Parameter estimation of one-class SVM on imbalance text classification

Compared with conventional two-class learning schemes, one-class classification simply uses a single class for training purposes. Applying one-class classification to the minorities in an imbalanced data has been shown to achieve better performance than the two-class one. In this paper, in order to make the best use of all the available information during the learning procedure, we propose a general framework which first uses the minority class for training in the one-class classification stage; and then uses both minority and majority class for estimating the generalization performance of the constructed classifier. Based upon this generalization performance measurement, parameter search algorithm selects the best parameter settings for this classifier. Experiments on UCI and Reuters text data show that one-class SVM embedded in this framework achieves much better performance than the standard one-class SVM alone and other learning schemes, such as one-class Naive Bayes, one-class nearest neighbour and neural network.

Development of a food composition database for the estimation of dietary intakes of glucosinolates, the biologically active constituents of cruciferous vegetables

Evidence indicates that cruciferous vegetables are protective against a range of cancers with glucosinolates and their breakdown products considered the biologically active constituents. To date, epidemiological studies have not investigated the intakes of these constituents due to a lack of food composition databases. The aim of the present study was to develop a database for the glucosinolate content of cruciferous vegetables that can be used to quantify dietary exposure for use in epidemiological studies of diet–disease relationships. Published food composition data sources for the glucosinolate content of cruciferous vegetables were identified and assessed for data quality using established criteria. Adequate data for the total glucosinolate content were available from eighteen published studies providing 140 estimates for forty-two items. The highest glucosinolate values were for cress (389 mg/100 g) while the lowest values were for Pe-tsai chinese cabbage (20 mg/100 g). There is considerable variation in the values reported for the same vegetable by different studies, with a median difference between the minimum and maximum values of 5·8-fold. Limited analysis of cooked cruciferous vegetables has been conducted; however, the available data show that average losses during cooking are approximately 36 %. This is the first attempt to collate the available literature on the glucosinolate content of cruciferous vegetables. These data will allow quantification of intakes of the glucosinolates, which can be used in epidemiological studies to investigate the role of cruciferous vegetables in cancer aetiology and prevention.

Simultaneous estimation of optical flow and object state: A modified approach to optical flow calculation

Differential optical flow methods are widely used within the computer vision community. They are classified as being either local, as in the Lucas-Kanade method, or global, as in the Horn-Schunck technique. As the physical dynamics of an object is inherently coupled into the behavior of its image in the video stream, in this paper, we use such dynamic parameter information in calculating optical flow when tracking a moving object using a video stream. Indeed, we use a modified error function in the minimization that contains physical parameter information. Further, the refined estimates of optical flow is used for better estimation of the physical parameters of the object in the simultaneous estimation of optical flow and object state(SEOS).

Depth estimation of metallic objects using multiwavelets scale-space representation

The problem of dimensional defects in aluminum die-castings is widespread throughout the foundry industry and their detection is of paramount importance in maintaining product quality. Due to the unpredictable factory environment and metallic with highly reflective nature, it is extremely hard to estimate true dimensionality of these metallic parts, autonomously. Some existing vision systems are capable of estimating depth to high accuracy, however are very much hardware dependent, involving the use of light and laser pattern projectors, integrated into vision systems or laser scanners. However, due to the reflective nature of these metallic parts and variable factory environments, the aforementioned vision systems tend to exhibit unpromising performance. Moreover, hardware dependency makes these systems cumbersome and costly. In this work, we propose a novel robust 3D reconstruction algorithm capable of reconstructing dimensionally accurate 3D depth models of the aluminum die-castings. The developed system is very simple and cost effective as it consists of only a pair of stereo cameras and a defused fluorescent light. The proposed vision system is capable of estimating surface depths within the accuracy of 0.5mm. In addition, the system is invariant to illuminative variations as well as orientation and location of the objects on the input image space, making the developed system highly robust. Due to its hardware simplicity and robustness, it can be implemented in different factory environments without a significant change in the setup. The proposed system is a major part of quality inspection system for the automotive manufacturing industry.

Object focused simultaneous estimation of optical flow and state dynamics

The framework of differential optical flow has been built upon to enhance the performance of motion estimation from optical flow. By coupling optical flow and object state parameters, an effective procedure for object tracking is implemented with the dasiaSimultaneous Estimation of Optical Flow and Object Statepsila (SEOS) technique. The SEOS method utilizes dynamic object parameter information when calculating optical flow for tracking a moving object within a video stream. Optical flow estimation for the SEOS method requires minimization of an error functional containing object physical parameter data. The convergence of an energy functional to a feasible or optimal solution set is not guaranteed. Convergence criteria is often assumed and not shown explicitly. Convergence of the SEOS method for both the Jacobi and Gauss-Seidel numerical resolution methods is evaluated.

Wireless estimation of canine pose for search and rescue

In this paper we discuss the use of accelerometers and Bluetooth to monitor canine pose in the context of common poses observed in urban search and rescue dogs. We discuss the use of the canine pose system in a disaster environment, and propose techniques for determining canine pose. In addition we discuss the challenges with this approach in such environments. The paper presents the experimental results obtained from the heavy urban search and rescue disaster simulation, where experiments were conducted using multiple canines, which show that angles can be derived from acceleration readings. Our experiments show that similar angles were measured for each of the poses, even when measured on multiple USAR canines of varying size. We also found measurable and consistent differences between each of the poses, making them clearly distinguishable from one another, again even when comparing with different USAR canines.

A simple 96-well microplate method for estimation of total polyphenolic content in seaweeds

Seaweed polyphenols are potent antioxidants and have also been shown to have α-glucosidase inhibiting activity. In our continuous efforts to develop new marine-based nutraceuticals and functional food ingredients, we have investigated many algal species collected on the Atlantic coast of Canada. A simple method for estimating the total polyphenol content in seaweeds and their extracts was developed based on the classic Folin-Ciocalteau colorimetric reaction. By using the 96-well microplate and a microplate reader, this new method saves experimental time, significantly reduces the amount of sample required, handles large number of samples in one experiment, and also improves the repeatability of the results.

A number of algal samples collected on the seashore of Nova Scotia, Canada, were analyzed for their levels of polyphenol content using this microplate-based method. The antioxidant activity of these samples was also assessed by using DPPH (2, 2-diphenyl-1-picrylhydrazyl) radical scavenging assay. The results showed that there is a strong correlation between the total polyphenol content and the potency of antioxidant effect.

Estimation of the dietary requirement for vitamin D in healthy adults

Background: Knowledge gaps have contributed to considerable variation among international dietary recommendations for vitamin D.

Objective: We aimed to establish the distribution of dietary vitamin D required to maintain serum 25-hydroxyvitamin D [25(OH)D] concentrations above several proposed cutoffs (ie, 25, 37.5, 50, and 80 nmol/L) during wintertime after adjustment for the effect of summer sunshine exposure and diet.

Design: A randomized, placebo-controlled, double-blind 22-wk intervention study was conducted in men and women aged 20–40 y (n = 238) by using different supplemental doses (0, 5, 10, and 15 µg/d) of vitamin D3 throughout the winter. Serum 25(OH)D concentrations were measured by using enzyme-linked immunoassay at baseline (October 2006) and endpoint (March 2007).

Results: There were clear dose-related increments (P < 0.0001) in serum 25(OH)D with increasing supplemental vitamin D3. The slope of the relation between vitamin D intake and serum 25(OH)D was 1.96 nmol·L–1·µg–1 intake. The vitamin D intake that maintained serum 25(OH)D concentrations of >25 nmol/L in 97.5% of the sample was 8.7 µg/d. This intake ranged from 7.2 µg/d in those who enjoyed sunshine exposure, 8.8 µg/d in those who sometimes had sun exposure, and 12.3 µg/d in those who avoided sunshine. Vitamin D intakes required to maintain serum 25(OH)D concentrations of >37.5, >50, and >80 nmol/L in 97.5% of the sample were 19.9, 28.0, and 41.1 µg/d, respectively.

Conclusion: The range of vitamin D intakes required to ensure maintenance of wintertime vitamin D status [as defined by incremental cutoffs of serum 25(OH)D] in the vast majority (>97.5%) of 20–40-y-old adults, considering a variety of sun exposure preferences, is between 7.2 and 41.1 µg/d.

Cell disruption optimization and covalent immobilization of beta-D-galactosidase from kluyveromyces marxianus YW-1 for lactose hydrolysis in milk

&beta;-D-galactosidase (EC 3.2.1.23) from Kluyveromyces marxianus YW-1, an isolate from whey, has been studied in terms of cell disruption to liberate the useful enzyme. The enzyme produced in a bioreactor on a wheat bran medium has been successfully immobilized with a view to developing a commercially usable technology for lactose hydrolysis in the food industry. Three chemical and three physical methods of cell disruption were tested and a method of grinding with river sand was found to give highest enzyme activity (720 U). The enzyme was covalently immobilized on gelatin. Immobilized enzyme had optimum pH and temperature of 7.0 and 40 °C, respectively and was found to give 49% hydrolysis of lactose in milk after 4 h of incubation. The immobilized enzyme was used for eight hydrolysis batches without appreciable loss in activity. The retention of high catalytic activity compared with the losses experienced with several previously reported immobilized versions of the enzyme is significant. The method of immobilization is simple, effective, and can be used for the immobilization of other enzymes.

Estimation of in vitro prebiotic activity of individual and mixed dietary fibres

Background: Epidemiological studies have demonstrated a link between dietary fibre deficiency and prevalence of many “Western diseases” particularly colon related diseases. Many of the health benefits associated with dietary fibre are attributed to their prebiotic effect. However, not all fibres have the same prebiotic potential or the same impact on colon health.

Objective: To examine the in vitro fermentation properties of individual and mixed dietary fibres by measuring fermentation byproducts over time.

Design: Wheat bran and guar gum were selected for this study. Individual and mixed dietary fibres were added to batch fermentation system and were inoculated with fresh faecal inoculum (n= 4). Positive (inulin) and negative (no substrate) fermenters were also used to determine the differences. The pH of the five fermenters was adjusted to a baseline of 5.5 and 6.8 representing the pH of the proximal and distal sections of the colon respectively. Samples were drawn out of the fermenters at 0, 3, 9 and 24 hours for the analysis of pH, ammonia and short chain fatty acids (SCFAs).

Outcomes: There were no significant differences in the pH levels at various time points between fermenters adjusted to pH 5.5 at baseline. However, in fermenters adjusted to pH 6.8 the pH of the fermenter containing wheat bran increased over the time (24h (P = 0.017)) due to production of a high amount of ammonia. The total SCFAs production was greater in fermenters containing combined fibres.

Conclusion: There is a large inter-individual variation in the prebiotic effect of all types of dietary fibres, however, in the present study, dietary fibre combinations showed greater prebiotic potential compared with the individual fibres.

On the simultaneous estimation of optical flow and object state

This thesis focuses on the enhancement of differential optical flow techniques. The framwork of differential optical flow has been built upon to improve object motion estimation within a video stream or image sequence. This augmentation comes in the form of a combined optical flow and object state estimation method (SEOS)

Robust ODF smoothing for accurate estimation of fiber orientation

Q-ball imaging was presented as a model free, linear and multimodal diffusion sensitive approach to reconstruct diffusion orientation distribution function (ODF) using diffusion weighted MRI data. The ODFs are widely used to estimate the fiber orientations. However, the smoothness constraint was proposed to achieve a balance between the angular resolution and noise stability for ODF constructs. Different regularization methods were proposed for this purpose. However, these methods are not robust and quite sensitive to the global regularization parameter. Although, numerical methods such as L-curve test are used to define a globally appropriate regularization parameter, it cannot serve as a universal value suitable for all regions of interest. This may result in over smoothing and potentially end up in neglecting an existing fiber population. In this paper, we propose to include an interpolation step prior to the spherical harmonic decomposition. This interpolation based approach is based on Delaunay triangulation provides a reliable, robust and accurate smoothing approach. This method is easy to implement and does not require other numerical methods to define the required parameters. Also, the fiber orientations estimated using this approach are more accurate compared to other common approaches.

Estimation of nitrogen and phosphorus in effluent from the striped catfish farming sector in the Mekong Delta, Vietnam

In this study an attempt is made to estimate nitrogen and phosphorus discharged to the environment from the striped catfish (Pangasianodon hypophthalmus) farming sector in the Mekong Delta (8°33̕–10°55̕ N, 104°30̕–106°50̕ E), South Vietnam. The sector accounted for 687,000 t production in 2007 and 1,094,879 t in 2008, with over 95% of the produce destined for export to over 100 countries. Commercial and farm-made feeds are used in catfish farming, currently the former being more predominant. Nitrogen discharge levels were similar for commercial feeds (median 46.0 kg/t fish) and farm-made feeds (median 46.8 kg/t fish); whilst, phosphorus discharge levels for commercial feeds (median 14.4 kg/t fish) were considerably lower than for farm-made feeds (median 18.4 kg/t fish). Based on the median nutrient discharge levels for commercial feeds, striped catfish production in the Mekong Delta discharged 31,602 t N and 9,893 t P, and 50,364 t N and 15,766 t P in 2007 and 2008, respectively. However, the amount of nutrients returned directly to the Mekong River may be substantially less than this as a significant proportion of the water used for catfish farming as well as the sludge is diverted to other agricultural farming systems. Striped catfish farming in the Mekong Delta compared favourably with other cultured species, irrespective of the type of feed used, when the total amounts of N and P discharged in the production of a tonne of production was estimated.

Estimation of basis frequencies for time-varying SIMO channels : a second-order method

The complex exponential basis expansion model (CE-BEM) provides an accurate description for the time-varying (TV) channels encountered in mobile communications. Many blind channel identification and equalization approaches based on the CE-BEM require precise knowledge of the basis frequencies of TV channels. Existing methods for basis frequency estimation usually resort to the higher-order statistics of channel outputs and impose strict constraints on the source signal. In this paper, we propose a novel method to estimate the basis frequencies for blind identification and equalization of time-varying single-input multiple-output (SIMO) finite-impulse-response (FIR) channels. The proposed method exploits only the second-order statistics of channel outputs and does not require strong conditions on the source signal. As a result, it exhibits superior performance to the existing basis frequency estimation methods. The validity of our method is demonstrated by numerical simulations.