Special issue on advances in data mining and its applications

Data mining is the process to identify valid, implicit, previously unknown, potentially useful and understandable information from large databases. It is an important step in the process of knowledge discovery in databases, (Olaru & Wehenkel, 1999). In a data mining process, input data can be structured, seme-structured, or unstructured. Data can be in text, categorical or numerical values. One of the important characteristics of data mining is its ability to deal data with large volume, distributed, time variant, noisy, and high dimensionality. A large number of data mining algorithms have been developed for different applications. For example, association rules mining can be useful for market basket problems, clustering algorithms can be used to discover trends in unsupervised learning problems, classification algorithms can be applied in decision-making problems, and sequential and time series mining algorithms can be used in predicting events, fault detection, and other supervised learning problems (Vapnik, 1999). Classification is among the most important tasks in the data mining, particularly for data mining applications into engineering fields. Together with regression, classification is mainly for predictive modelling. So far, there have been a number of classification algorithms in practice. According to (Sebastiani, 2002), the main classification algorithms can be categorized as: decision tree and rule based approach such as C4.5 (Quinlan, 1996); probability methods such as Bayesian classifier (Lewis, 1998); on-line methods such as Winnow (Littlestone, 1988) and CVFDT (Hulten 2001), neural networks methods (Rumelhart, Hinton & Wiliams, 1986); example-based methods such as k-nearest neighbors (Duda & Hart, 1973), and SVM (Cortes & Vapnik, 1995). Other important techniques for classification tasks include Associative Classification (Liu et al, 1998) and Ensemble Classification (Tumer, 1996).

A common inhibitory binding site for zinc and odorants at the voltage-gated K+ channel of rat olfactory receptor neurons

This study compared the effects of zinc and odorants on the voltage-gated K+ channel of rat olfactory neurons. Zinc reduced current magnitude, depolarized the voltage activation curve and slowed activation kinetics without affecting inactivation or deactivation kinetics. Zinc inhibition was potentiated by the NO compound, S-nitroso-cysteine. The pH- and diethylpyrocarbonate-dependence of zinc inhibition suggested that zinc acted by binding to histidine residues. Cysteine residues were eliminated as contributing to the zinc-binding site. The odorants, acetophenone and amyl acetate, also reduced current magnitude, depolarized the voltage activation curve and selectively slowed activation kinetics. Furthermore, the diethylpyrocarbonate- and pH-dependence of odorant inhibition implied that the odorants also bind to histidine residues. Zinc inhibitory potency was dramatically diminished in the presence of odorants, implying competition for a common binding site. These observations indicate that the odorants and zinc share a common inhibitory binding site on the external surface of the voltage-gated K+ channel.

Ciguatoxins: Cyclic polyether modulators of voltage-gated Iion channel function

Ciguatoxins are cyclic polyether toxins, derived from marine dinoflagellates, which are responsible for the symptoms of ciguatera poisoning. Ingestion of tropical and subtropical fin fish contaminated by ciguatoxins results in an illness characterised by neurological, cardiovascular and gastrointestinal disorders. The pharmacology of ciguatoxins is characterised by their ability to cause persistent activation of voltage-gated sodium channels, to increase neuronal excitability and neurotransmitter release, to impair synaptic vesicle recycling, and to cause cell swelling. It is these effects, in combination with an action to block voltage-gated potassium channels at high doses, which are believed to underlie the complex of symptoms associated with ciguatera. This review examines the sources, structures and pharmacology of ciguatoxins. In particular, attention is placed on their cellular modes of actions to modulate voltage-gated ion channels and other Na+-dependent mechanisms in numerous cell types and to current approaches for detection and treatment of ciguatera.

Presentation and detection of invasive melanoma in a high-risk population

Background: Early detection of melanoma has been encouraged in Queensland for many years, yet little is known about the patterns of detection and the way in which they relate to tumor thickness. Objective: Our purpose was to describe current patterns of melanoma detection in Queensland. Methods: This was a population-based study, comprising 3772 Queensland residents diagnosed with a histologically confirmed melanoma between 2000 and 2003. Results: Almost half (44.0%) of the melanomas were detected by the patients themselves, with physicians detecting one fourth (25.3%) and partners one fifth (18.6%). Melanomas detected by doctors were more likely to be thin (\0.75 mm) than those detected by the patient or other layperson. Melanomas detected during a deliberate skin examination were thinner than those detected incidentally. Limitations: Although a participation rate of 78% was achieved, as in any survey, nonresponse bias cannot be completely excluded, and the ability of the results to be generalized to other geographical areas is unknown. Conclusion: There are clear differences in the depth distribution of melanoma in terms of method of detection and who detects the lesions that are consistent with, but do not automatically lead to, the conclusion that promoting active methods of detection may be beneficial. ( J Am Acad Dermatol 2006;54:783-92.)

Load profiling and data mining techniques in electricity deregulated market

This paper presents load profiles of electricity customers, using the knowledge discovery in databases (KDD) procedure, a data mining technique, to determine the load profiles for different types of customers. In this paper, the current load profiling methods are compared using data mining techniques, by analysing and evaluating these classification techniques. The objective of this study is to determine the best load profiling methods and data mining techniques to classify, detect and predict non-technical losses in the distribution sector, due to faulty metering and billing errors, as well as to gather knowledge on customer behaviour and preferences so as to gain a competitive advantage in the deregulated market. This paper focuses mainly on the comparative analysis of the classification techniques selected; a forthcoming paper will focus on the detection and prediction methods.

Off-line signature verification and forgery detection system based on fuzzy modeling

This paper presents an innovative approach for signature verification and forgery detection based on fuzzy modeling. The signature image is binarized and resized to a fixed size window and is then thinned. The thinned image is then partitioned into a fixed number of eight sub-images called boxes. This partition is done using the horizontal density approximation approach. Each sub-image is then further resized and again partitioned into twelve further sub-images using the uniform partitioning approach. The features of consideration are normalized vector angle (α) from each box. Each feature extracted from sample signatures gives rise to a fuzzy set. Since the choice of a proper fuzzification function is crucial for verification, we have devised a new fuzzification function with structural parameters, which is able to adapt to the variations in fuzzy sets. This function is employed to develop a complete forgery detection and verification system.

The effects of prepulse-blink reflex trial repetition and prepulse change on blink reflex modification at short and long lead intervals

Prepulse inhibition and facilitation of the blink reflex are said to reflect different responses elicited by the lead stimulus, transient detection and orienting response respectively. Two experiments investigated the effects of trial repetition and lead stimulus change on blink modification. It was hypothesized that these manipulations will affect orienting and thus blink facilitation to a greater extent than they will affect transient detection and thus blink inhibition. In Experiment 1 (N = 64), subjects were trained with a sequence of 12 lead stimulus and 12 blink stimulus alone presentations, and 24 lead stimulus-blink stimulus pairings. Lead interval was 120 ms for 12 of the trials and 2000 ms for the other 12. For half the subjects this sequence was followed by a change in pitch of the lead stimulus. In Experiment 2 (N = 64), subjects were trained with a sequence of 36 blink alone stimuli and 36 lead stimulus-blink stimulus pairings. The lead interval was 120 ms for half the subjects and 2000 ms for the other half. The pitch of the lead stimulus on prestimulus trials 31-33 was changed for half the subjects in each group. In both experiments, the amount of blink inhibition decreased during training whereas the amount of blink facilitation remained unchanged. Lead stimulus change had no effect on blink modification in either experiment although it resulted in enhanced skin conductance responses and greater heart rate deceleration in Experiment 2. The present results are not consistent with the notion that blink facilitation is linked to orienting whereas blink inhibition reflects a transient detection mechanism. (C) 1998 Elsevier Science B.V.

Prediction of MHC class II-binding peptides using an evolutionary algorithm and artificial neural network

Motivation: Prediction methods for identifying binding peptides could minimize the number of peptides required to be synthesized and assayed, and thereby facilitate the identification of potential T-cell epitopes. We developed a bioinformatic method for the prediction of peptide binding to MHC class II molecules. Results: Experimental binding data and expert knowledge of anchor positions and binding motifs were combined with an evolutionary algorithm (EA) and an artificial neural network (ANN): binding data extraction --> peptide alignment --> ANN training and classification. This method, termed PERUN, was implemented for the prediction of peptides that bind to HLA-DR4(B1*0401). The respective positive predictive values of PERUN predictions of high-, moderate-, low- and zero-affinity binder-a were assessed as 0.8, 0.7, 0.5 and 0.8 by cross-validation, and 1.0, 0.8, 0.3 and 0.7 by experimental binding. This illustrates the synergy between experimentation and computer modeling, and its application to the identification of potential immunotheraaeutic peptides.

Effects of habitat modification in mangroves on the structure of mollusc and crab assemblages

The abundance and species richness of mollusc and crab assemblages were examined in a subtropical mangrove forest in Moreton Bay, Queensland, Australia, which has been disturbed and damaged by the construction of a wooden boardwalk and a path. Sections of the forest immediately adjacent to the boardwalk and path were compared with reference areas to determine whether changes to the small-scale structural complexity within the forest affected the benthic fauna. The disturbed area was characterised by having 65-80% fewer pneumatophores, significantly fewer species and individuals of molluscs, but significantly more species and individuals of crabs than the reference areas. The abundance of mangrove pneumatophores and the attached epiphytic algae were manipulated at two sites to determine whether observed differences in these features could account for the differences in the assemblage of molluscs in the disturbed area of the forest compared with reference areas. Five experimental treatments were used: undisturbed controls, pneumatophore removals (abundance reduced by ca. 65%), epiphytic algal removals (algae removed from ca. 65% of pneumatophores), pneumatophore disturbance controls and algal disturbance controls. The experimental reduction of the abundance of mangrove pneumatophores and the associated epiphytic algae led to significant declines (by as much as 83%) in the number of molluscs utilising the substratum in the modified plots. There was no significant difference in the abundance of molluscs in the pneumatophore and algal removal plots suggesting any effect was primarily related to removal of the epiphytic algae from the surface of the pneumatophores. The responses by the biota to the changes in the physical environment demonstrate that even relatively small-scale modifications to the physical structure of subtropical mangrove forests can lead to significant effects on the diversity and abundance of macrobenthic organisms in these habitats. Such modifications have the potential to cause cascading effects at higher trophic levels with a deterioration in the value of these habitats as nursery and feeding grounds. Future efforts at conservation of these estuarine environments must focus on the prevention or reduction of modifications to the physical structure and integrity of the system, rather than just on the prevention of loss of entire patches of habitat. (C) 2000 Elsevier Science B.V. All rights reserved.