granting of credit. Its core objective consists of the generation of a score by means of which potential clients can be listed in the order of the probability of default. A critical factor is whether a credit scoring model is accurate enough in order to provide correct classification of the client as a good or bad payer. In this context the concept of bootstraping aggregating (bagging) arises. The basic idea is to generate multiple classifiers by obtaining the predicted values from the fitted models to several replicated datasets and then combining them into a single predictive classification in order to improve the classification accuracy. In this paper we propose a new bagging-type variant procedure, which we call poly-bagging, consisting of combining predictors over a succession of resamplings. The study is derived by credit scoring modelling. The proposed poly-bagging procedure was applied to some different artificial datasets and to a real granting of credit dataset up to three successions of resamplings. We observed better classification accuracy for the two-bagged and the three-bagged models for all considered setups. These results lead to a strong indication that the poly-bagging approach may promote improvement on the modelling performance measures, while keeping a flexible and straightforward bagging-type structure easy to implement.     

Input-output properties of the Page-Hinkley detector

We consider the stochastic input-output properties of a simple non-linear dynamical system, the so-called Page-Hinkley detector, playing a key role in change detection, and also in queuing theory. We show that for L-mixing inputs with negative expectation the output process of this system is L-mixing. The result is applied to get an upper bound for the false alarm rate. The proof is then adapted to get a similar result for the case of random i.i.d. inputs. Possible extensions and open problems are given in the discussion.     

Competitive exploration of rectilinear polygons

Exploring a polygon with robots when the robots do not have knowledge of the surroundings can be viewed as an online problem. Typical for online problems is that decisions must be made based on past events without complete information about the future. In our case the robots do not have complete information about the environment. Competitive analysis can be used to measure the performance of methods solving online problems. The competitive ratio of such a method is the ratio between the method's performance and the performance of the best method having full knowledge of the future. We prove constant competitive strategies and lower bounds for exploring a simple rectilinear polygon in the _L1$L_1$ metric.     

A platform for the development of patient applications in the domain of personalized health

Personalized health (p-health) systems can contribute significantly to the sustainability of healthcare systems, though their feasibility is yet to be proven. One of the problems related to their development is the lack of well-established development tools for this domain. As the p-health paradigm is focused on patient self-management, big challenges arise around the design and implementation of patient systems. This paper presents a reference platform created for the development of these applications, and shows the advantages of its adoption in a complex project dealing with cardio-vascular diseases.     

On the internal frost resistance of self-compacting concrete, with and without polypropylene fibres

In this article experimental and numerical studies of internal frost resistance of self-compacting and normal concrete, with and without fibres, are outlined. For this purpose self-compacting concrete with low water-cement ratio was studied, with varying amounts of filler, crystalline or sedimentary, different pouring pressures and different mixing procedure with two ages at the start of testing. The concrete was frozen twice a day at ±20^∘C all around the specimen up to 300 cycles. Measurement of length, weight and internal fundamental frequency were performed at the start of testing, at 100 cycles and at 300 frost cycles. Tests were also carried out on submerged cast self-compacting concrete and on self-compacting concrete with fibres. In reference tests normal concrete was studied in parallel. In general self-compacting concrete behaved well or better than normal concrete to internal frost except for the submerged cast concrete, where large segregation occurred and low internal frost resistance followed. The main reasons for low internal frost resistance in submerged applications were probably differences in water-cement ratio in the casting due to cement, water or/and aggregate segregation. Polypropylene fibres in concrete seemed to prohibit the movement of water in the air void system so that a sudden internal collapse occurred before 300 frost cycles.     

The effect of ionising radiation on photosynthetic oxygenic microorganisms for survival in space flight revealed by automatic photosystem II-based biosensors

Photosynthetic microorganisms are expected to be useful to maintain an oxygenic atmosphere and to provide biomass for astronauts in the International Space Station as well as in future long-term space flights. However, fluxes of complex ionizing radiation of various intensities and energies make space an extreme environment for the microorganisms, affecting their photosynthetic efficiency. To automatically monitor the photosynthetic Photosystem II (PSII) activity of microorganisms under space conditions an optical biosensor, which utilizes chlorophyll fluorescence as biological transduction system, was built; the PSII activity was monitored by the biosensor during balloon flights at stratospheric altitudes of about 40 km. The effect of space stress on quantum yield of PSII varied among the tested species depending on the growth light conditions at which they were exposed during the flights.     

Modeling and gradient pattern analysis of irregular SFM structures of porous silicon

Technological applications in opto-electronic devices have increased the interest in characterizing porous silicon structure patterns. Due to its physical properties, solutions from KPZ 2D are adopted to simulate the structure of porous material interface whose spatial characteristics are equivalent to those found in porous silicon samples. The analysis of the simulated and real Scanning Force Microscopy (SFM) surfaces was done using the Gradient Pattern Analysis (GPA). We found that the KPZ 2D model presented asymmetry levels compatible with the irregular surfaces observed by means of SFM images of π-Si.     

Key principles for user-centred systems design

The concept of user-centred systems design (UCSD) has no agreed upon definition. Consequently, there is a great variety in the ways it is applied, which may lead to poor quality and poor usability in the resulting systems, as well as misconceptions about the effectiveness of UCSD. The purpose of this paper is to propose a definition of UCSD. We have identified 12 key principles for the adoption of a user-centred development process, principles that are based on existing theory, as well as research in and experiences from a large number of software development projects. The initial set of principles were applied and evaluated in a case study and modified accordingly. These principles can be used to communicate the nature of UCSD, evaluate a development process or develop systems development processes that support a user-centred approach. We also suggest activity lists and some tools for applying UCSD.