Efficient Leave-<Emphasis Type="Italic">m</Emphasis>-out Cross-Validation of Support Vector Regression by Generalizing Decremental Algorithm

We propose a computationally efficient method for cross-validation of the Support Vector Regression (SVR) by generalizing the decremental algorithm of SVR. Incremental and decremental algorithm of Support Vector Machines (SVM) ^{2, 8, 9)} efficiently update the trained SVM model when a single data point is added to or removed from the training set. The computational cost of leave-one-out cross-validation can be reduced using the decremental algorithm. However, when we perform leave-m-out cross-validation (m > 1), we have to repeatedly apply the decremental algorithm for each data point. In this paper, we extend the decremental algorithm of SVR^{8, 9)} in such a way that several data points can be removed more efficiently. Experimental results indicate that the proposed approach can reduce the computational cost. In particular, we observed that the number of breakpoints, which is the main computational cost of the involved path-following, were reduced from \({\mathcal O}(m)\) to \({\mathcal O}(\sqrt{m})\).     

A dynamic default revision mechanism for speculative computation

In this work a default revision mechanism is introduced into speculative computation to manage incomplete information. The default revision is supported by a method for the generation of default constraints based on Bayesian networks. The method enables the generation of an initial set of defaults which is used to produce the most likely scenarios during the computation, represented by active processes. As facts arrive, the Bayesian network is used to derive new defaults. The objective with such a new dynamic mechanism is to keep the active processes coherent with arrived facts. This is achieved by changing the initial set of default constraints during the reasoning process in speculative computation. A practical example in clinical decision support is described.     

Object detection based on a robust and accurate statistical multi-point-pair model

In this paper, we propose a robust and accurate background model, called grayscale arranging pairs (GAP). The model is based on the statistical reach feature (SRF), which is defined as a set of statistical pair-wise features. Using the GAP model, moving objects are successfully detected under a variety of complex environmental conditions. The main concept of the proposed method is the use of multiple point pairs that exhibit a stable statistical intensity relationship as a background model. The intensity difference between pixels of the pair is much more stable than the intensity of a single pixel, especially in varying environments. Our proposed method focuses more on the history of global spatial correlations between pixels than on the history of any given pixel or local spatial correlations. Furthermore, we clarify how to reduce the GAP modeling time and present experimental results comparing GAP with existing object detection methods, demonstrating that superior object detection with higher precision and recall rates is achieved by GAP.     

Revealing microbial community structures in large- and small-scale activated sludge systems by barcoded pyrosequencing of 16S rRNA gene

The diversity of bacterial groups in activated sludge from large- and small-scale wastewater treatment plants was explored by barcoded pyrosequencing of 16S rRNA gene. Activated sludge samples (three small and 17 large scale) were collected from 12 wastewater treatment plants to clarify precise taxonomy and relative abundances. DNA was extracted, and amplified by 4 base barcoded 27f/519r primer set. The 454 Titanium (Roche) pyrosequences were obtained and analyses performed by Quantitative Insight Into Microbial Ecology (QIIME) with around 100,000 reads. Sequence statistics were computed, while constructing a phylogenetic tree and heatmap. Computed results explained total microbial diversity at phylum and class level and resolution was further extended to Operational Taxonomic Unit (OTU) based taxonomic assignment for investigating community distribution based on individual sample. Composition of sequence reads were compared and microbial community structures for large- and small-scale treatment plants were identified as major phyla (Proteobacteria and Bacteroidetes) and classes (Betaproteobacteria and Bacteroidetes). Also, family level breakdowns were explained and differences in family Nitrospiraceae and phylum Actinobacteria found at their species level were also illustrated. Thus, the pyrosequencing method provides high resolution insight into microbial community structures in activated sludge that might have been unnoticed with conventional approaches.     

Characterization of microbial community structures and their activities in single anaerobic granules by beta imaging, microsensors and fluorescence in situ hybridization

The spatial distribution of microorganisms and their in situ activities in anaerobic granules were investigated by fluorescence in situ hybridization (FISH), beta imaging and microsensors. FISH results revealed a layered structure of microorganisms in the granule, where Chloroflexi was present in the outermost layer, Smithella spp. and Syntrophobacter spp. were found in a depth of ca. 100 μm, and Archaea was restricted to the inner layer (below ca. 300 μm from the surface). Substrate uptake patterns elucidated by beta imaging demonstrated that glucose uptake was highest at 50 μm depth, whereas propionate uptake had a peak at 200 μm depth. In addition, microsensor measurements revealed that acid was produced mainly at 100 μm depth and H(2) production was detected at a depth from 100 to 200 μm. H(2) consumption and corresponding CH(4) production were found below 200 μm from the surface. Direct comparison of these results implied sequential degradation of complex organic compounds in anaerobic granules; Chloroflexi contributed to fermentation of organic compounds and acid production in the outermost layer, volatile fatty acids were oxidized and H(2) was produced mainly by Smithella spp. and Syntrophobacter spp. at a depth from 100 to 200 μm, and Archaea produced CH(4) below ca. 300 μm from the surface.     

Construction of a brain–machine hybrid system to evaluate adaptability of an insect

Insects perform adaptive behavior according to changing environmental conditions using comparatively small brains. Because adaptability is generated through the relationship among brain, body and environment, it is necessary to examine how a brain works under these conditions. In this study, to understand neural processing involved in adaptive behavior, we constructed a brain–machine hybrid system using motor signals related to the steering behavior of the male silkworm moth for controlling a two-wheeled mobile robot. We developed this hybrid system according to the following steps. (1) We selected steering signals corresponding to walking direction that were activated during neck swinging induced by optic flow and pheromone stimuli. (2) To control a robot by neural activity, we implemented a spike-behavior conversion rule such that frequency of the left and right neck motor neurons’ spikes was linearly converted into rotation of the wheels. (3) For electrophysiological multi-unit recordings on a robot, we developed small amplifiers. Using this hybrid system, we could observe the programmed behavioral pattern and orientation toward a pheromone source. Moreover, we compared the orientation behavior of moths and that of the hybrid system at different pheromone stimulus frequencies. From these experiments, we concluded that we could reconstruct silkworm moth behavior on the hybrid system.     

Biplane computed cerebral angiotomography of moyamoya disease (author's transl)

In this paper, the authors report biplane computed cerebral angiotomographic findings in 5 cases of Moyamoya disease. The specific features of Moyamoya disease on the CT image were as follows: On the axial plane, the linear structures of the anterior half of the circle of Willis and the proximal portion of the middle cerebral arteries disappeared, and instead of these normal structures, irregular, tortuous or patchy high-density areas just like a "caterpillar" were shown in the basal cistern and medial Sylvian fissures. On the modified coronal plane, the supraclinoid internal carotid arteries and the carotid fork could be identified only with a difficulty, and abnormal, "nebular" high-density areas consisting of irregular, tortuous or patchy high-density vascular components became visible in the basal cistern extending to the basal ganglia. A modified coronal plane and intravenous minimum dose bolus injection method seemed to be more useful for the visualization of these specific features on the CT image. Even before carotid angiography, we can suspect Moyamoya disease for finding these specific features on the CT image. Carotid angiography has been the only method of diagnosing Moyamoya disease. Instead of this invasive examination, computed cerebral angiotomography is useful in detecting Moyamoya disease conveniently and non-invasively. Therefore, we may conclude that computed cerebral angiotomography is very useful method for screening and follow up study of Moyamoya disease.     

Decomposition Characteristics of an Artificial Biogas in a Low‐Pressure DC Glow Discharge

The decomposition characteristics of an artificial biogas, which is a mixture of CH ₄, CO ₂, and H ₂ S, using a low‐pressure DC glow discharge have been investigated. It is found that H ₂, CO, C ₂ H ₂, H ₂ O, CS ₂, and COS are produced from the artificial biogas in the glow discharge. About 65% of the hydrogen atoms in CH ₄ are converted into H ₂ at an input energy of 800 J, at which CH ₄ is completely decomposed, and the decomposition characteristics of the artificial biogas are minimally dependent on the H ₂ S additive. Further, H ₂ S has a tendency to be decomposed earlier than the other components of the artificial biogas. When the glow discharge is generated in the artificial biogas with H ₂ S, some of the carbon atoms are found to deposit on the electrodes and the wall of the discharge chamber. © 2013 Wiley Periodicals, Inc. Electr Eng Jpn, 186(1): 26–33, 2014; Published online in Wiley Online Library ( wileyonlinelibrary.com ). DOI 10.1002/eej.22304     

‘Lipase and protease production of dairy Penicillium sp. on milk‐protein‐based solid substrates’

The lipolytic and proteolytic activity of Penicillium camemberti PC TT033 and Penicillium roqueforti PR G3, cultured on the whey solids or simulated cheese media, were compared under several pH reaction conditions. Lipolytic activity was higher when both strains had been cultured on the whey medium than on the simulated cheese medium, whereas proteolytic activity was less influenced by the culture medium. The relationship between the reaction pH and these enzyme activities was dependent on the culture medium, which suggested that the expression level and balance of isozyme rely on the culture substrate.     

Evaluation of multiple features for violent scenes detection

Multimedia Tools and Applications - Violent scenes detection (VSD) is a challenging problem because of the heterogeneous content, large variations in video quality, and complex semantic meanings of...