New kernel functions and learning methods for text and data mining

Recent advances in machine learning methods enable increasingly the automatic construction of various types of computer assisted methods that have been difficult or laborious to program by human experts. The tasks for which this kind of tools are needed arise in many areas, here especially in the fields of bioinformatics and natural language processing. The machine learning methods may not work satisfactorily if they are not appropriately tailored to the task in question. However, their learning performance can often be improved by taking advantage of deeper insight of the application domain or the learning problem at hand. This thesis considers developing kernel-based learning algorithms incorporating this kind of prior knowledge of the task in question in an advantageous way. Moreover, computationally efficient algorithms for training the learning machines for specific tasks are presented. In the context of kernel-based learning methods, the incorporation of prior knowledge is often done by designing appropriate kernel functions. Another well-known way is to develop cost functions that fit to the task under consideration. For disambiguation tasks in natural language, we develop kernel functions that take account of the positional information and the mutual similarities of words. It is shown that the use of this information significantly improves the disambiguation performance of the learning machine. Further, we design a new cost function that is better suitable for the task of information retrieval and for more general ranking problems than the cost functions designed for regression and classification. We also consider other applications of the kernel-based learning algorithms such as text categorization, and pattern recognition in differential display. We develop computationally efficient algorithms for training the considered learning machines with the proposed kernel functions. We also design a fast cross-validation algorithm for regularized least-squares type of learning algorithm. Further, an efficient version of the regularized least-squares algorithm that can be used together with the new cost function for preference learning and ranking tasks is proposed. In summary, we demonstrate that the incorporation of prior knowledge is possible and beneficial, and novel advanced kernels and cost functions can be used in algorithms efficiently.

Investigation and data adaptive analysis of CARS spectra

In this diploma work advantages of coherent anti-Stokes Raman scattering spectrometry (CARS) and various methods of the quantitative analysis of substance structure with its help are considered. The basic methods and concepts of the adaptive analysis are adduced. On the basis of these methods the algorithm of automatic measurement of a scattering strip size of a target component in CARS spectrum is developed. The algorithm uses known full spectrum of target substance and compares it with a CARS spectrum. The form of a differential spectrum is used as a feedback to control the accuracy of matching. To exclude the influence of a background in CARS spectra the differential spectrum is analysed by means of its second derivative. The algorithm is checked up on the simulated simple spectra and on the spectra of organic compounds received experimentally.

Breast Imaging Reporting and Data System - BI-RADS®: valor preditivo positivo das categorias 3, 4 e 5. revisão sistemática da literatura

OBJETIVO: Avaliar artigos, na literatura, que verificam o valor preditivo positivo das categorias 3, 4 e 5 do Breast Imaging Reporting and Data System (BI-RADS®). MATERIAIS E MÉTODOS: Foi realizada pesquisa na base de dados Medline utilizando os termos "predictive value" e "BI-RADS". Foram incluídos 11 artigos nesta revisão. RESULTADOS: O valor preditivo positivo das categorias 3, 4 e 5 variou entre 0% e 8%, 4% e 62%, 54% e 100%, respectivamente. Três artigos avaliaram, concomitantemente, os critérios morfológicos das lesões que apresentaram maior valor preditivo positivo na mamografia, sendo nódulo espiculado o critério com maior valor preditivo positivo. CONCLUSÃO: Houve grande variabilidade do valor preditivo positivo das categorias 3, 4 e 5 do BI-RADS® em todos os estudos, porém foram identificadas diferenças metodológicas que limitaram a comparação desses estudos.

Research networks and inventors’ mobility as drivers of innovation evidence from Europe

We investigate the importance of the labour mobility of inventors, as well as the scale, extent and density of their collaborative research networks, for regional innovation outcomes. To do so, we apply a knowledge production function framework at the regional level and include inventors’ networks and their labour mobility as regressors. Our empirical approach takes full account of spatial interactions by estimating a spatial lag model together, where necessary, with a spatial error model. In addition, standard errors are calculated using spatial heteroskedasticity and autocorrelation consistent estimators to ensure their robustness in the presence of spatial error autocorrelation and heteroskedasticity of unknown form. Our results point to the existence of a robust positive correlation between intraregional labour mobility and regional innovation, whilst the relationship with networks is less clear. However, networking across regions positively correlates with a region’s innovation intensity.

Valor preditivo positivo das categorias 3, 4 e 5 do Breast Imaging Reporting and Data System (BI-RADS®)

OBJETIVO: O objetivo deste trabalho foi avaliar o BI-RADS® como fator preditivo de suspeição de malignidade em lesões mamárias não palpáveis nas categorias 3, 4 e 5, correlacionando as mamografias com os resultados histopatológicos através do cálculo do valor preditivo positivo do exame mamográfico. MATERIAIS E MÉTODOS: Trezentas e setenta e uma pacientes encaminhadas a um serviço de referência em tratamento de câncer em Teresina, PI, para realização de exames histopatológicos em mama no período de julho de 2005 a março de 2008, por terem mamografia de categorias 3, 4 ou 5, tiveram seus exames revisados. Das 371 pacientes, 265 foram submetidas a biópsia por agulha grossa e 106, a marcação pré-cirúrgica. RESULTADOS: Em relação às mamografias, 11,32% foram classificadas como categoria 3, 76,28% como categoria 4 e 12,4% como categoria 5. Os resultados histológicos demonstraram 24% de exames positivos para malignidade. Os valores preditivos positivos das categorias 3, 4 e 5 foram, respectivamente, de 7,14%, 16,96% e 82,61%. Foram calculados os valores preditivos positivos, separadamente, para as biópsias percutâneas (7,14%, 15,76%, 76,47%) e para as marcações pré-cirúrgicas (7,14%, 20%, 100%). CONCLUSÃO: Achados malignos foram subestimados pelo laudo radiológico e houve superestimação de achados benignos, o que resultou na realização desnecessária de alguns procedimentos invasivos.

Positive predictive values of Breast Imaging Reporting and Data System (BI-RADS®) categories 3, 4 and 5 in breast lesions submitted to percutaneous biopsy

Objective To evaluate the BI-RADS as a predictive factor of suspicion for malignancy in breast lesions by correlating radiological with histological results and calculating the positive predictive value for categories 3, 4 and 5 in a breast cancer reference center in the city of São Paulo. Materials and Methods Retrospective, analytical and cross-sectional study including 725 patients with mammographic and/or sonographic findings classified as BI-RADS categories 3, 4 and 5 who were referred to the authors' institution to undergo percutaneous biopsy. The tests results were reviewed and the positive predictive value was calculated by means of a specific mathematical equation. Results Positive predictive values found for categories 3, 4 and 5 were respectively the following: 0.74%, 33.08% and 92.95%, for cases submitted to ultrasound-guided biopsy, and 0.00%, 14.90% and 100% for cases submitted to stereotactic biopsy. Conclusion The present study demonstrated high suspicion for malignancy in lesions classified as category 5 and low risk for category 3. As regards category 4, the need for systematic biopsies was observed.

Tools and data services registry: a community effort to document bioinformatics resources.

Life sciences are yielding huge data sets that underpin scientific discoveries fundamental to improvement in human health, agriculture and the environment. In support of these discoveries, a plethora of databases and tools are deployed, in technically complex and diverse implementations, across a spectrum of scientific disciplines. The corpus of documentation of these resources is fragmented across the Web, with much redundancy, and has lacked a common standard of information. The outcome is that scientists must often struggle to find, understand, compare and use the best resources for the task at hand.Here we present a community-driven curation effort, supported by ELIXIR-the European infrastructure for biological information-that aspires to a comprehensive and consistent registry of information about bioinformatics resources. The sustainable upkeep of this Tools and Data Services Registry is assured by a curation effort driven by and tailored to local needs, and shared amongst a network of engaged partners.As of November 2015, the registry includes 1785 resources, with depositions from 126 individual registrations including 52 institutional providers and 74 individuals. With community support, the registry can become a standard for dissemination of information about bioinformatics resources: we welcome everyone to join us in this common endeavour. The registry is freely available at https://bio.tools.

Learning 2.0 : concepts and experiences with social networks and software

Peer-reviewed

Trigger and data link system for CMS Resistive Plate Chambers at the LHC accelerator

The purpose of the work was to realize a high-speed digital data transfer system for RPC muon chambers in the CMS experiment on CERN’s new LHC accelerator. This large scale system took many years and many stages of prototyping to develop, and required the participation of tens of people. The system interfaces to Frontend Boards (FEB) at the 200,000-channel detector and to the trigger and readout electronics in the control room of the experiment. The distance between these two is about 80 metres and the speed required for the optic links was pushing the limits of available technology when the project was started. Here, as in many other aspects of the design, it was assumed that the features of readily available commercial components would develop in the course of the design work, just as they did. By choosing a high speed it was possible to multiplex the data from some the chambers into the same fibres to reduce the number of links needed. Further reduction was achieved by employing zero suppression and data compression, and a total of only 660 optical links were needed. Another requirement, which conflicted somewhat with choosing the components a late as possible was that the design needed to be radiation tolerant to an ionizing dose of 100 Gy and to a have a moderate tolerance to Single Event Effects (SEEs). This required some radiation test campaigns, and eventually led to ASICs being chosen for some of the critical parts. The system was made to be as reconfigurable as possible. The reconfiguration needs to be done from a distance as the electronics is not accessible except for some short and rare service breaks once the accelerator starts running. Therefore reconfigurable logic is extensively used, and the firmware development for the FPGAs constituted a sizable part of the work. Some special techniques needed to be used there too, to achieve the required radiation tolerance. The system has been demonstrated to work in several laboratory and beam tests, and now we are waiting to see it in action when the LHC will start running in the autumn 2008.

Research networks and scientific production in Economics: The recent spanish experience

This paper studies Spanish scientific production in Economics from 1994 to 2004. It focuses on aspects that have received little attention in other bibliometric studies, such as the impact of research and the role of scientific collaborations in the publications produced by Spanish universities. Our results show that national research networks have played a fundamental role in the increase in Spanish scientific production in this discipline.

Automatic evaluation and data generation for analytical chemistry instrumental analysis exercises

In general, laboratory activities are costly in terms of time, space, and money. As such, the ability to provide realistically simulated laboratory data that enables students to practice data analysis techniques as a complementary activity would be expected to reduce these costs while opening up very interesting possibilities. In the present work, a novel methodology is presented for design of analytical chemistry instrumental analysis exercises that can be automatically personalized for each student and the results evaluated immediately. The proposed system provides each student with a different set of experimental data generated randomly while satisfying a set of constraints, rather than using data obtained from actual laboratory work. This allows the instructor to provide students with a set of practical problems to complement their regular laboratory work along with the corresponding feedback provided by the system's automatic evaluation process. To this end, the Goodle Grading Management System (GMS), an innovative web-based educational tool for automating the collection and assessment of practical exercises for engineering and scientific courses, was developed. The proposed methodology takes full advantage of the Goodle GMS fusion code architecture. The design of a particular exercise is provided ad hoc by the instructor and requires basic Matlab knowledge. The system has been employed with satisfactory results in several university courses. To demonstrate the automatic evaluation process, three exercises are presented in detail. The first exercise involves a linear regression analysis of data and the calculation of the quality parameters of an instrumental analysis method. The second and third exercises address two different comparison tests, a comparison test of the mean and a t-paired test.