Machine Learning Algorithms in Heavy Process Manufacturing

In a global economy, manufacturers mainly compete with cost efficiency of production, as the price of raw materials are similar worldwide. Heavy industry has two big issues to deal with. On the one hand there is lots of data which needs to be analyzed in an effective manner, and on the other hand making big improvements via investments in cooperate structure or new machinery is neither economically nor physically viable. Machine learning offers a promising way for manufacturers to address both these problems as they are in an excellent position to employ learning techniques with their massive resource of historical production data. However, choosing modelling a strategy in this setting is far from trivial and this is the objective of this article. The article investigates characteristics of the most popular classifiers used in industry today. Support Vector Machines, Multilayer Perceptron, Decision Trees, Random Forests, and the meta-algorithms Bagging and Boosting are mainly investigated in this work. Lessons from real-world implementations of these learners are also provided together with future directions when different learners are expected to perform well. The importance of feature selection and relevant selection methods in an industrial setting are further investigated. Performance metrics have also been discussed for the sake of completion.

Faceparts for recognition

A facial classification system that utilises images of faceparts is presented in this paper. Each facepart region is allocated a degree of importance. The random forests approach is employed for classification. The approach grows many classification trees where each tree gives a classification decision. The forest selects the classification that gives the most votes. Experimental results are presented and discussed

Lung nodules detection by ensemble classification

A method is presented that achieves lung nodule detection by classification of nodule and non-nodule patterns. It is based on random forests which are ensemble learners that grow classification trees. Each tree produces a classification decision, and an integrated output is calculated. The performance of the developed method is compared against that of the support vector machine and the decision tree methods. Three experiments are performed using lung scans of 32 patients including thousands of images within which nodule locations are marked by expert radiologists. The classification errors and execution times are presented and discussed. The lowest classification error (2.4%) has been produced by the developed method.

Hybrid classification of pulmonary nodules

Automated classification of lung nodules is challenging because of the variation in shape and size of lung nodules, as well as their associated differences in their images. Ensemble based learners have demonstrated the potentialof good performance. Random forests are employed for pulmonary nodule classification where each tree in the forest produces a classification decision, and an integrated output is calculated. A classification aided by clustering approach is proposed to improve the lung nodule classification performance. Three experiments are performed using the LIDC lung image database of 32 cases. The classification performance and execution times are presented and discussed.

Pulmonary nodule classification aided by clustering

Lung nodules can be detected through examining CT scans. An automated lung nodule classification system is presented in this paper. The system employs random forests as it base classifier. A unique architecture for classification-aided-by-clustering is presented. Four experiments are conducted to study the performance of the developed system. 5721 CT lung image slices from the LIDC database are employed in the experiments. According to the experimental results, the highest sensitivity of 97.92%, and specificty of 96.28% are achieved by the system. The results demonstrate that the system has improved the performances of its tested counterparts.

Multilabel classification using error correction codes

This paper presents a multilabel classification method that employs an error correction code together with a base ensemble learner to deal with multilabel data. It explores two different error correction codes: convolutional code and BCH code. A random forest learner is used as its based learner. The performance of the proposed method is evaluated experimentally. The popular multilabel yeast dataset is used for benchmarking. The results are compared against those of several exiting approaches. The proposed method performs well against its counterparts.

Integrating multibeam backscatter angular response, mosaic and bathymetry data for benthic habitat mapping

Multibeam echosounders (MBES) are increasingly becoming the tool of choice for marine habitat mapping applications. In turn, the rapid expansion of habitat mapping studies has resulted in a need for automated classification techniques to efficiently map benthic habitats, assess confidence in model outputs, and evaluate the importance of variables driving the patterns observed. The benthic habitat characterisation process often involves the analysis of MBES bathymetry, backscatter mosaic or angular response with observation data providing ground truth. However, studies that make use of the full range of MBES outputs within a single classification process are limited. We present an approach that integrates backscatter angular response with MBES bathymetry, backscatter mosaic and their derivatives in a classification process using a Random Forests (RF) machine-learning algorithm to predict the distribution of benthic biological habitats. This approach includes a method of deriving statistical features from backscatter angular response curves created from MBES data collated within homogeneous regions of a backscatter mosaic. Using the RF algorithm we assess the relative importance of each variable in order to optimise the classification process and simplify models applied. The results showed that the inclusion of the angular response features in the classification process improved the accuracy of the final habitat maps from 88.5% to 93.6%. The RF algorithm identified bathymetry and the angular response mean as the two most important predictors. However, the highest classification rates were only obtained after incorporating additional features derived from bathymetry and the backscatter mosaic. The angular response features were found to be more important to the classification process compared to the backscatter mosaic features. This analysis indicates that integrating angular response information with bathymetry and the backscatter mosaic, along with their derivatives, constitutes an important improvement for studying the distribution of benthic habitats, which is necessary for effective marine spatial planning and resource management.

Wave exposure as a predictor of benthic habitat distribution on high energy temperate reefs

The new found ability to measure physical attributes of the marine environment at high resolution across broad spatial scales has driven the rapid evolution of benthic habitat mapping as a field in its own right. Improvement of the resolution and ecological validity of seafloor habitat distribution models has, for the most part, paralleled developments in new generations of acoustic survey tools such as multibeam echosounders. While sonar methods have been well demonstrated to provide useful proxies of the relatively static geophysical patterns that reflect distribution of benthic species and assemblages, the spatially and temporally variable influence of hydrodynamic energy on habitat distribution have been less well studied. Here we investigate the role of wave exposure on patterns of distribution of near-shore benthic habitats. A high resolution spectral wave model was developed for a 624 km2 site along Cape Otway, a major coastal feature of western Victoria, Australia. Comparison of habitat classifications implemented using the Random Forests algorithm established that significantly more accurate estimations of habitat distribution were obtained by including a fine-scale numerical wave model, extended to the seabed using linear wave theory, than by using depth and seafloor morphology information alone. Variable importance measures and map interpretation indicated that the spatial variation in wave-induced bottom orbital velocity was most influential in discriminating habitat classes containing the canopy forming kelp Ecklonia radiata, a foundation kelp species that affects biodiversity and ecological functioning on shallow reefs across temperate Australasia. We demonstrate that hydrodynamic models reflecting key environmental drivers on wave-exposed coastlines are important in accurately defining distributions of benthic habitats. This study highlights the suitability of exposure measures for predictive habitat modeling on wave-exposed coastlines and provides a basis for continuing work relating patterns of biological distribution to remotely-sensed patterns of the physical environment.

Avaliação de compostos com actividade inibidora da glicoproteína-P

A resistência a múltiplos fármacos é um grande problema na terapia anti-cancerígena, sendo a glicoproteína-P (P-gp) uma das responsáveis por esta resistência. A realização deste trabalho incidiu principalmente no desenvolvimento de modelos matemáticos/estatísticos e “químicos”. Para os modelos matemáticos/estatísticos utilizamos métodos de Machine Learning como o Support Vector Machine (SVM) e o Random Forest, (RF) em relação aos modelos químicos utilizou-se farmacóforos. Os métodos acima mencionados foram aplicados a diversas proteínas P-gp, p53 e complexo p53-MDM2, utilizando duas famílias: as pifitrinas para a p53 e flavonóides para P-gp e, em menor medida, um grupo diversificado de moléculas de diversas famílias químicas. Nos modelos obtidos pelo SVM quando aplicados à P-gp e à família dos flavonóides, obtivemos bons valores através do kernel Radial Basis Function (RBF), com precisão de conjunto de treino de 94% e especificidade de 96%. Quanto ao conjunto de teste com previsão de 70% e especificidade de 67%, sendo que o número de falsos negativos foi o mais baixo comparativamente aos restantes kernels. Aplicando o RF à família dos flavonóides verificou-se que o conjunto de treino apresenta 86% de precisão e uma especificidade de 90%, quanto ao conjunto de teste obtivemos uma previsão de 70% e uma especificidade de 60%, existindo a particularidade de o número de falsos negativos ser o mais baixo. Repetindo o procedimento anterior (RF) e utilizando um total de 63 descritores, os resultados apresentaram valores inferiores obtendo-se para o conjunto de treino 79% de precisão e 82% de especificidade. Aplicando o modelo ao conjunto de teste obteve-se 70% de previsão e 60% de especificidade. Comparando os dois métodos, escolhemos o método SVM com o kernel RBF como modelo que nos garante os melhores resultados de classificação. Aplicamos o método SVM à P-gp e a um conjunto de moléculas não flavonóides que são transportados pela P-gp, obteve-se bons valores através do kernel RBF, com precisão de conjunto de treino de 95% e especificidade de 93%. Quanto ao conjunto de teste, obtivemos uma previsão de 70% e uma especificidade de 69%, existindo a particularidade de o número de falsos negativos ser o mais baixo. Aplicou-se o método do farmacóforo a três alvos, sendo estes, um conjunto de inibidores flavonóides e de substratos não flavonóides para a P-gp, um grupo de piftrinas para a p53 e um conjunto diversificado de estruturas para a ligação da p53-MDM2. Em cada um dos quatro modelos de farmacóforos obtidos identificou-se três características, sendo que as características referentes ao anel aromático e ao dador de ligações de hidrogénio estão presentes em todos os modelos obtidos. Realizando o rastreio em diversas bases de dados utilizando os modelos, obtivemos hits com uma grande diversidade estrutural.

Combining ALOS/PALSAR derived vegetation structure and inundation patterns to characterize major vegetation types in the Mamiraua Sustainable Development Reserve, Central Amazon floodplain, Brazil

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Aberrant lipid metabolism in hepatocellular carcinoma revealed by plasma metabolomics and lipid profiling

There has been limited analysis of the effects of hepatocellular carcinoma (HCC) on liver metabolism and circulating endogenous metabolites. Here, we report the findings of a plasma metabolomic investigation of HCC patients by ultraperformance liquid chromatography-electrospray ionization-quadrupole time-of-flight mass spectrometry (UPLC-ESI-QTOFMS), random forests machine learning algorithm, and multivariate data analysis. Control subjects included healthy individuals as well as patients with liver cirrhosis or acute myeloid leukemia. We found that HCC was associated with increased plasma levels of glycodeoxycholate, deoxycholate 3-sulfate, and bilirubin. Accurate mass measurement also indicated upregulation of biliverdin and the fetal bile acids 7α-hydroxy-3-oxochol-4-en-24-oic acid and 3-oxochol-4,6-dien-24-oic acid in HCC patients. A quantitative lipid profiling of patient plasma was also conducted by ultraperformance liquid chromatography-electrospray ionization-triple quadrupole mass spectrometry (UPLC-ESI-TQMS). By this method, we found that HCC was also associated with reduced levels of lysophosphocholines and in 4 of 20 patients with increased levels of lysophosphatidic acid [LPA(16:0)], where it correlated with plasma α-fetoprotein levels. Interestingly, when fatty acids were quantitatively profiled by gas chromatography-mass spectrometry (GC-MS), we found that lignoceric acid (24:0) and nervonic acid (24:1) were virtually absent from HCC plasma. Overall, this investigation illustrates the power of the new discovery technologies represented in the UPLC-ESI-QTOFMS platform combined with the targeted, quantitative platforms of UPLC-ESI-TQMS and GC-MS for conducting metabolomic investigations that can engender new insights into cancer pathobiology.

Metabolomics reveals attenuation of the SLC6A20 kidney transporter in nonhuman primate and mouse models of type 2 diabetes mellitus

To enhance understanding of the metabolic indicators of type 2 diabetes mellitus (T2DM) disease pathogenesis and progression, the urinary metabolomes of well characterized rhesus macaques (normal or spontaneously and naturally diabetic) were examined. High-resolution ultra-performance liquid chromatography coupled with the accurate mass determination of time-of-flight mass spectrometry was used to analyze spot urine samples from normal (n = 10) and T2DM (n = 11) male monkeys. The machine-learning algorithm random forests classified urine samples as either from normal or T2DM monkeys. The metabolites important for developing the classifier were further examined for their biological significance. Random forests models had a misclassification error of less than 5%. Metabolites were identified based on accurate masses (<10 ppm) and confirmed by tandem mass spectrometry of authentic compounds. Urinary compounds significantly increased (p < 0.05) in the T2DM when compared with the normal group included glycine betaine (9-fold), citric acid (2.8-fold), kynurenic acid (1.8-fold), glucose (68-fold), and pipecolic acid (6.5-fold). When compared with the conventional definition of T2DM, the metabolites were also useful in defining the T2DM condition, and the urinary elevations in glycine betaine and pipecolic acid (as well as proline) indicated defective re-absorption in the kidney proximal tubules by SLC6A20, a Na(+)-dependent transporter. The mRNA levels of SLC6A20 were significantly reduced in the kidneys of monkeys with T2DM. These observations were validated in the db/db mouse model of T2DM. This study provides convincing evidence of the power of metabolomics for identifying functional changes at many levels in the omics pipeline.

Application and comparison of classification algorithms for recognition of Alzheimer's disease in electrical brain activity (EEG)

The early detection of subjects with probable Alzheimer's disease (AD) is crucial for effective appliance of treatment strategies. Here we explored the ability of a multitude of linear and non-linear classification algorithms to discriminate between the electroencephalograms (EEGs) of patients with varying degree of AD and their age-matched control subjects. Absolute and relative spectral power, distribution of spectral power, and measures of spatial synchronization were calculated from recordings of resting eyes-closed continuous EEGs of 45 healthy controls, 116 patients with mild AD and 81 patients with moderate AD, recruited in two different centers (Stockholm, New York). The applied classification algorithms were: principal component linear discriminant analysis (PC LDA), partial least squares LDA (PLS LDA), principal component logistic regression (PC LR), partial least squares logistic regression (PLS LR), bagging, random forest, support vector machines (SVM) and feed-forward neural network. Based on 10-fold cross-validation runs it could be demonstrated that even tough modern computer-intensive classification algorithms such as random forests, SVM and neural networks show a slight superiority, more classical classification algorithms performed nearly equally well. Using random forests classification a considerable sensitivity of up to 85% and a specificity of 78%, respectively for the test of even only mild AD patients has been reached, whereas for the comparison of moderate AD vs. controls, using SVM and neural networks, values of 89% and 88% for sensitivity and specificity were achieved. Such a remarkable performance proves the value of these classification algorithms for clinical diagnostics.

Geospatial analysis of western juniper (Juniperus occidentalis) encroachment utilizing remotely sensed data

Utilizing remote sensing methods to assess landscape-scale ecological change are rapidly becoming a dominant force in the natural sciences. Powerful and robust non-parametric statistical methods are also actively being developed to compliment the unique characteristics of remotely sensed data. The focus of this research is to utilize these powerful, robust remote sensing and statistical approaches to shed light on woody plant encroachment into native grasslands--a troubling ecological phenomenon occurring throughout the world. Specifically, this research investigates western juniper encroachment within the sage-steppe ecosystem of the western USA. Western juniper trees are native to the intermountain west and are ecologically important by means of providing structural diversity and habitat for many species. However, after nearly 150 years of post-European settlement changes to this threatened ecosystem, natural ecological processes such as fire regimes no longer limit the range of western juniper to rocky refugia and other areas protected from short fire return intervals that are historically common to the region. Consequently, sage-steppe communities with high juniper densities exhibit negative impacts, such as reduced structural diversity, degraded wildlife habitat and ultimately the loss of biodiversity. Much of today's sage-steppe ecosystem is transitioning to juniper woodlands. Additionally, the majority of western juniper woodlands have not reached their full potential in both range and density. The first section of this research investigates the biophysical drivers responsible for juniper expansion patterns observed in the sage-steppe ecosystem. The second section is a comprehensive accuracy assessment of classification methods used to identify juniper tree cover from multispectral 1 m spatial resolution aerial imagery.

Human urinary metabolomic profile of PPARalpha induced fatty acid beta-oxidation

Activation of the peroxisome proliferator-activated receptor alpha (PPARalpha) is associated with increased fatty acid catabolism and is commonly targeted for the treatment of hyperlipidemia. To identify latent, endogenous biomarkers of PPARalpha activation and hence increased fatty acid beta-oxidation, healthy human volunteers were given fenofibrate orally for 2 weeks and their urine was profiled by UPLC-QTOFMS. Biomarkers identified by the machine learning algorithm random forests included significant depletion by day 14 of both pantothenic acid (>5-fold) and acetylcarnitine (>20-fold), observations that are consistent with known targets of PPARalpha including pantothenate kinase and genes encoding proteins involved in the transport and synthesis of acylcarnitines. It was also concluded that serum cholesterol (-12.7%), triglycerides (-25.6%), uric acid (-34.7%), together with urinary propylcarnitine (>10-fold), isobutyrylcarnitine (>2.5-fold), (S)-(+)-2-methylbutyrylcarnitine (5-fold), and isovalerylcarnitine (>5-fold) were all reduced by day 14. Specificity of these biomarkers as indicators of PPARalpha activation was demonstrated using the Ppara-null mouse. Urinary pantothenic acid and acylcarnitines may prove useful indicators of PPARalpha-induced fatty acid beta-oxidation in humans. This study illustrates the utility of a pharmacometabolomic approach to understand drug effects on lipid metabolism in both human populations and in inbred mouse models.