Supervised Aggregative Feature Extraction for Big Data Time Series Regression

In many applications, and especially those where batch processes are involved, a target scalar output of interest is often dependent on one or more time series of data. With the exponential growth in data logging in modern industries such time series are increasingly available for statistical modeling in soft sensing applications. In order to exploit time series data for predictive modelling, it is necessary to summarise the information they contain as a set of features to use as model regressors. Typically this is done in an unsupervised fashion using simple techniques such as computing statistical moments, principal components or wavelet decompositions, often leading to significant information loss and hence suboptimal predictive models. In this paper, a functional learning paradigm is exploited in a supervised fashion to derive continuous, smooth estimates of time series data (yielding aggregated local information), while simultaneously estimating a continuous shape function yielding optimal predictions. The proposed Supervised Aggregative Feature Extraction (SAFE) methodology can be extended to support nonlinear predictive models by embedding the functional learning framework in a Reproducing Kernel Hilbert Spaces setting. SAFE has a number of attractive features including closed form solution and the ability to explicitly incorporate first and second order derivative information. Using simulation studies and a practical semiconductor manufacturing case study we highlight the strengths of the new methodology with respect to standard unsupervised feature extraction approaches.

Evaluation of the ionic liquids 1-alkyl-3-methylimidazolium hexafluorophosphate as a solvent for the extraction of benzene from cyclohexane : ( Liquid + liquid ) equilibria

In the catalytic hydrogenation of benzene to cyclohexane, the separation of unreacted benzene from the product stream is inevitable and essential for an economically viable process. In order to evaluate the separation efficiency of ionic liquids (ILs) as a solvent in this extraction processes, the ternary (liquid + liquid) equilibrium of 1-alkyl-3-methylimidazolium hexafluorophosphate, [Cnmim][PF6] (n = 4, 5, 6), with benzene and cyclohexane was studied at T = 298.15 K and atmospheric pressure. The reliability of the experimentally determined tie-line data was confirmed by applying the Othmer–Tobias equation. The solute distribution coefficient and solvent selectivity for the systems studied were calculated and compared with literature data for other ILs and sulfolane. It turns out that the benzene distribution coefficient increases and solvent selectivity decreases as the length of the cation alkyl chain grows, and the ionic liquids [Cnmim][PF6] proved to be promising solvents for benzene–cyclohexane extractive separation. Finally, an NRTL model was applied to correlate and fit the experimental LLE data for the ternary systems studied.

Deep separation of benzene from cyclohexane by liquid extraction using ionic liquids as the solvent

Separation of benzene and cyclohexane is one of the most important and difficult processes in the petrochemical industry, especially for low benzene concentration. In this work, three ionic liquids (ILs), [Bmim][BF ₄], [Bpy][BF ₄], and [Bmim][SCN], were investigated as the solvent in the extraction of benzene from cyclohexane. The corresponding ternary liquid-liquid equilibria (LLE) were experimentally determined at T = 298.15 K and atmospheric pressure. The LLE data were correlated with the nonrandom two-liquid model, and the parameters were fitted. The separation capabilities of the ILs were evaluated in terms of the benzene distribution coefficient and solvent selectivity. The effect of the IL structure on the separation was explained based on a well-founded physical model, COSMO-RS. Finally, the extraction processes were defined, and the operation parameters were analyzed. It shows that the ILs studied are suitable solvents for the extractive separation of benzene and cyclohexane, and their separation efficiency can be generally ranked as [Bmim][BF ₄] > [Bpy][BF ₄] > [Bmim][SCN]. The extraction process for a feed with 15 mol % benzene was optimized. High product purity (cyclohexane 0.997) and high recovery efficiency (cyclohexane 96.9% and benzene 98.1%) can be reached. © 2012 American Chemical Society.

Improving Recall of Regular Expressions for Information Extraction

Learning or writing regular expressions to identify instances of a specific
concept within text documents with a high precision and recall is challenging.
It is relatively easy to improve the precision of an initial regular expression
by identifying false positives covered and tweaking the expression to avoid the
false positives. However, modifying the expression to improve recall is difficult
since false negatives can only be identified by manually analyzing all documents,
in the absence of any tools to identify the missing instances. We focus on partially
automating the discovery of missing instances by soliciting minimal user
feedback. We present a technique to identify good generalizations of a regular
expression that have improved recall while retaining high precision. We empirically
demonstrate the effectiveness of the proposed technique as compared to
existing methods and show results for a variety of tasks such as identification of
dates, phone numbers, product names, and course numbers on real world datasets

Query Suggestions for Textual Problem Solution Repositories

Textual problem-solution repositories are available today in
various forms, most commonly as problem-solution pairs from community
question answering systems. Modern search engines that operate on
the web can suggest possible completions in real-time for users as they
type in queries. We study the problem of generating intelligent query
suggestions for users of customized search systems that enable querying
over problem-solution repositories. Due to the small scale and specialized
nature of such systems, we often do not have the luxury of depending on
query logs for finding query suggestions. We propose a retrieval model
for generating query suggestions for search on a set of problem solution
pairs. We harness the problem solution partition inherent in such
repositories to improve upon traditional query suggestion mechanisms
designed for systems that search over general textual corpora. We evaluate
our technique over real problem-solution datasets and illustrate that
our technique provides large and statistically significant

Method for rapid extraction of pectic substances from plant materials

A method is described for the rapid extraction of pectic substances from alcohol insoluble solids (AIS) from material of plant origin, especially fruit. Samples of AIS can be prepared for galacturonic acid assay within 60 min using extraction with 0·5m HCl in a Fibertec-1 system (Tecator) for 30 min. The extraction conditions are carefully standardised and operator error is reduced by the elimination of transfer steps, particularly during filtration. The results obtained for plant-derived alcohol insoluble solids containing from 10% to 33% pectic substances were in close agreement with those obtained by enzymic hydrolysis using a commercially available enzyme preparation (Ultrazyme). The method will have application in the rapid, routine estimation of pectic substances in plant materials. © 1987.

Visual outcomes and astigmatism after sutureless, manual cataract extraction in rural China: study of cataract outcomes and up-take of services (SCOUTS) in the caring is hip project, report 1.

OBJECTIVE: To study the visual acuity and astigmatism of persons undergoing cataract extraction by local surgeons in rural China. METHODS: Visual acuity, keratometry, and refraction were measured 10 to 14 months postoperatively for all cataract cases during 4 months in Sanrao, China. RESULTS: Among 313 eligible subjects, 242 (77%) could be contacted, of whom 176 (73%) were examined. Of those who were examined, mean +/- SD age was 69.3 +/- 10.5 years, 66.5% were female, 35 had been operated on bilaterally at Sanrao, and 85.2% had a preoperative presenting visual acuity of 6/60 or worse. Presenting and best-corrected postoperative acuity in the eye that was operated on were 6/18 or better in 83.4% and 95.7%, respectively. Among 27 fellow eyes operated on elsewhere, 40.7% had a presenting acuity of 6/18 or better and 40.7% were blind (P < .001). Mean +/- SD postoperative astigmatism did not differ between 211 eyes that were operated on (-1.13 +/- 0.84 diopters) and 109 eyes that were not (-1.13 +/- 1.17 diopters; P = .27). Presence of operative complications (8.5%) and older age were associated with worse vision; bilateral surgery was associated with better vision. CONCLUSIONS: These results confirm the effectiveness of skill transfer in this setting, with superior outcomes to most studies in rural Asia and to eyes in this cohort operated on at other facilities.

Endothelial cell loss and surgically induced astigmatism after sutureless large-incision manual cataract extraction (SLIMCE).

OBJECTIVES:

To describe a modified manual cataract extraction technique, sutureless large-incision manual cataract extraction (SLIMCE), and to report its clinical outcomes.

METHODS:

Case notes of 50 consecutive patients with cataract surgery performed using the SLIMCE technique were retrospectively reviewed. Clinical outcomes 3 months after surgery were analyzed, including postoperative uncorrected visual acuity, best-corrected visual acuity, intraoperative and postoperative complications, endothelial cell loss, and surgically induced astigmatism using the vector analysis method.

RESULTS:

At the 3-month follow-up, all 50 patients had postoperative best-corrected visual acuity of at least 20/60, and 37 patients (74%) had visual acuity of at least 20/30. Uncorrected visual acuity was at least 20/68 in 28 patients (56%) and was between 20/80 and 20/200 in 22 patients (44%). No significant intraoperative complications were encountered, and sutureless wounds were achieved in all but 2 patients. At the 3-month follow-up, endothelial cell loss was 3.9%, and the mean surgically induced astigmatism was 0.69 diopter.

CONCLUSIONS:

SLIMCE is a safe and effective manual cataract extraction technique with low rates of surgically induced astigmatism and endothelial cell loss. In view of its low cost, SLIMCE may have a potential role in reducing cataract blindness in developing countries.

The challenge of MNCs and development: oil extraction, CSR, Nigeria and corruption

Globalisation has led to a shift in world order with the rise of the corporate non-state actor. This rise has led to an assumption that Multi-National Corporations (MNCs) must assume responsibilities beyond profit maximisation for shareholders. With the rise of the MNC as a corporate non-state actor there have been discussions around the role of business with regard to human rights.

This article looks at the case of oil extraction in Nigeria. Focussing on the historical dependency of Nigeria and the evolution of the state into a resource-dependent country, it looks at the limitations of existing human rights obligations as they relate to business. This article proposes that corporate social responsibility (CSR) policies of MNCs can act as a preliminary stage in the quest for wider human rights protections. It is in motivating MNCs to design and implement effective CSR policies in dependent states like Nigeria, that the challenge lies.

Sediment metal bioavailability in Lake Taihu, China: evaluation of sequential extraction, DGT, and PBET techniques

The European “Community Bureau of Reference” (BCR) sequential extraction procedure, diffusive gradient in thin-films technique (DGT), and physiologically based extraction test were applied to assess metal bioavailability in sediments of Lake Taihu (n = 13). Findings from the three methods showed that Cd was a significant problem in the western lake whereas Cu, Zn, and Ni pollution was most severe in the northern lake. Results from the sequential extraction revealed that more than 50 % of the Cu and Zn were highly mobile and defined within the extractable fraction (AS1 + FM2 + OS3) in the majority of the sediments, in contrast extractable fractions of Ni and Cd were lower than 50 % in most of the sampling sites. Average Cu, Zn, Ni, and Cd bioaccessibilities were <50 % in the gastric phase. Zn and Cd bioaccessibility in the intestinal phase was ∼50 % lower than the gastric phase while bioaccessibilities of Cu and Ni were 47–57 % greater than the gastric phase. Linear regression analysis between DGT and BCR measurements indicated that the extractable fractions (AS1 + FM2 + OS3) in the reducing environment were the main source of DGT uptake, suggesting that DGT is a good in situ evaluation tool for metal bioavailability in sediments.

Medical information extraction in European Portuguese

The electronic storage of medical patient data is becoming a daily experience in most of the practices and hospitals worldwide. However, much of the data available is in free-form text, a convenient way of expressing concepts and events, but especially challenging if one wants to perform automatic searches, summarization or statistical analysis. Information Extraction can relieve some of these problems by offering a semantically informed interpretation and abstraction of the texts. MedInX, the Medical Information eXtraction system presented in this document, is the first information extraction system developed to process textual clinical discharge records written in Portuguese. The main goal of the system is to improve access to the information locked up in unstructured text, and, consequently, the efficiency of the health care process, by allowing faster and reliable access to quality information on health, for both patient and health professionals. MedInX components are based on Natural Language Processing principles, and provide several mechanisms to read, process and utilize external resources, such as terminologies and ontologies, in the process of automatic mapping of free text reports onto a structured representation. However, the flexible and scalable architecture of the system, also allowed its application to the task of Named Entity Recognition on a shared evaluation contest focused on Portuguese general domain free-form texts. The evaluation of the system on a set of authentic hospital discharge letters indicates that the system performs with 95% F-measure, on the task of entity recognition, and 95% precision on the task of relation extraction. Example applications, demonstrating the use of MedInX capabilities in real applications in the hospital setting, are also presented in this document. These applications were designed to answer common clinical problems related with the automatic coding of diagnoses and other health-related conditions described in the documents, according to the international classification systems ICD-9-CM and ICF. The automatic review of the content and completeness of the documents is an example of another developed application, denominated MedInX Clinical Audit system.

Extraction and purification of triterpenic acids from eucalypt bark

A indústria da pasta e do papel é um sector importante da economia mundial, particularmente a que assenta em espécies de Eucalyptus. No entanto, essas indústrias geram quantidades significativas de correntes secundárias de subprodutos e resíduos de biomassa que podem ser exploradas em aplicações de valor acrescentado em vez de serem queimadas para produção de energia. Um exemplo nobre pode ser a produção de ácidos triterpénicos com estruturas dos tipos lupano, oleanano e ursano, dada a sua abundância em alguns destes resíduos, em particular na casca, adotando o conceito de biorrefinaria integrada numa fábrica de pasta. Estes compostos são conhecidos pelas suas inúmeras actividades biológicas, por exemplo, antioxidante, anti-inflamatória e antitumoral, e podem encontrar aplicações em produtos de elevado valor, tais como cosméticos, nutracêuticos ou farmacêuticos. Neste sentido, o estudo das potencialidades das cascas das espécies de eucalipto mais exploradas enquanto fontes de compostos triterpénicos é um tópico relevante. Por conseguinte, foram analisados e comparados em pormenor os teores e composições em ácidos triterpénicos (TTAs) das cascas externas de várias espécies de eucalipto (E. globulus, E. grandis, E. urograndis, E. maidenii e E. nitens). Os teores dos principais TTAs identificados nestas espécies variaram entre 4.5 g/kg no E. urograndis e 21.6 g/kg no E. nitens. Observou-se que as cascas externas de Eucalyptus de zonas temperadas e Mediterrânicas, nomeadamente E. nitens e E. globulus, são mais ricas em TTAs que as espécies de regiões tropicais e subtropicais. Além disso, a casca externa do E. globulus é claramente a mais rica em ácidos com estruturas do tipo ursano enquanto a do E. nitens é a mais rica em ácidos do tipo oleanano e lupano. Estes resultados levaram-nos a estudar a extração dos TTAs da casca de Eucalyptus, bem como a sua posterior concentração e purificação, a qual foi efetuada por extração sólido-líquido convencional combinada com a precipitação de solutos, e por extração com fluidos supercríticos (SFE). No que diz respeito à primeira abordagem referida, foi desenvolvido neste trabalho um método patenteado que permite obter extratos enriquecidos em TTAs das cascas de eucalipto baseado em tecnologias disponíveis no imediato. Em relação à segunda abordagem, e de forma a apostar em processos de baixo impacto ambiental exigidos pelas biorrefinarias do futuro, a SFE surge como uma opção natural. Assim, foi efetuada a SFE da casca caduca do E. globulus com dióxido de carbono puro e modificado para recuperar a fração de TTAs, e os resultados foram comparados com os obtidos por extração em Soxhlet com diclorometano. Foram realizados estudos preliminares sobre a influência da pressão (100-200 bar), a adição de co-solvente (0, 5 e 8% m/m de etanol), e operação em múltiplos passos a fim de avaliar a aplicabilidade da alternativa supercrítica para a sua produção eficiente e selectiva. Os resultados destacaram a influência da pressão e o importante papel resumo (cont.) desempenhado pelo co-solvente neste processo, cujo efeito foi mais relevante do que o aumento da pressão em várias dezenas de bar. Este trabalho foi depois otimizado, usando o planeamento factorial de experiências e a metodologia de superfícies de resposta, para analisar a influência da temperatura (40-60 ºC), pressão (100-200 bar), e teor de etanol (0.0-5.0% m/m) na recuperação dos TTAs e respectiva concentração nos extractos. Nestes intervalos, as melhores condições de operação encontradas foram 200 bar, 40 °C e 5% de etanol, para as quais os modelos de regressão estatisticamente validados previram um rendimento de extração de 1.2% com 50% de concentração em TTAs, correspondendo ao rendimento em TTAs de 5.1 g/kg de casca e uma recuperação de 79.2% comparativamente ao valor do Soxhlet. Os TTAs livres e acetilados apresentaram tendências de extracção bastante distintas devido às suas diferentes afinidades para o CO2 causadas pelas diferentes polaridades: os derivados acetilados aproximam-se de um patamar máximo a cerca de 200 bar e 5% de etanol, enquanto a extração dos TTAs livres apresenta uma tendência sempre crescente no intervalo de condições estudado. Foram também medidas curvas cumulativas de SFE da casca do E. globulus de forma a analisar o comportamento cinético do processo em termos de rendimento total, rendimento em TTAs, rendimento em TTAs livres, rendimento em TTAs acetilados, e concentração dos TTAs nos extractos. Foi analisada a influência da pressão, temperatura, teor de co-solvente e caudal do dióxido de carbono sobre as respostas anteriores. Os dados experimentais foram modelados com os modelos Logístico, de Dessorção, de Placa Plana Simples, e de Difusão. Na globalidade, os resultados confirmaram que a pressão e o teor de etanol têm um efeito significativo sobre as curvas de extração, os rendimentos finais e as concentrações dos extratos, e mostraram a presença de limitações externas à transferência de massa em alguns ensaios. Mais uma vez, as famílias individuais de TTAs livres e acetilados apresentaram diferentes tendências de extracção. A modelação permitiu-nos confirmar não só o importante papel desempenhado pela difusão intraparticular na SFE, mas também a contribuição da resistência no filme em alguns ensaios. Após a análise de todos os resultados, foi efetuado um ensaio em duas etapas em série, possibilitando o enriquecimento do teor em TTAs no extracto devido às diferentes condições adotadas em cada etapa. Por último, um éster metílico de um ácido triterpénico do tipo oleanano - morolato de metilo - foi identificado pela primeira vez enquanto componente da casca de Eucalyptus na casca externa do Eucalyptus grandis x globulus, onde ocorre em teores elevados. A sua extração com CO2 supercrítico foi também realizada, visando a conceção de uma alternativa de extração ambientalmente benigna para este composto. A 200 bar e 60 ºC, a remoção do morolato de metilo atingiu um patamar às 6 h para 5.1 kg h-1 de CO2 / kg de casca. Em geral, e de forma semelhante à SFE da casca do E. globulus, os TTAs acetilados foram mais significativamente extraídos quando comparados com os seus ácidos livres, o que está diretamente relacionado com a natureza menos polar destas moléculas. O trabalho apresentado nesta tese é uma contribuição para a valorização de uma corrente de biomassa com baixo valor na indústria de pasta em duas vertentes complementares. Por um lado, aumentou o conhecimento da composição lipofílica das cascas de Eucalyptus spp. com interesse comercial para a produção de pasta, destacando o seu potencial enquanto fontes de ácidos triterpénicos. Por outro lado, foram desenvolvidos dois processos alternativos e facilmente integráveis numa fábrica de pasta para a sua exploração a partir da casca: um baseado em tecnologias convencionais bem estabelecidas a nível industrial, prevendo a sua aplicação a curto prazo, e um outro baseado na SFE, seguindo as tendências das futuras biorrefinarias.