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.

Image retrieval based on the combination of color and keyword

In this paper, color information and keyword information are combined for image retrieval. In detail, each image is divided into several blocks and then the color histograms of each block are derived. Users could feed back some annotations represented by keywords. Then, the keywords may spread in the image database so that both color-based and keyword-based retrieval could be utilized together. A prototype system shows that the proposed method is effective and efficient in performing image retrieval tasks.

Dietary combination of soy with a probiotic or prebiotic food significantly reduces total and LDL cholesterol in mildly hypercholesterolaemic subjects

Objective: We hypothesized that a dietary combination of soy with either a probiotic (yoghurt) or a prebiotic (resistant starch) would result in enhanced lipid-lowering effects compared with a control soy diet, possibly via improvements in isoflavone bioavailability.

Subjects: Mildly hypercholesterolaemic subjects (men and post-menopausal women) older than 45 years were recruited via the local media. Thirty-six subjects commenced the study; five withdrew.

Results: Soy+probiotic significantly decreased total cholesterol (4.72.0%; P=0.038) and soy+prebiotic significantly decreased total and low-density lipoprotein cholesterol (5.51.6%; P=0.003 and 7.32.2%; P=0.005, respectively). The bioavailabilities of daidzein, genistein or equol were not affected by probiotic or prebiotic consumption or associated with lipid changes.

Conclusion: Dietary combination of soy with either a probiotic or a prebiotic resulted in significant lipid lowering, not related to isoflavone bioavailability.

Dual responsive chemosensors for anions : the combination of fluorescent PET (Photoinduced Electron Transfer) and colorimetric chemosensors in a single molecule

The design and synthesis of two novel fluorescent PET anion sensors is described, based on the principle of ‘fluorophore-spacer-(anion)receptor’. The sensors 1 and 2 employ simple diaromatic thioureas as anion receptors, and the fluorophore is a naphthalimide moiety that absorbs in the visible part of the spectrum and emits in the green. Upon recognition of anions such as F⁻ and AcO⁻ in DMSO, the fluorescence emission of 1 and 2 was ‘switched off’, with no significant changes in the UV–vis spectra. This recognition shows a 1:1 binding between the receptor and the anions. In the case of F⁻, further additions of the anion, gave rise to large changes in the UV–vis spectra, where the λ_max at 455 nm was shifted to 550 nm. These changes are thought to be due to the deprotonation of the 4-amino moiety of the naphthalimide fluorophore. This was in fact found to be the case, using simple naphthalimide derivatives such as 6. Sensors 1 and 2 can thus display dual sensing action; where at low concentrations, the fluorescence emission is quenched, and at higher concentrations the absorption spectra are modulated.

Epidemiologic merit of obese-years, the combination of degree and duration of obesity

This study aims to test the effect of combining the degree and the duration of obesity into a single variable-obese-years-and to examine whether obese-years is a better predictor of the risk of diabetes than simply body mass index (BMI) or duration of obesity. Of the original cohort of the Framingham Heart Study, 5,036 participants were followed up every 2 years for up to 48 years (from 1948). The variable, obese-years, was defined by multiplying for each participant the number of BMI units above 30 kg/m(2) by the number of years lived at that BMI. Associations with diabetes were analyzed by using time-dependent Cox proportional hazards regression models adjusted for potential confounders. The incidence of type-2 diabetes increased as the number of obese-years increased, with adjusted hazard ratios of 1.07 (95% confidence interval: 1.06, 1.09) per additional 10 obese-years. The dose-response relation between diabetes incidence and obese-years varied by sex and smoking status. The Akaike Information Criterion was lowest in the model containing obese-years compared with models containing either the degree or duration of obesity alone. A construct of obese-years is strongly associated with risk of diabetes and could be a better indicator of the health risks associated with increasing body weight than BMI or duration of obesity alone.

Combination of decisions from a multiple neural network classifier system

In this paper, a study of the effectiveness of a multiple classifier system (MCS) in a medical diagnostic task is described. A hybrid network, based on the integration of a fuzzy ARTMAP and the probabilistic neural network, is employed as the basis of the MCS. Outputs from multiple networks are combined using some decision combination method to reach a final prediction. By using a real medical database, a set of experiments has been conducted to evaluate the performance of the MSC with different network configurations. The experimental results reveal the potential of the MCS as a useful decision support tool in the medical field.

A multi-tier ensemble construction of classifiers for phishing email detection and filtering

This paper is devoted to multi-tier ensemble classifiers for the detection and filtering of phishing emails. We introduce a new construction of ensemble classifiers, based on the well known and productive multi-tier approach. Our experiments evaluate their performance for the detection and filtering of phishing emails. The multi-tier constructions are well known and have been used to design effective classifiers for email classification and other applications previously. We investigate new multi-tier ensemble classifiers, where diverse ensemble methods are combined in a unified system by incorporating different ensembles at a lower tier as an integral part of another ensemble at the top tier. Our novel contribution is to investigate the possibility and effectiveness of combining diverse ensemble methods into one large multi-tier ensemble for the example of detection and filtering of phishing emails. Our study handled a few essential ensemble methods and more recent approaches incorporated into a combined multi-tier ensemble classifier. The results show that new large multi-tier ensemble classifiers achieved better performance compared with the outcomes of the base classifiers and ensemble classifiers incorporated in the multi-tier system. This demonstrates that the new method of combining diverse ensembles into one unified multi-tier ensemble can be applied to increase the performance of classifiers if diverse ensembles are incorporated in the system.

Improving load forecasting accuracy through combination of best forecasts

Neural network (NN) models have been widely used in the literature for short-term load forecasting. Their popularity is mainly due to their excellent learning and approximation capability. However, their forecasting performance significantly depends on several factors including initializing parameters, training algorithm, and NN structure. To minimize negative effects of these factors, this paper proposes a practically simple, yet effective and an efficient method to combine forecasts generated by NN models. The proposed method includes three main phases: (i) training NNs with different structures, (ii) selecting best NN models based on their forecasting performance for a validation set, and (iii) combination of forecasts for selected best NNs. Forecast combination is performed through calculating the mean of forecasts generated by best NN models. The performance of the proposed method is examined using real world data set. Comparative studies demonstrate that the accuracy of combined forecasts is significantly superior to those obtained from individual NN models.

Do leatherback turtles Dermochelys coriacea forage during the breeding season? A combination of data-logging devices provide new insights

Animals which undertake migrations from foraging grounds to suitable breeding areas must adopt strategies in these new conditions in order to minimise the rate at which body condition deteriorates (which will occur due to oogenesis or provisioning for young). For some animals this involves continuing foraging, whereas for others the optimal strategy is to fast during the breeding season. The leatherback turtle undertakes long-distance migrations from temperate zones to tropical breeding areas, and in some of these areas it has been shown to exhibit diving behaviour indicative of foraging. We used conventional time–depth recorders and a single novel mouth-opening sensor to investigate the foraging behaviour of leatherback turtles in the southern Caribbean. Diving behaviour suggested attempted foraging on vertically migrating prey with significantly more diving to a more consistent depth occurring during the night. No obvious prey manipulation was detected by the mouth sensor, but rhythmic mouth opening did occur during specific phases of dives, suggesting that the turtle was relying on gustatory cues to sense its immediate environment. Patterns of diving in conjunction with these mouth-opening activities imply that leatherbacks are attempting to forage during the breeding season and that gustatory cues are important to leatherbacks.

Detection of fingertips based on the combination of color information and circle detection

Fingertips of human hand play an important role in hand-based interaction with computers. Therefore, identification of fingertips' positions on hand image is vital for developing a human computer interaction system. All most all of the research works for fingertips detection, initially isolate hand image from the background image. Most of these techniques develop color based segmentation methods because human skin color possess an exceptional characterises that can be used to isolate hand from the rest of the image quite easily. Sometimes color image segmentation becomes difficult due to illumination and background variations. To make it simple and reliable, this paper proposes a robust method for detecting fingertips of a hand image based on the combination of color segmentation and circle detection. Due to the characteristics of circularity of fingertips regions of hand boundary, any existing circle detection algorithms can be applied to detect circles at fingertips region. It is difficult to detect fingertips solely based on the circle detection method. For this reason, initially the proposed method detects all the circular regions on the image applying Circle Hough Transformation (CHT) then the fingertips are selected based on the color characteristics of the fingertips regions. Experimental results show that the proposed approach is promising.

Hybrid controller with the combination of FLC and neural network-based IMC for nonlinear processes

This work presents a hybrid controller based on the combination of fuzzy logic control (FLC) mechanism and internal model-based control (IMC). Neural network-based inverse and forward models are developed for IMC. After designing the FLC and IMC independently, they are combined in parallel to produce a single control signal. Mean averaging mechanism is used to combine the prediction of both controllers. Finally, performance of the proposed hybrid controller is studied for a nonlinear numerical plant model (NNPM). Simulation result shows the proposed hybrid controller outperforms both FLC and IMC.

Combination of calcium and magnesium ions prevents substrate inhibition and promotes biomass and lipid production in thraustochytrids under higher glycerol concentration

Effect of calcium and magnesium ions was studied in detail in batch mode in shake flask cultures of two fast growing strains of thraustochytrids (Aurantiochytrium sp. DBTIOC-18 and Schizochytrium sp. DBTIOC-1) for biomass and lipid production. These strains were previously isolated from Indian marine biodiversity. Screening of these two strains on different carbon and nitrogen sources revealed the suitability of glycerol over glucose and sodium nitrate over yeast extract for the cultivation of these strains. The presence of higher concentration of glycerol in the medium inhibited the glycerol utilization by the cell thus resulting in lower biomass and lipid production in both the strains. Supplementing media with calcium and magnesium ions promoted glycerol utilization thus resulted in a substantial rise in volumetric production of biomass (55.12 g L-1, 48.12 g L-1), fatty acid for biodiesel (27.14 g L-1, 22.15 g L-1) and docosahexaenoic acid (14.57 g L-1, 10.12 g L-1) with both strains Aurantiochytrium sp. DBTIOC-18 and Schizochytrium sp. DBTIOC-1, respectively. Growth profile study of these two strains showed further improvement in production of biomass, fatty acid for biodiesel and docosahexaenoic acid when cultures were extended up to 7 days. Finding of this work underlines the importance of calcium and magnesium salts in designing new fermentation strategies to prevent substrate inhibition and achieve high cell density culture under high nutrient concentration especially carbon sources.

The combination of plyometric and balance training improves sprint and shuttle run performances more often than plyometric-only training with children

Because balance is not fully developed in children and studies have shown functional improvements with balance only training studies, a combination of plyometric and balance activities might enhance static balance, dynamic balance, and power. The objective of this study was to compare the effectiveness of plyometric only (PLYO) with balance and plyometric (COMBINED) training on balance and power measures in children. Before and after an 8-week training period, testing assessed lower-body strength (1 repetition maximum leg press), power (horizontal and vertical jumps, triple hop for distance, reactive strength, and leg stiffness), running speed (10-m and 30-m sprint), static and dynamic balance (Standing Stork Test and Star Excursion Balance Test), and agility (shuttle run). Subjects were randomly divided into 2 training groups (PLYO [n = 14] and COMBINED [n = 14]) and a control group (n = 12). Results based on magnitude-based inferences and precision of estimation indicated that the COMBINED training group was considered likely to be superior to the PLYO group in leg stiffness (d = 0.69, 91% likely), 10-m sprint (d = 0.57, 84% likely), and shuttle run (d = 0.52, 80% likely). The difference between the groups was unclear in 8 of the 11 dependent variables. COMBINED training enhanced activities such as 10-m sprints and shuttle runs to a greater degree. COMBINED training could be an important consideration for reducing the high velocity impacts of PLYO training. This reduction in stretch-shortening cycle stress on neuromuscular system with the replacement of balance and landing exercises might help to alleviate the overtraining effects of excessive repetitive high load activities.

A combination of omega-3 fatty acids, folic acid and B-group vitamins is superior at lowering homocysteine than omega-3 alone: a meta-analysis

The aim of the study was to assess whether omega-3 polyunsaturated fatty acid supplementation alone or in combination with folic acid and B-group vitamins is effective in lowering homocysteine. The Medline Ovid, Embase and Cochrane databases were searched for randomized-controlled trial studies that intervened with omega-3 supplementation (with or without folic acid) and measured changes in homocysteine concentration. Studies were pooled using a random effects model for meta-analysis. Three different models were analyzed: all trials combined, omega-3 polyunsaturated fatty acid trials, and omega-3 polyunsaturated fatty acids with folic acid and B-group vitamin trials. Nineteen studies were included, consisting of 3267 participants completing 21 trials. Studies were heterogeneous; varying by dose, duration and participant health conditions. Across all trials, omega-3 supplementation was effective in lowering homocysteine by an average of 1.18μmol/L (95%CI: (-1.89, -0.48), P=.001). The average homocysteine-lowering effect was greater when omega-3 supplementation was combined with folic acid and B-group vitamins (-1.37μmol/L, 95%CI: (-2.38, -0.36), P<.01) compared to omega-3 supplementation alone (-1.09μmol/L 95%CI: (-2.04, -0.13), P=.03). Omega-3 polyunsaturated fatty acid supplementation was associated with a modest reduction in homocysteine. For the purposes of reducing homocysteine, a combination of omega-3s (0.2-6g/day), folic acid (150 - 2500μg/day) and vitamins B6 and B12 may be more effective than omega-3 supplementation alone.