a backtracking search tool for constructing combinatorial test suites

Combinatorial testing is an important testing method. It requires the test cases to cover various combinations of parameters of the system under test. The test generation problem for combinatorial testing can be modeled as constructing a matrix which has certain properties. This paper first discusses two combinatorial testing criteria: covering array and orthogonal array, and then proposes a backtracking search algorithm to construct matrices satisfying them. Several search heuristics and symmetry breaking techniques are used to reduce the search time. This paper also introduces some techniques to generate large covering array instances from smaller ones. All the techniques have been implemented in a tool called EXACT (EXhaustive seArch of Combinatorial Test suites). A new optimal covering array is found by this tool.

Impact of cognitive load and frustration on drivers’ speech [Abstract]

Secondary tasks such as cell phone calls or interaction with automated speech dialog systems (SDSs) increase the driver’s cognitive load as well as the probability of driving errors. This study analyzes speech production variations due to cognitive load and emotional state of drivers in real driving conditions. Speech samples were acquired from 24 female and 17 male subjects (approximately 8.5 h of data) while talking to a co-driver and communicating with two automated call centers, with emotional states (neutral, negative) and the number of necessary SDS query repetitions also labeled. A consistent shift in a number of speech production parameters (pitch, first format center frequency, spectral center of gravity, spectral energy spread, and duration of voiced segments) was observed when comparing SDS interaction against co-driver interaction; further increases were observed when considering negative emotion segments and the number of requested SDS query repetitions. A mel frequency cepstral coefficient based Gaussian mixture classifier trained on 10 male and 10 female sessions provided 91% accuracy in the open test set task of distinguishing co-driver interactions from SDS interactions, suggesting—together with the acoustic analysis—that it is possible to monitor the level of driver distraction directly from their speech.

Ratchetting of railhead in the vicinity of the gap of the insulated rail joints

Insulated Rail Joints (IRJs) are designed to electrically isolate two rails in rail tracks to control the signalling system for safer train operations. Unfortunately the gapped section of the IRJs is structurally weak and often fails prematurely especially in heavy haul tracks, which adversely affects service reliability and efficiency. The IRJs suffer from a number of failure modes; the railhead ratchetting at the gap is, however, regarded as the root cause and attended to in this thesis. Ratchetting increases with the increase in wheel loads; in the absence of a life prediction model, effective management of the IRJs for increased wagon wheel loads has become very challenging. Therefore, the main aim of this thesis is to determine method to predict IRJs' service life. The distinct discontinuity of the railhead at the gap makes the Hertzian theory and the rolling contact shakedown map, commonly used in the continuously welded rails, not applicable to examine the metal ratchetting of the IRJs. Finite Element (FE) technique is, therefore, used to explore the railhead metal ratchetting characteristics in this thesis, the boundary conditions of which has been determined from a full scale study of the IRJ specimens under rolling contact of the loaded wheels. A special purpose test set up containing full-scale wagon wheel was used to apply rolling wheel loads on the railhead edges of the test specimens. The state of the rail end face strains was determined using a non-contact digital imaging technique and used for calibrating the FE model. The basic material parameters for this FE model were obtained through independent uniaxial, monotonic tensile tests on specimens cut from the head hardened virgin rails. The monotonic tensile test data have been used to establish a cyclic load simulation model of the railhead steel specimen; the simulated cyclic load test has provided the necessary data for the three decomposed kinematic hardening plastic strain accumulation model of Chaboche. A performance based service life prediction algorithm for the IRJs was established using the plastic strain accumulation obtained from the Chaboche model. The predicted service lives of IRJs using this algorithm have agreed well with the published data. The finite element model has been used to carry out a sensitivity study on the effects of wheel diameter to the railhead metal plasticity. This study revealed that the depth of the plastic zone at the railhead edges is independent of the wheel diameter; however, large wheel diameter is shown to increase the IRJs' service life.

Transforming morning to afternoon using linear regression techniques

Visual localization in outdoor environments is often hampered by the natural variation in appearance caused by such things as weather phenomena, diurnal fluctuations in lighting, and seasonal changes. Such changes are global across an environment and, in the case of global light changes and seasonal variation, the change in appearance occurs in a regular, cyclic manner. Visual localization could be greatly improved if it were possible to predict the appearance of a particular location at a particular time, based on the appearance of the location in the past and knowledge of the nature of appearance change over time. In this paper, we investigate whether global appearance changes in an environment can be learned sufficiently to improve visual localization performance. We use time of day as a test case, and generate transformations between morning and afternoon using sample images from a training set. We demonstrate the learned transformation can be generalized from training data and show the resulting visual localization on a test set is improved relative to raw image comparison. The improvement in localization remains when the area is revisited several weeks later.

Overview of the ShARe/CLEF eHealth Evaluation Lab 2013

Discharge summaries and other free-text reports in healthcare transfer information between working shifts and geographic locations. Patients are likely to have difficulties in understanding their content, because of their medical jargon, non-standard abbreviations,and ward-specific idioms. This paper reports on an evaluation lab with an aim to support the continuum of care by developing methods and resources that make clinical reports in English easier to understand for patients, and which helps them in finding information related to their condition.

Generating natural language texts from business process models

Process Modeling is a widely used concept for understanding, documenting and also redesigning the operations of organizations. The validation and usage of process models is however affected by the fact that only business analysts fully understand them in detail. This is in particular a problem because they are typically not domain experts. In this paper, we investigate in how far the concept of verbalization can be adapted from object-role modeling to process models. To this end, we define an approach which automatically transforms BPMN process models into natural language texts and combines different techniques from linguistics and graph decomposition in a flexible and accurate manner. The evaluation of the technique is based on a prototypical implementation and involves a test set of 53 BPMN process models showing that natural language texts can be generated in a reliable fashion.

Experimental study of load-bearing cold-formed steel walls exposed to realistic design fires

This paper presents the details of full scale fire tests of LSF wall panels conducted using realistic fire time-temperature curves. Tests included eight LSF wall specimens of various configurations exposed to both parametric design and natural fire curves. Details of the fire test set-up, test procedure and the results including the measured time-temperature and deformation curves of LSF wall panels are presented along with wall stud failure modes and times. This paper also compares the structural and thermal behavioural characteristics of LSF wall studs with those based on the standard time-temperature curve. Finally, the stud failure times and temperatures are summarized for both standard and realistic design fire curves. This study provides the necessary test data to validate the numerical models of LSF wall panels and to undertake a detailed study into the structural and thermal performance of LSF wall panels exposed to realistic design fire curves.

Fine-grained plant classification using convolutional neural networks for feature extraction

We present an overview of the QUT plant classification system submitted to LifeCLEF 2014. This system uses generic features extracted from a convolutional neural network previously used to perform general object classification. We examine the effectiveness of these features to perform plant classification when used in combination with an extremely randomised forest. Using this system, with minimal tuning, we obtained relatively good results with a score of 0:249 on the test set of LifeCLEF 2014.

Investigation of structural requirements of anticancer activity at the paclitaxel/tubulin binding site using CoMFA and CoMSIA

CoMFA and CoMSIA analysis were utilized in this investigation to define the important interacting regions in paclitaxel/tubulin binding site and to develop selective paclitaxel-like active compounds. The starting geometry of paclitaxel analogs was taken from the crystal structure of docetaxel. A total of 28 derivatives of paclitaxel were divided into two groups—a training set comprising of 19 compounds and a test set comprising of nine compounds. They were constructed and geometrically optimized using SYBYL v6.6. CoMFA studies provided a good predictability (q2 = 0.699, r2 = 0.991, PC = 6, S.E.E. = 0.343 and F = 185.910). They showed the steric and electrostatic properties as the major interacting forces whilst the lipophilic property contribution was a minor factor for recognition forces of the binding site. These results were in agreement with the experimental data of the binding activities of these compounds. Five fields in CoMSIA analysis (steric, electrostatic, hydrophobic, hydrogen-bond acceptor and donor properties) were considered contributors in the ligand–receptor interactions. The results obtained from the CoMSIA studies were: q2 = 0.535, r2 = 0.983, PC = 5, S.E.E. = 0.452 and F = 127.884. The data obtained from both CoMFA and CoMSIA studies were interpreted with respect to the paclitaxel/tubulin binding site. This intuitively suggested where the most significant anchoring points for binding affinity are located. This information could be used for the development of new compounds having paclitaxel-like activity with new chemical entities to overcome the existing pharmaceutical barriers and the economical problem associated with the synthesis of the paclitaxel analogs. These will boost the wide use of this useful class of compounds, i.e. in brain tumors as the most of the present active compounds have poor blood–brain barrier crossing ratios and also, various tubulin isotypes has shown resistance to taxanes and other antimitotic agents.

Automatic ICD-10 classification of cancers from free-text death certificates

Objective Death certificates provide an invaluable source for cancer mortality statistics; however, this value can only be realised if accurate, quantitative data can be extracted from certificates – an aim hampered by both the volume and variable nature of certificates written in natural language. This paper proposes an automatic classification system for identifying cancer related causes of death from death certificates. Methods Detailed features, including terms, n-grams and SNOMED CT concepts were extracted from a collection of 447,336 death certificates. These features were used to train Support Vector Machine classifiers (one classifier for each cancer type). The classifiers were deployed in a cascaded architecture: the first level identified the presence of cancer (i.e., binary cancer/nocancer) and the second level identified the type of cancer (according to the ICD-10 classification system). A held-out test set was used to evaluate the effectiveness of the classifiers according to precision, recall and F-measure. In addition, detailed feature analysis was performed to reveal the characteristics of a successful cancer classification model. Results The system was highly effective at identifying cancer as the underlying cause of death (F-measure 0.94). The system was also effective at determining the type of cancer for common cancers (F-measure 0.7). Rare cancers, for which there was little training data, were difficult to classify accurately (F-measure 0.12). Factors influencing performance were the amount of training data and certain ambiguous cancers (e.g., those in the stomach region). The feature analysis revealed a combination of features were important for cancer type classification, with SNOMED CT concept and oncology specific morphology features proving the most valuable. Conclusion The system proposed in this study provides automatic identification and characterisation of cancers from large collections of free-text death certificates. This allows organisations such as Cancer Registries to monitor and report on cancer mortality in a timely and accurate manner. In addition, the methods and findings are generally applicable beyond cancer classification and to other sources of medical text besides death certificates.

11β-Hydroxysteroid dehydrogenase inhibitors and selectivity modeling of 11β-HSDs

11β-hydroksisteroididehydrogenaasientsyymit (11β-HSD) 1 ja 2 säätelevät kortisonin ja kortisolin määrää kudoksissa. 11β-HSD1 -entsyymin ylimäärä erityisesti viskeraalisessa rasvakudoksessa aiheuttaa metaboliseen oireyhtymän klassisia oireita, mikä tarjoaa mahdollisuuden metabolisen oireyhtymän hoitoon 11β-HSD1 -entsyymin selektiivisellä estämisellä. 11β-HSD2 -entsyymin inhibitio aiheuttaa kortisonivälitteisen mineralokortikoidireseptorien aktivoitumisen, mikä puolestaan johtaa hypertensiivisiin haittavaikutuksiin. Haittavaikutuksista huolimatta 11β-HSD2 -entsyymin estäminen saattaa olla hyödyllistä tilanteissa, joissa halutaan nostaa kortisolin määrä elimistössä. Lukuisia selektiivisiä 11β-HSD1 inhibiittoreita on kehitetty, mutta 11β-HSD2-inhibiittoreita on raportoitu vähemmän. Ero näiden kahden isotsyymin aktiivisen kohdan välillä on myös tuntematon, mikä vaikeuttaa selektiivisten inhibiittoreiden kehittämistä kummallekin entsyymille. Tällä työllä oli kaksi tarkoitusta: (1) löytää ero 11β-HSD entsyymien välillä ja (2) kehittää farmakoforimalli, jota voitaisiin käyttää selektiivisten 11β-HSD2 -inhibiittoreiden virtuaaliseulontaan. Ongelmaa lähestyttiin tietokoneavusteisesti: homologimallinnuksella, pienmolekyylien telakoinnilla proteiiniin, ligandipohjaisella farmakoforimallinnuksella ja virtuaaliseulonnalla. Homologimallinnukseen käytettiin SwissModeler -ohjelmaa, ja luotu malli oli hyvin päällekäinaseteltavissa niin templaattinsa (17β-HSD1) kuin 11β-HSD1 -entsyymin kanssa. Eroa entsyymien välillä ei löytynyt tarkastelemalla päällekäinaseteltuja entsyymejä. Seitsemän yhdistettä, joista kuusi on 11β-HSD2 -selektiivisiä, telakoitiin molempiin entsyymeihin käyttäen ohjelmaa GOLD. 11β-HSD1 -entsyymiin yhdisteet kiinnittyivät kuten suurin osa 11β-HSD1 -selektiivisistä tai epäselektiivisistä inhibiittoreista, kun taas 11β-HSD2 -entsyymiin kaikki yhdisteet olivat telakoituneet käänteisesti. Tällainen sitoutumistapa mahdollistaa vetysidokset Ser310:een ja Asn171:een, aminohappoihin, jotka olivat nähtävissä vain 11β-HSD2 -entsyymissä. Farmakoforimallinnukseen käytettiin ohjelmaa LigandScout3.0, jolla ajettiin myös virtuaaliseulonnat. Luodut kaksi farmakoforimallia, jotka perustuivat aiemmin telakointiinkin käytettyihin kuuteen 11β-HSD2 -selektiiviseen yhdisteeseen, koostuivat kuudesta ominaisuudesta (vetysidosakseptori, vetysidosdonori ja hydrofobinen), ja kieltoalueista. 11β-HSD2 -selektiivisyyden kannalta tärkeimmät ominaisuudet ovat vetysidosakseptori, joka voi muodostaa sidoksen Ser310 kanssa ja vetysidosdonori sen vieressä. Tälle vetysidosdonorille ei löytynyt vuorovaikutusparia 11β-HSD2-mallista. Sopivasti proteiiniin orientoitunut vesimolekyyli voisi kuitenkin olla sopiva ratkaisu puuttuvalle vuorovaikutusparille. Koska molemmat farmakoforimallit löysivät 11β-HSD2 -selektiivisiä yhdisteitä ja jättivät epäselektiivisiä pois testiseulonnassa, käytettiin molempia malleja Innsbruckin yliopistossa säilytettävistä yhdisteistä (2700 kappaletta) koostetun tietokannan seulontaan. Molemmista seulonnoista löytyneistä hiteistä valittiin yhteensä kymmenen kappaletta, jotka lähetettiin biologisiin testeihin. Biologisien testien tulokset vahvistavat lopullisesti sen kuinka hyvin luodut mallit edustavat todellisuudessa 11β-HSD2 -selektiivisyyttä.

Flexural bond strength of masonry using various blocks and mortars

This paper deals with an experimental study on flexural bond strength of masonry using various blocks in combination with different mortars. Flexural bond strength of masonry has been determined by testing stack-bonded prisms using a modified bond wrench test set-up. The effect of mortar composition and strength on the masonry's. flexural bond strength using three types of masonry units (stabilized mud blocks, stabilized soil-sand blocks and burnt brick) has been examined. The effect of the masonry unit's moisture content on flexural bond strength has also been studied. Increases in mortar strength lead to increased flexural bond strength for cement mortar, irrespective of the type of masonry unit. It has been found that combination mortars, such as soil-cement mortar and cement-lime mortar, lead to better bond strength compared to cement mortars. The moisture content of the masonry unit at the time of casting has displayed significant influence on the flexural bond strength of the masonry. It has been found that for each type of masonry unit, an optimum moisture content exists, beyond which the flexural bond strength falls off quickly.

Semi-supervised SVMs for classification with unknown class proportions and a small labeled dataset

In the design of practical web page classification systems one often encounters a situation in which the labeled training set is created by choosing some examples from each class; but, the class proportions in this set are not the same as those in the test distribution to which the classifier will be actually applied. The problem is made worse when the amount of training data is also small. In this paper we explore and adapt binary SVM methods that make use of unlabeled data from the test distribution, viz., Transductive SVMs (TSVMs) and expectation regularization/constraint (ER/EC) methods to deal with this situation. We empirically show that when the labeled training data is small, TSVM designed using the class ratio tuned by minimizing the loss on the labeled set yields the best performance; its performance is good even when the deviation between the class ratios of the labeled training set and the test set is quite large. When the labeled training data is sufficiently large, an unsupervised Gaussian mixture model can be used to get a very good estimate of the class ratio in the test set; also, when this estimate is used, both TSVM and EC/ER give their best possible performance, with TSVM coming out superior. The ideas in the paper can be easily extended to multi-class SVMs and MaxEnt models.

Minimization of grid current distortion in parallel-connected converters through carrier interleaving

Identical parallel-connected converters with unequal load sharing have unequal terminal voltages. The difference in terminal voltages is more pronounced in case of back-to-back connected converters, operated in power-circulation mode for the purpose of endurance tests. In this paper, a synchronous reference frame based analysis is presented to estimate the grid current distortion in interleaved, grid-connected converters with unequal terminal voltages. Influence of carrier interleaving angle on rms grid current ripple is studied theoretically as well as experimentally. Optimum interleaving angle to minimize the rms grid current ripple is investigated for different applications of parallel converters. The applications include unity power factor rectifiers, inverters for renewable energy sources, reactive power compensators, and circulating-power test set-up used for thermal testing of high-power converters. Optimum interleaving angle is shown to be a strong function of the average of the modulation indices of the two converters, irrespective of the application. The findings are verified experimentally on two parallel-connected converters, circulating reactive power of up to 150 kVA between them.

A STUDY INTO THE BEHAVIOR OF AN ALUMINUM FOAM UNDER COMPRESSION

The characterization of a closed-cell aluminum foam with the trade name Alporas is carried out here under compression loading for a nominal cross-head speed of 1 mm/min. Foam samples in the form of cubes are tested in a UTM and the average stress-strain behavior is obtained which clearly displays a plateau strength of approximately 2 MPa. It is noted that the specific energy absorption capacity of the foam can be high despite its low strength which makes it attractive as a material for certain energy-absorbing countermeasures. The mechanical behavior of the present Alporas foam is simulated using cellular (i.e. so-called microstructure-based) and solid element-based finite element models. The efficacy of the cellular approach is shown, perhaps for the first time in published literature, in terms of prediction of both stress-strain response and inclined fold formation during axial crush under compression loading. Keeping in mind future applications under impact loads, limited results are presented when foam samples are subjected to low velocity impact in a drop-weight test set-up.