Improved support vector machine generalization using normalized input space

Data pre-processing always plays a key role in learning algorithm performance. In this research we consider data pre-processing by normalization for Support Vector Machines (SVMs). We examine the normalization affect across 112 classification problems with SVM using the rbf kernel. We observe a significant classification improvement due to normalization. Finally we suggest a rule based method to find when normalization is necessary for a specific classification problem. The best normalization method is also automatically selected by SVM itself.

Nonlinear discrimination using support vector machine

Appropriate training data always play an important role in constructing an efficient classifier to solve the data mining classification problem. Support Vector Machine (SVM) is a comparatively new approach in constructing a model/classifier for data analysis, based on Statistical Learning Theory (SLT). SVM utilizes a transformation of the basic constrained optimization problem compared to that of a quadratic programming method, which can be solved parsimoniously through standard methods. Our research focuses on SVM to classify a number of different sizes of data sets. We found SVM to perform well in the case of discrimination compared to some other existing popular classifiers.

A need for national regulation on property investment advice and marketeering - suggestions put forward

The property investment advice and marketeering industry is currently unregulated in Australia. There is no uniform national or state regulation in this area. The only protection and remedies currently available are those under the general consumer protection laws scattered in various Acts, and even so, these have numerous problems.This article sets out to argue for a new set of laws to regulate property investment advice and marketeering. In providing suggestions for reform, the article also argues that, to overcome the constitutional difficulty, a national co-operative approach is the only way to move forward in this area and suggests that a new regulator be set up to administer and enforce the new proposed laws on property investment advice and marketeering.

An ensemble of classifiers with genetic algorithmBased Feature Selection

Different data classification algorithms have been developed and applied in various areas to analyze and extract valuable information and patterns from large datasets with noise and missing values. However, none of them could consistently perform well over all datasets. To this end, ensemble methods have been suggested as the promising measures. This paper proposes a novel hybrid algorithm, which is the combination of a multi-objective Genetic Algorithm (GA) and an ensemble classifier. While the ensemble classifier, which consists of a decision tree classifier, an Artificial Neural Network (ANN) classifier, and a Support Vector Machine (SVM) classifier, is used as the classification committee, the multi-objective Genetic Algorithm is employed as the feature selector to facilitate the ensemble classifier to improve the overall sample classification accuracy while also identifying the most important features in the dataset of interest. The proposed GA-Ensemble method is tested on three benchmark datasets, and compared with each individual classifier as well as the methods based on mutual information theory, bagging and boosting. The results suggest that this GA-Ensemble method outperform other algorithms in comparison, and be a useful method for classification and feature selection problems.

Alterations in renal and myocardial adrenoceptors associated with ethynyloestradiol- and levonorgestrel-induced hypertension in the rat

Hypertension is one of many side effects of oral contraceptive use in a small percentage of women. Although the underlying pathology has yet to be fully resolved, alterations in the renin-angiotensin-aldosterone axis, sympathetic nervous system/ renal and cardiac function have been implicated. In the thesis to be presented, the possible involvement of alterations in renal and myocardial adrenoceptor characteristics in the pathogenesis of steroid contraceptive-induced hypertension in rats was examined by radioligand binding techniques. In Chapter 2, a rat model of OC hypertension is described. Chronic low-dose administration of ethynyloestradiol (EE2), levonorgestrel (NG) or a combination of both steroids (EE2/NG) to female Sprague-Dawley rats was shown to significantly increase systolic blood pressure (SBP). Renal and cardiac hypertrophy developed in association with EE2-, EE2/NG- but not NG-induced hypertension. Moreover, whereas administration of NG alone attenuated body weight gain, combined EE2/NG administration increased body weight gain from the second week of treatment onwards. Based on the above observations, it is proposed that EE2 and NG induce hypertension in rats via different mechanisms. Although SBP was elevated to a similar maximum in all steroid-treated groups (+ 20 mmHg compared to controls), only with EE2 administration did SBP remain elevated for the duration of the 17 week treatment regimen. NG may therefore have a protective effect on blood pressure with long-term combined steroid contraceptive treatment. In Chapter 4, renal adrenoceptors were characterized using radioactively labelled adrenocephor antagonists. Under appropriate conditions, binding of [3H]-prazosin and [3H]-rauwolscine to membrane preparations of whole rat kidney displayed the kinetics, saturability and specificity of α1- and α2 -adrenoceptors respectively, which were present in a ratio 3:1. In contrast, [3H]-dihydroergocryptine ([3H]-DHE) apparently bound to both α1 and α2-adrenoceptors. Binding sites identified by [125I] –iodocyanopindolol (ICYP) had the recognition characteristics of β-adrenoceptors. In drug competition studies using the subtype-selective antagonists practolol (β1) and ICI 118,551 (β2)/ the ratio of β1- to β2 -adrenoceptors was found to be approximately 2:1. Subsequently, renal adrenoceptors were investigated at various stages during the development of hypertension with the different steroid contraceptive treatments (Chapters 5 and 6). Preliminary binding studies with [3H]-DHE and [3H]-prazosin suggested that the number of renal α2 - but not α1-adrenoceptors was reduced in rats with established EE2-induced hypertension (17 weeks treatment). This was subsequently confirmed using [3H]-rauwolscine, which in addition showed that the reduction in renal α2 -adrenoceptor number occurred during the developmental stage of EE2/NG~induced hypertension (6 weeks treatment) and established EE2-induced hypertension (12 weeks treatment). NG induced hypertension was unassociated with changes in renal α1- and α2-adrenoceptor characteristics. Renal β-adrenoceptor affinity was reduced in established EE2-, but not NG- or EE2/NG- induced hypertension. Moreover, the β-adrenoceptor agonist (-)-isoprenaline bound to renal β-adrenoceptors with reduced affinity following EE2 administration. Several endogenous and synthetic steroids were found to be ineffective inhibitors of [3H] –prazosin, [3H] –rauwolscine and ICYP binding excluding a direct interaction of these steroids with renal α1-, α2- and β -adrenoceptors. In Chapter 7, myocardial adrenoceptors were characterized and investigated in steroid-treated rats. In membrane preparations of whole myocardium, [3H]-prazosin binding was characteristically to α1- adrenoceptors, whereas there was a notable absence of [3H]-rauwolscine binding. Using ICYP, β-adrenoceptors were also detected, the ratio of β1- to β2~adrenoceptors being 3:1. Steroid contraceptive-induced hypertension was not associated with myocardial α1-adrenoceptor changes. Similarly, myocardial β-adrenoceptors were unchanged in established EE2-, NG- and EE2/NG-induced hypertension (12 weeks treatment). The affinity of (-)-isoprenaline for myocardial β-adrenoceptors was unaffected by EE2 aditiinistration. These studies suggest that established EE2- but not NG-induced hypertension in rats is associated with selective alterations in renal α2- and (β-adrenoceptors. These adrenoceptor changes may help to maintain elevated blood pressure by affecting the control of renal function by the sympathetic nervous system, catecholamines and several hormones which affect renin release and the transport of fluid and electrolytes in the nephron.

The effect of organisational culture perceptions on the relationship between budgetary participation and managerial job-related outcomes

We examine the impact of managers' perceptions of their organisational culture (OC) on the relationship between budgetary participation (BP) and managerial job-related outcomes, operationalised as managerial performance and job-related tension (JRT). Does the relationship between BP and job-related outcomes would depend on managers' perceptions of innovation and attention to detail? Data supported predictions that increasing BP would lower JRT for managers perceiving a high emphasis on innovation within their OC, regardless of their perceptions of an emphasis on attention to detail. When managers perceived low innovation, however, their perception of level of attention to detail had a significant effect on the relationship between BP and JRT. More specifically, increasing BP was found to decrease JRT for managers who perceived low innovation and low attention to detail. For managers who perceived low innovation and high attention to detail, however, this effect was attenuated. Finally, the positive relationship between BP and managerial performance was not found to he affected significantly by managers' OC perceptions.

Minimizing the limitations of GL analyser of fusion based email classification

In the last decade, the Internet email has become one of the primary method of communication used by everyone for the exchange of ideas and information. However, in recent years, along with the rapid growth of the Internet and email, there has been a dramatic growth in spam. Classifications algorithms have been successfully used to filter spam, but with a certain amount of false positive trade-offs. This problem is mainly caused by the dynamic nature of spam content, spam delivery strategies, as well as the diversification of the classification algorithms. This paper presents an approach of email classification to overcome the burden of analyzing technique of GL (grey list) analyser as further refinements of our previous multi-classifier based email classification [10]. In this approach, we introduce a “majority voting grey list (MVGL)” analyzing technique with two different variations which will analyze only the product of GL emails. Our empirical evidence proofs the improvements of this approach, in terms of complexity and cost, compared to existing GL analyser. This approach also overcomes the limitation of human interaction of existing analyzing technique.

Prediction of wool knitwear pilling propensity using support vector machines

The propensity of wool knitwear to form entangled fiber balls, known as pills, on the surface is affected by a large number of factors. This study examines, for the first time, the application of the support vector machine (SVM) data mining tool to the pilling propensity prediction of wool knitwear. The results indicate that by using the binary classification method and the radial basis function (RBF) kernel function, the SVM is able to give high pilling propensity prediction accuracy for wool knitwear without data over-fitting. The study also found that the number of records available for each pill rating greatly affects the learning and prediction capability of SVM models.

Heterometallic CeIII–FeIII–salicylate networks : models for corrosion mitigation of steel surfaces by the ′Green′ inhibitor, Ce(salicylate)3

The syntheses and structures of the novel Ce–Fe bimetallic complexes [{Fe(sal)₂(bpy)}₂Ce(NO₃)(H₂O)₃]·EtOH and [{Fe(sal)₂(bpy)}₄Ce₂(H₂O)₁₁][salH]₂·EtOH·3H₂O (salH₂ = salicylic acid) suggest Fe³⁺–sal²⁻ units and Ce–OC(R)O–Fe bridging contribute to the formation of corrosion inhibitive layers on steel surfaces exposed to [Ce(salH)₃(H₂O)].

Statutory verification of water rights: the 'insuperable' difficulties of propertising water entitlements

The abolition of riparian entitlements in the early stages of colonial Australia and the vesting of these rights in the Crown represented a turning point for the evolution of private water rights. The extinguishment of common law rights connected to vested land interests and the introduction of new, unaligned statutory entitlements provided a new and fundamentally different system for the creation and regulation of private water entitlements. Unlike riparian entitlements, in the absence of express definition, statutory water entitlements may only be verified as property where such a construction is consistent with the nature and scope of the entitlement. In this respect, the statutory framework has disaggregated the propertisation of water rights from land ownership and linked the process to broader statutory interpretation principles. The shift away from institutional property has generated concerns about the interpretive approaches appropriate for the verification of legislative water entitlements. This article examines the existing interpretive approaches and argues that the blurring of the propertisation process with the separate issue of whether any change or modification of such water rights attracts s 51(xxxi) of the Commonwealth Constitution has produced a situation where core property indicia is increasingly overshadowed by legislative defeasibility. In the recent High Court decision of ICM Agriculture Pty Ltd v Commonwealth, the focus of the majority judgements upon the inherent susceptibility of legislative entitlements to variation or extinguishment acted as a catalyst for the non-propertisation of statutory bore water licences in New South Wales. The emphasis the majority judgements gave to legislative defeasibility precluded a full and balanced assessment of other highly relevant property indicia, in particular the expectation interests of the holders. Conflating property and constitutional evaluation in this way is inappropriate in an era where entitlements to natural resource interests are increasingly statute based and the verification process has significant social and economic repercussions. Determining whether a statutory entitlement constitutes property requires a careful balancing of legislative intent, social and environmental context and individual expectation and the vicissitudes of a regulatory context should not eclipse this process.

Delivering categorized news items using RSS feeds and web services

In the past decade the massive growth of the Internet brought huge changes in the way humans live their daily life; however, the biggest concern with rapid growth of digital information is how to efficiently manage and filter unwanted data. In this paper, we propose a method for managing RSS feeds from various news websites. A Web service was developed to provide filtered news items extracted from RSS feeds and these were categorized based on classical text categorization algorithms. A client application consuming this Web service retrieves and displays such filtered information. A prototype was implemented using Rapidminer 4.3 as a data mining tool and SVM as a classification algorithm. Experimental results suggest that the proposed method is effective and saves a significant amount of user processing time.

Vehicle detection and classification by measuring and processing magnetic signal

This paper presents novel vehicle detection and classification method by measuring and processing magnetic signal based on single micro-electro- mechanical system (MEMS) magnetic sensor. When a vehicle moves over the ground, it generates a succession of impacts on the earth's magnetic field, which can be detected by single magnetic sensor. The magnetic signal measured by the magnetic sensor is related to the moving direction and the type of the vehicle. Generally, the recognition rate using single sensor detector is not high. In order to improve the recognition rate, a novel feature extraction algorithm and a novel vehicle classification and recognition algorithm are presented. The concavity and convexity areas, and the angles of concave and convex parts of the waveform are extracted. An improved support vector machine (ISVM) classifier is developed to perform vehicle classification and recognition. The effectiveness of the proposed approach is verified by outdoor experiments.