Predicting CMS datasets popularity with machine learning

In CMS è stato lanciato un progetto di Data Analytics e, all’interno di esso, un’attività specifica pilota che mira a sfruttare tecniche di Machine Learning per predire la popolarità dei dataset di CMS. Si tratta di un’osservabile molto delicata, la cui eventuale predizione premetterebbe a CMS di costruire modelli di data placement più intelligenti, ampie ottimizzazioni nell’uso dello storage a tutti i livelli Tiers, e formerebbe la base per l’introduzione di un solito sistema di data management dinamico e adattivo. Questa tesi descrive il lavoro fatto sfruttando un nuovo prototipo pilota chiamato DCAFPilot, interamente scritto in python, per affrontare questa sfida.

Big data, nosql e machine learning: un'applicazione di prediction e recommendation basata sulle api di Amazon

In questa tesi sono stati introdotti e studiati i Big Data, dando particolare importanza al mondo NoSQL, approfondendo MongoDB, e al mondo del Machine Learning, approfondendo PredictionIO. Successivamente è stata sviluppata un'applicazione attraverso l'utilizzo di tecnologie web, nodejs, node-webkit e le tecnologie approfondite prima. L'applicazione utilizza l'interpolazione polinomiale per predirre il prezzo di un bene salvato nello storico presente su MongoDB. Attraverso PredictionIO, essa analizza il comportamento degli altri utenti consigliando dei prodotti per l'acquisto. Infine è stata effetuata un'analisi dei risultati dell'errore prodotto dall'interpolazione.

A unified approach to modeling multivariate binary data using copulas over partitions

Many seemingly disparate approaches for marginal modeling have been developed in recent years. We demonstrate that many current approaches for marginal modeling of correlated binary outcomes produce likelihoods that are equivalent to the proposed copula-based models herein. These general copula models of underlying latent threshold random variables yield likelihood based models for marginal fixed effects estimation and interpretation in the analysis of correlated binary data. Moreover, we propose a nomenclature and set of model relationships that substantially elucidates the complex area of marginalized models for binary data. A diverse collection of didactic mathematical and numerical examples are given to illustrate concepts.

Tracking a non-cooperative target using a Doppler radar wireless sensor network

Tracking or target localization is used in a wide range of important tasks from knowing when your flight will arrive to ensuring your mail is received on time. Tracking provides the location of resources enabling solutions to complex logistical problems. Wireless Sensor Networks (WSN) create new opportunities when applied to tracking, such as more flexible deployment and real-time information. When radar is used as the sensing element in a tracking WSN better results can be obtained; because radar has a comparatively larger range both in distance and angle to other sensors commonly used in WSNs. This allows for less nodes deployed covering larger areas, saving money. In this report I implement a tracking WSN platform similar to what was developed by Lim, Wang, and Terzis. This consists of several sensor nodes each with a radar, a sink node connected to a host PC, and a Matlab© program to fuse sensor data. I have re-implemented their experiment with my WSN platform for tracking a non-cooperative target to verify their results and also run simulations to compare. The results of these tests are discussed and some future improvements are proposed.

The European League for Non-Western Studies and E-learning

Leiden’s Faculty of Arts invited on May 20th the members of the ELNWS for a workshop on e-learning for Non-Western studies. The aim of the workshop was to take a closer look at the possibilities of e-learning as a means of realizing co-operation between our universities. During our discussions, and looking back on the past academic years, we attempted to position our own academic community on the Internet, in the sense of 'looking for, finding and putting it on, or giving it a location' in the omnipresent Internet. In order to investigate the ways in which e-learning can be used as a tool to stimulate European co-operation within the field of Non-Western studies, Leiden’s Faculty of Arts rewards one initiative between two or more European universities of the ELNWS with financial support.

Combining Wireless Sensor Networks and Machine Learning for Flash Flood Nowcasting

This paper addresses an investigation with machine learning (ML) classification techniques to assist in the problem of flash flood now casting. We have been attempting to build a Wireless Sensor Network (WSN) to collect measurements from a river located in an urban area. The machine learning classification methods were investigated with the aim of allowing flash flood now casting, which in turn allows the WSN to give alerts to the local population. We have evaluated several types of ML taking account of the different now casting stages (i.e. Number of future time steps to forecast). We have also evaluated different data representation to be used as input of the ML techniques. The results show that different data representation can lead to results significantly better for different stages of now casting.

Facial Nerve Image Enhancement from CBCT using Supervised Learning Technique

Facial nerve segmentation plays an important role in surgical planning of cochlear implantation. Clinically available CBCT images are used for surgical planning. However, its relatively low resolution renders the identification of the facial nerve difficult. In this work, we present a supervised learning approach to enhance facial nerve image information from CBCT. A supervised learning approach based on multi-output random forest was employed to learn the mapping between CBCT and micro-CT images. Evaluation was performed qualitatively and quantitatively by using the predicted image as input for a previously published dedicated facial nerve segmentation, and cochlear implantation surgical planning software, OtoPlan. Results show the potential of the proposed approach to improve facial nerve image quality as imaged by CBCT and to leverage its segmentation using OtoPlan.

Prediction of Trabecular Bone Anisotropy from Quantitative Computed Tomography using Supervised Learning and a Novel Morphometric Feature Descriptor

Patient-specific biomechanical models including local bone mineral density and anisotropy have gained importance for assessing musculoskeletal disorders. However the trabecular bone anisotropy captured by high-resolution imaging is only available at the peripheral skeleton in clinical practice. In this work, we propose a supervised learning approach to predict trabecular bone anisotropy that builds on a novel set of pose invariant feature descriptors. The statistical relationship between trabecular bone anisotropy and feature descriptors were learned from a database of pairs of high resolution QCT and clinical QCT reconstructions. On a set of leave-one-out experiments, we compared the accuracy of the proposed approach to previous ones, and report a mean prediction error of 6% for the tensor norm, 6% for the degree of anisotropy and 19◦ for the principal tensor direction. These findings show the potential of the proposed approach to predict trabecular bone anisotropy from clinically available QCT images.

Using service-learning to encourage community health advocacy among high school students

The project outlined throughout this program management plan aims to develop a health-focused student advocacy group in the San Antonio Independent School District (SAISD). At its core, this project will be an opportunity for SAISD students to engage in service-learning, through which they will learn and develop by designing, organizing and participating in meaningful public health service experiences. ^ This program management plan addresses the genuine need for public health community education by using the service-learning model as a framework to engage students to effect change. The plan delineates the process by which the student advocacy group is to be assembled, selection of service-learning project, project objectives, technical objectives, and communication requirements. Ideally, the plan should help to facilitate project coordination, communication, and planning, and to support the direction of resources. The appendices that follow also provide useful tools with which to follow through with project implementation. ^ The plan is about more than providing a tool to educate students about the health issues in their community. It is about providing a way to teach health advocacy and self-interest and encourage civic engagement via public health. Students have the potential to positively effect lasting change among their peers, in their schools and in the community.^

Description and analysis of variation in dairy product and concomitant nutrient purchases between Hispanic and non -Hispanic households using a panel of shoppers' purchase records

The purpose of this research was development of a method of estimating nutrient availability in populations as approximated by supermarket purchase records. Demographic information describing 12,516 panel households was obtained from a marketing and advertising program operated by H. E. Butt Grocery Company of San Antonio, Texas. A non-probability sample of 2,161 households meeting expenditure criteria was selected and all purchases of dairy products for this sample of households were organized into a database constructed to facilitate the retrieval, aggregation, and analysis of dairy product purchases and their nutrient contents. Two hypotheses were tested: (1) no difference would be found between Hispanic and non-Hispanic purchases of dairy product categories during the study period and (2) no difference would be found between Hispanic and non-Hispanic purchases of nutrients contained in those dairy products during the thirteen-week study period.^ Food purchase records were used to estimate nutrient exposure on a weekly, per capita basis for Hispanic and non-Hispanic households by linking some 40,000 dairy purchase Universal Product code (UPC) numbers with food composition values contained in USDA Handbook 8-1. Results of this study suggest Hispanic sample households consistently purchased fewer dairy products than did non-Hispanic sample households and consequently had fewer nutrients available from dairy purchases. While weekly expenditures for dairy products among the sample households remained relatively constant during the study period, shifts in the types of dairy products purchased were observed. The effect of ethnicity on dairy product and nutrient purchases was significant over the thirteen-week period. A database consisting of customer, household, and purchase information can be developed to successfully associate food item UPC numbers with a standard reference of food composition to estimate nutrient availability in a population over extended periods of time. ^

Genetic diversity among alfalfa genotypes (Medicago sativa L.) of non-dormant cultivars using SSR markers and agronomic traits

The aim of this study was to assess genetic diversity among 40 alfalfa (Medicago sativa L.) genotypes of different non-dormant (FD=8) cultivars. Biomass yield, regrowth speed and reaction to spring black stem, lepto leaf spot, and rust were evaluated. Analyses of variances were performed using a mixed model to examine the agronomic variation among individuals. A principal component analysis on standardized agronomic data was performed. Agronomic data were also used to calculate Gower's distance and UPGMA algorithm. For the molecular analysis, six SSR markers were evaluated and 84 alleles were identified. The genetic distance was estimated using standard Nei's distance. Average standard genetic diversity was 0.843, indicating a high degree of variability among genotypes. Finally, a generalized procrustes analysis was performed to calculate the correlation between molecular and agronomic distance, indicating a 65.4% of consensus. This value is likely related to the low number of individuals included in the study, which might have underestimated the real phenotypic variability among genotypes. Despite the low number of individuals and SSR markers analyzed, this study provides a baseline for future diversity studies to identify genetically distant alfalfa individuals or cultivars.

Microarray analysis of autoimmune diseases by machine learning procedures

—Microarray-based global gene expression profiling, with the use of sophisticated statistical algorithms is providing new insights into the pathogenesis of autoimmune diseases. We have applied a novel statistical technique for gene selection based on machine learning approaches to analyze microarray expression data gathered from patients with systemic lupus erythematosus (SLE) and primary antiphospholipid syndrome (PAPS), two autoimmune diseases of unknown genetic origin that share many common features. The methodology included a combination of three data discretization policies, a consensus gene selection method, and a multivariate correlation measurement. A set of 150 genes was found to discriminate SLE and PAPS patients from healthy individuals. Statistical validations demonstrate the relevance of this gene set from an univariate and multivariate perspective. Moreover, functional characterization of these genes identified an interferon-regulated gene signature, consistent with previous reports. It also revealed the existence of other regulatory pathways, including those regulated by PTEN, TNF, and BCL-2, which are altered in SLE and PAPS. Remarkably, a significant number of these genes carry E2F binding motifs in their promoters, projecting a role for E2F in the regulation of autoimmunity.