Yet another induction algorithm

Inducing general functions from specific training examples is a central problem in the machine learning. Using sets of If-then rules is the most expressive and readable manner. To find If-then rules, many induction algorithms such as ID3, AQ, CN2 and their variants, were proposed. Sequential covering is the kernel technique of them. To avoid testing all possible selectors, Entropy gain is used to select the best attribute in ID3. Constraint of the size of star was introduced in AQ and beam search was adopted in CN2. These methods speed up their induction algorithms but many good selectors are filtered out. In this work, we introduce a new induction algorithm that is based on enumeration of all possible selectors. Contrary to the previous works, we use pruning power to reduce irrelative selectors. But we can guarantee that no good selectors are filtered out. Comparing with other techniques, the experiment results demonstrate
that the rules produced by our induction algorithm have high consistency and simplicity.

An examination on the performance of MML causal induction

This paper presents an examination report on the performance of the improved MML based causal model discovery algorithm. In this paper, We firstly describe our improvement to the causal discovery algorithm which introduces a new encoding scheme for measuring the cost of describing the causal structure. Stiring function is also applied to further simplify the computational complexity and thus works more efficiently. It is followed by a detailed examination report on the performance of our improved discovery algorithm. The experimental results of the current version of the discovery system show that: (l) the current version is capable of discovering what discovered by previous system; (2) current system is capable of discovering more complicated causal networks with large number of variables; (3) the new version works more efficiently compared with the previous version in terms of time complexity.

Conductive fabrics for heating applications

By coating textiles with electrically conductive organic polymers, we are able to produce functional, intelligent fabrics. These fabrics can be utilised in applications such as gas sensors, actuators, electromagnetic shielding, radar absorption, selected frequency filtering in indoor wireless applications, and heating applications where vital parts of the body can be heated without embedding any wiring through the fabric.

Heat generation in fabrics coated with the conductive polymer polypyrrole was investigated. The fabrics were coated by chemical synthesis methods by oxidizing the pyrrole monomer in the presence of the fabric substrate. Ferric chloride was selected as the oxidizing agent and anthraquinone-2-sulfonic acid (AQSA) sodium salt monohydrate as the dopant.

Conductive fabrics were characterized by resistivity measurements, scanning electron microscopy, thermal imaging, current transmission over a period of time and calculations of power density per unit area. Effects of reaction conditions on the electrical properties and heat generated are presented. Polypyrrole coated fabrics were stable and possessed high electrical conductivity. Resistivity values ranged from 100-500 ohms/square depending on the reaction parameters. When subjected to a constant voltage of 24V, the polypyrrole coated polyester-Lycra® fabric doped with AQSA reached a maximum temperature of 42°C and a power density per unit area of 430 W/m² was achieved.

Computational simulation of heat transfer from alternate heating sources

The alternant heat transfer induced by particle packet and gas bubbles on an object surface in a gas fluidised bed is computationally studied. The particle packet and bubble are modelled by a DPPM (double particle-layer and Porous Medium) model and a hemispherical model, respectively. Different meshing schemes are applied and different mesh sizes are used in meshing particle packet and heated object and a very large geometrical size difference between them was considered. Two parallel solver processes were proposed to perform the simulation of heat transfer for different purposes and implemented with the Fluent CFD package.

Induction of epithelial to mesenchymal transition in PMC42-LA human breast carcinoma cells by carcinoma-associated fibroblast secreted factors

Background
Breast carcinoma is accompanied by changes in the acellular and cellular components of the microenvironment, the latter typified by a switch from fibroblasts to myofibroblasts.


Methods
We utilised conditioned media cultures, Western blot analysis and immunocytochemistry to investigate the differential effects of normal mammary fibroblasts (NMFs) and mammary cancer-associated fibroblasts (CAFs) on the phenotype and behaviour of PMC42-LA breast cancer cells. NMFs were obtained from a mammary gland at reduction mammoplasty, and CAFs from a mammary carcinoma after resection.


Results
We found greater expression of myofibroblastic markers in CAFs than in NMFs. Medium from both CAFs and NMFs induced novel expression of α-smooth muscle actin and cytokeratin-14 in PMC42-LA organoids. However, although conditioned media from NMFs resulted in distribution of vimentin-positive cells to the periphery of PMC42-LA organoids, this was not seen with CAF-conditioned medium. Upregulation of vimentin was accompanied by a mis-localization of E-cadherin, suggesting a loss of adhesive function. This was confirmed by visualizing the change in active β-catenin, localized to the cell junctions in control cells/cells in NMF-conditioned medium, to inactive β-catenin, localized to nuclei and cytoplasm in cells in CAF-conditioned medium.


Conclusion
We found no significant difference between the influences of NMFs and CAFs on PMC42-LA cell proliferation, viability, or apoptosis; significantly, we demonstrated a role for CAFs, but not for NMFs, in increasing the migratory ability of PMC42-LA cells. By concentrating NMF-conditioned media, we demonstrated the presence of factor(s) that induce epithelial-mesenchymal transition in NMF-conditioned media that are present at higher levels in CAF-conditioned media. Our in vitro results are consistent with observations in vivo showing that alterations in stroma influence the phenotype and behaviour of surrounding cells and provide evidence for a role for CAFs in stimulating cancer progression via an epithelial-mesenchymal transition. These findings have implications for our understanding of the roles of signalling between epithelial and stromal cells in the development and progression of mammary carcinoma.

Solar heating systems for recirculation aquaculture

The literature over the past 25 years indicates that there has been a continued interest in using passive and active solar technologies to reduce the conventional energy required to maintain water temperatures in small recirculation aquaculture systems. Although all of the experimental systems reviewed report favourable results, there is little information available to guide system designers. This paper describes the use of a simulation model to predict the annual conventional energy consumption of a 10.6 m3 RAS enclosed in a double layer polyethylene greenhouse in two different climates. The water was maintained at 22.5 °C and the recirculation rate was 10% of tank volume per day. Simple unglazed solar collectors have also been combined with the greenhouse to further reduce energy consumption. The effect of increasing collector area on the solar fraction and utilization of useful energy was predicted. Finally, the model was used to investigate the relationship between the occurrence of condensation on the inner cover, ventilation rates and energy use. It was found that in a hot dry climate, the greenhouse alone was sufficient to reduce the conventional energy requirements by 87%; while in the cooler temperate climate reductions of 66% were possible. When solar collectors were added to the system, conventional energy requirements were reduced further and depended on the area of collector used. For example, in the temperate climate location, conventional energy requirements were reduced to 23% of a RAS enclosed in a non-solar building when 26 m2 of solar collector inclined at the optimum angle for winter energy collection were used. Although condensation could be successfully reduced by ventilation of the greenhouse, this increased conventional energy requirements because the potential for evaporation was increased. Covering the tanks at night was found to be a more effective strategy because it reduced condensation and conventional energy use simultaneously.

The effect of cure cycle heating rate on the fibre/matrix interface

Development of civil aerospace composites is key to future “greener” aircraft. Aircraft manufacturers must improve efficiency of their product and manufacturing processes to remain viable. The aerospace industry is undergoing a materials revolution in the design and manufacture of composite airframes. The Airbus A350 and Boeing 787 (both due to enter service in the latter part of this decade) will push utilisation levels of  composite materials beyond 50% of the total airframe by weight. This  change requires massive investment in materials technology, manufacturing capability and skills development. The Quickstep process provides the ability to rapidly cure aerospace standard composite materials whilst providing enhanced mechanical properties. Utilising fluid to transfer heat to the   composite component during the curing process allows far higher heat rates than with conventional cure techniques. The rapid heat-up rates reduce the viscosity of the resin system greatly to provide a longer processing window introducing greater flexibility and removing the need for high pressure during cure. Interlaminar fracture toughness (Mode I) and Interfacial Shear Strength of aerospace standard materials cured using Quickstep have been    compared to autoclave cured laminates. Results suggest an improvement in fibre-matrix adhesion.

Heating effects on water repellency in Australian eucalypt forest soils and their value in estimating wildfire soil temperatures

Wildfires can induce or enhance soil water repellency under a range of vegetation communities. According to mainly USA-based laboratory studies, repellency is eliminated at a maximum soil temperature (T) of 280–400°C. Knowledge of T reached during a wildfire is important in evaluating post-fire soil physical properties, fertility and seedbed status. T is, however, notoriously difficult to ascertain retrospectively and often based on indicative observations with a large potential error. Soils under fire-prone Australian eucalypt forests tend to be water repellent when dry or moderately moist even if long unburnt. This study aims to quantify the temperature of water repellency destruction for Australian topsoil material sampled under three sites with contrasting eucalypt cover (Eucalyptus sieberi, E. ovata and E. baxteri). Soil water repellency was present prior to heating in all samples, increased during heating, but was abruptly eliminated at a specific T between 260 and 340°C. Elimination temperature varied somewhat between samples, but was found to be dependent on heating duration, with longest duration resulting in lowest elimination temperature. Results suggest that post-fire water repellency may be used as an aid in hindcasting soil temperature reached during the passage of a fire within repellency-prone environments.

Early induction of chalpains in rotenone-mediated neuronal apoptosis

Rotenone is an inhibitor of mitochondrial complex I that produces a model of Parkinson's disease (PD), where neurons undergo apoptosis by caspase-dependent and/or caspase-independent pathways. Inhibition of calpains has recently been shown to attenuate neuronal apoptosis. This study aims to establish for the first time, the time-point of calpain activation with respect to the caspase activation and the possibility of cell cycle re-entry in rotenone-mediated cell death. Immunoblot results revealed calpain activation occurred at 5, 10 h prior to caspase-3 activation (at 15 h), suggesting calpain activation was an earlier cellular event compared to caspase activation in the rotenone-mediated apoptosis. In addition, an upregulation of phospho-p53 was observed at 21 h. However, no expression or upregulation of cell cycle regulatory proteins including cdc25a, cyclin-D1 and cyclin-D3 were observed, strongly suggesting that cell cycle re-entry did not occur. These findings provide new insights into the differential patterns of calpain and caspase activation that result from rotenone poisoning and which may be relevant to the therapeutic management of PD.

Induction of propranolol metabolism by Ginkgo biloba extract EGb 761 in rats

Ginkgo biloba is one of the most popular herbal medicines in the world, due to its purported pharmacological effects, including memory-enhancing, cognition-improving, and antiplatelet effects. When used in the elderly, Ginkgo has a high potential for interactions with cardiovascular drugs. This study aimed to investigate the effects of the standard Ginkgo biloba extract (EGB 761) treatment on the pharmacokinetics of propranolol and its metabolism to form Ndesisopropylpropranolol (NDP) in rats. We also examined the activity and expression of cytochrome P450 (CYP) 1A and other CYPs in rats treated with EGb 761 at 10 and 100 mg/kg/day for 10 days. A single oral dose of propranolol (10 mg/kg) was administered on day 11 and the concentrations of both propranolol and NDP were determined using validated liquid chromatography-mass spectrometry (LC-MS) methods. The levels of mRNA and protein of various CYPs were determined by RT-PCR and Western blotting analysis, respectively. Pretreatment of EGb 761 at 100 mg/kg, but not 10 mg/kg, for 10 days significantly reduced the area under the plasma concentration-time curve (AUC) and maximum plasma concentration (C max) of propranolol, whereas those values of NDP were significantly increased. CYP1A1, 1A2, 2B1/2, and 3A1 activities and gene expression in the rat liver were significantly increased in a dose-dependent manner by pretreatment with EGb 761. The ex-vivo formation of NDP in liver microsomes from rats pretreated with EGb 761 was markedly enhanced. The formation of NDP from propranolol in liver microsomes was significantly inhibited by α- naphthoflavone (ANF, a selective CYP1A2 inhibitor), but not by quinidine (a CYP2D inhibitor). These results indicated that EGb 761 pretreatment decreased the plasma concentrations of propranolol by accelerated conversion of parental drug to NDP due to induction of CYP1A2. EGb 761 pretreatment also significantly induced CYP2B1/2 and CYP3A1, suggesting potential interactions with substrate drugs for these two enzymes. Further study is needed to explore the potential for gingko-drug interactions and the clinical impact.

Short-term heating tests on doped polypyrrole-coated polyester fabrics

Polyethyleneterephtalate–polypyrrole (PET–PPy) textile complexes incorporating different anionic dopants have been heat treated at 60 °C, 80 °C, 105 °C, 125 °C and 150 °C to investigate effects of short-term heating on conductivity and stability. In most cases heat treatments below 80 °C did not significantly change the final resistance of the conducting textiles. Only the anionic dye-dopant Indigo Carmine acted in a heat-stabilizing manner during treatment at 150 °C, while all samples containing other anionic dopant underwent some degradation. A treatment temperature of 125 °C was the most effective for lowering the final resistance, with sulphonic group containing dopants being particularly effective in improving conductivity and stability. A 29% decrease in the final resistance of a PET–PPy/para-toluene-2-sulphonic acid (pTSA) sample was achievable after 900 s at 125 °C.

Convex hulls in concept induction

Classification learning is dominated by systems which induce large numbers of small axis-orthogonal decision surfaces. This strongly biases such systems towards particular hypothesis types but there is reason believe that many domains have underlying concepts which do not involve axis orthogonal surfaces. Further, the multiplicity of small decision regions mitigates against any holistic appreciation of the theories produced by these systems, notwithstanding the fact that many of the small regions are individually comprehensible. This thesis investigates modeling concepts as large geometric structures in n-dimensional space. Convex hulls are a superset of the set of axis orthogonal hyperrectangles into which axis orthogonal systems partition the instance space. In consequence, there is reason to believe that convex hulls might provide a more flexible and general learning bias than axis orthogonal regions. The formation of convex hulls around a group of points of the same class is shown to be a usable generalisation and is more general than generalisations produced by axis-orthogonal based classifiers, without constructive induction, like decision trees, decision lists and rules. The use of a small number of large hulls as a concept representation is shown to provide classification performance which can be better than that of classifiers which use a large number of small fragmentary regions for each concept. A convex hull based classifier, CH1, has been implemented and tested. CH1 can handle categorical and continuous data. Algorithms for two basic generalisation operations on hulls, inflation and facet deletion, are presented. The two operations are shown to improve the accuracy of the classifier and provide moderate classification accuracy over a representative selection of typical, largely or wholly continuous valued machine learning tasks. The classifier exhibits superior performance to well-known axis-orthogonal-based classifiers when presented with domains where the underlying decision surfaces are not axis parallel. The strengths and weaknesses of the system are identified. One particular advantage is the ability of the system to model domains with approximately the same number of structures as there are underlying concepts. This leads to the possibility of extraction of higher level mathematical descriptions of the induced concepts, using the techniques of computational geometry, which is not possible from a multiplicity of small regions.