101,994 research outputs found
A cluster computing approach applied to machine learning for earth observation big data analysis
No full textMachine learning and Deep Learning techniques are becoming increasingly popular as a promising approach for analysis of large business and scientific data sets for developing more sophisticated models for different intelligent applications such as image recognition, speech categorization, and automatic machine translation. Achieving high accurate data models requires massive amount of input training data, however most of the current machine learning tools still try to scale up both storage and processing on a single machine, to be able to process such large amount of data, along computation intensive models. The execution time of training phase of an accurate and efficient model could become a bottleneck when the data size scales, since the enormous computation and data input are still a huge burden to a single machine. Research scientists need to setup the network with initial configurations and wait for a long time to get back the trained model, which reduces the performance in the model training process. In order to solve this problem, there are frameworks and solutions that are utilizing accelerators such as GPU [18] and FPGA[19], for doing the computation part and offload certain operations onto GPU cores. Unfortunately, a single system can scale-up to some extents, which makes some limitation for the feasibility of current frameworks. On the other hand, there are frameworks as Spark [20] that distributes computing tasks across the nodes of a cluster. To find a possible solutions for reducing the execution time of the training phase of an accurate and efficient model in machine learning environments, this work proposes a distributed machine-learning framework based on Apache Spark and a cluster of distributed CPUs (no GPUs) to build implementations of Neural Network computational models to demonstrate that this scalable architecture provides a training speedup that increases uniformly with increase in the number of worker nodes. The key contributions of this approach are: • Leveraging distributed feature in Spark, this library can scale out both dataset and computation, by distributing both data and operations across the cluster nodes. This library targets a Multilayer Perceptron Neural Network computational model for an automatic pixel-based image classification aimed to cloud detection from iii Word Template by Friedman & Morgan 2014 LANDASAT 8 Satellite images and a Volcanic ash mass retrieval using MODIS data. The generation of training data, the training of the network, and the classification of an image are implemented by applying parallel programming primitives to multidimensional arrays of data, which are distributed across a computer cluster nodes. • In addition to a complete description of the Multilayer Perceptron Neural Network model implementations, this work also provides a computational benchmark comparing different machine learning frameworks (Spark MLlib, Tensorflow and BigDL) and evaluate the Spark cluster scalability using the speed up metric. The results show that the Spark-based system obtains near-linear scalability in its distributed configuration for the tested dataset: by adding two or three more nodes into the system it is possible to reduce the running time up to 2-3 times of the original stand-alone single node configuration. • Parameter synchronization is a performance critical operation for data parallel distributed model training (in terms of speed and scalability). BigDL, a distributed machine learning framework for big data platforms and workflows, has been used as a library on top of Apache Spark, because it has been demonstrated that, unlike existing machine learning frameworks, it can efficiently train large machine neural network across large (e.g., hundreds of servers) clusters. BigDL allows users to build machine learning applications for Earth Observation Big Data using a single unified data pipeline; the entire pipeline can directly run on top of existing big data systems as Apache Spark in a distributed fashion. • The results have demonstrated that the use of a unified data analytics and machine learning system like BigDL over Spark improve the ease of use (including development, deployment and operations) and enhances the performance of machine learning process in the Earth Observation big data context, where for achieving high accurate data models it is required massive amount of input training data
Pseudotemporal ordering of spatial lymphoid tissue microenvironment profiles trails Unclassified DLBCL at the periphery of the follicle
No full text: We have established a pseudotemporal ordering for the transcriptional signatures of distinct microregions within reactive lymphoid tissues, namely germinal center dark zones (DZ), germinal center light zones (LZ), and peri-follicular areas (Peri). By utilizing this pseudotime trajectory derived from the functional microenvironments of DZ, LZ, and Peri, we have ordered the transcriptomes of Diffuse Large B-cell Lymphoma cases. The apex of the resulting pseudotemporal trajectory, which is characterized by enrichment of molecular programs fronted by TNFR signaling and inhibitory immune checkpoint overexpression, intercepts a discrete peri-follicular biology. This observation is associated with DLBCL cases that are enriched in the Unclassified/type-3 COO category, raising questions about the potential extra-GC microenvironment imprint of this peculiar group of cases. This report offers a thought-provoking perspective on the relationship between transcriptional profiling of functional lymphoid tissue microenvironments and the evolving concept of the cell of origin in Diffuse Large B-cell Lymphomas
On the potential of Big Data capabilities for the validation of a weather forecasting system
No full textA key component of the EO projects is the validation of the EO data products through a Ground Truth Validation. In the validation process data can be collected from various ground-based sources and sensors (in situ measurements, instruments, crowd-sourcing, open source platform), then quality-controlled, and finally compared with the satellite products in order to get validated retrievals. The objective of this work is to develop a system that uses big data capabilities and tools for validation purposes, in particular for the assessment of a new weather nowcasting system, based on a predictive model exploiting Meteosat Second Generation (MSG) imagery
COAGULASE POSITIVE STAPHYLOCOCCI ENUMERATION AND ENTEROTOXINS DETECTION IN MILK AND DAIRY PRODUCTS FROM CENTRAL ITALY
No full textThis study aims at enumerating coagulase positive staphylococci (CPS) in 404 samples of milk and dairy products collected in own-checks or during the official controls from different dairy industries located in Central Italy. These microorganisms were enumerated using ISO 6888-2:1999/Amd. 1:2003 and only when they exceeded 10(b) CFU/g, the presence of any of the seven more common staphylococcal enterotoxins (SEA, SEB, SEC1, SEC2, SEC3, SED and SEE) was also investigated. Own-checks samples resulted always below the detection limit, whereas among those collected by the competent authorities in the framework of official controls, provola (100%) and mozzarella (22.9%) samples were positive to CPS, with mean values of 1.8x10(2) and 2.8x10(5) CFU/g respectively. Such values exceeded the maximum limits set by Commission Regulation (EC) No. 2073/2005, resulting in a request of hygiene improvements in the first case; in the second case, the presence of staphylococcal enterotoxins in 8 (2.7%) mozzarella samples out of 298 investigated cheese products was also observed, resulting in their withdrawn from the market. Therefore, this study aims at highlighting that monitoring of CPS incidence in dairy products and subsequent testing of cheeses for enterotoxins when appropriate represent an important tool for public health in order to avoid the occurrence of foodborne outbreaks
Nuove prospettive nel Managment dei pazienti con osteonecrosi dei mascellari da bifosfonati
No full textBleomycin, etoposide, doxorubicin, cyclophosphamide, vincristine, procarbazine, and prednisone outside German Hodgkin Study Group: The Italian experience
No full textSplenic marginal zone lymphoma with or without villous lymphocytes
No full textThe Integruppo Italiano Linfomi (IIL) carried out a study to assess the outcomes of splenic marginal zone lymphoma and to identify prognostic factors in 309 patients. The 5-year cause-specific survival (CSS) rate was 76%. In univariate analysis, the parameters predictive of shorter CSS were hemoglobin levels below 12 g/dL (P < .001), albumin levels below 3.5 g/dL (P = .001), International Prognostic Index (IPI) scores of 2 to 3 (P < .001), lactate dehydrogenase (LDH) levels above normal (P < .001), age older than 60 years (P = .01), platelet counts below 100,000/microL (P = .04), HbsAg-positivity (P = .01), and no splenectomy at diagnosis (P = .006). Values that maintained a negative influence on CSS in multivariate analysis were hemoglobin level less than 12 g/dL, LDH level greater than normal, and albumin level less than 3.5 g/dL. Using these 3 variables, we grouped patients into 3 prognostic categories: low-risk group (41%) with no adverse factors, intermediate-risk group (34%) with one adverse factor, and high-risk group (25%) with 2 or 3 adverse factors. The 5-year CSS rate was 88% for the low-risk group, 73% for the intermediate-risk group, and 50% for the high-risk group. The cause-specific mortality rate (x 1000 person-years) was 20 for the low-risk group, 47 for the intermediate-risk group, and 174 for the high-risk group. This latter group accounted for 54% of all lymphoma-related deaths. In conclusion, with the use of readily available factors, this prognostic index may be an effective tool for evaluating the need for treatment and the intensity of therapy in an individual patient
LA PRESA IN CARICO INTEGRATA DELL’ ADOLESCENTE CON PATOLOGIA CRONICA: PROGETTO PILOTA PER UN AMBULATORIO DI ADOLESCENTOLOGIA
No full textBibliographie Hilarion G. Petzold 1958 – 2009 mit Anhang als Einführung
No full textDieses Archiv enthält die Gesamtbibliographie der Werke des Autors nebst einiger Texte „Über H. G. Petzold“ im Schlussteil der Bibliographie sowie einen Anhang mit einer Einführung in die Architektur des Werkes in seinem wissenslogischen Aufbau als Ausarbeitung seines „Tree of Science Modells“ (2007).This archive contains the complete bibliography of the author and some texts about H. G. Petzold, moreover an epilogue with an introduction to the architecture of the works in its epistemological structure and composition and as an elaborations of Petzold’s „Tree of Science Modell (2007).https://www.fpi-publikation.de/polyloge/01-2009-petzold-h-g-gesamtbibliographie-h-g-petzold-1958-2009-updating-november2009/peerReviewedpublishedVersio
Dispelling the Myths Behind First-author Citation Counts
No full textWe conducted a full-scale evaluative citation analysis study of scholars in the XML research field to explore just how different from each other author rankings resulting from different citation counting methods actually are, and to demonstrate the capability of emerging data and tools on the Web in supporting more realistic citation counting methods. Our results contest some common arguments for the continued
use of first-author citation counts in the evaluation of scholars, such as high correlations between author rankings by first-author citation counts and other citation
counting methods, and high costs of using more realistic citation counting methods that are not well-supported by the ISI databases. It is argued that increasingly available digital full text research papers make it possible for citation analysis studies to go beyond what the ISI databases have directly supported and to employ more
sophisticated methods
- …
