Automatic information retrieval through text-mining

The dissertation presented for obtaining the Master’s Degree in Electrical Engineering and Computer Science, at Universidade Nova de Lisboa, Faculdade de Ciências e Tecnologia

electronic payments in mozambique: A Baseline on their adoption in Maputo and Matola

This study analyses the access and use of financial services by small business owners in the cities of Mozambique, as an important tool for boosting economic growth and diminishing inequality. It correlates owners’ and business characteristics with the probability of adopting Points-of-Sale (POS), Mobile Banking and Mobile Money in everyday transactions. The main findings highlight that what mostly affects the use of POS is the size of business and the volume of transactions (positively correlated with POS adoption), while using mobile phone technologies for payments predominantly depends on the owner’s age and whether he/she is a frequent cellphone user. Moreover, to increase the use of electronic means of payment it is necessary to increase financial literacy and improve the banking services.

Spatio-temporal patterns of tuberculosis incidence in Ribeirão Preto, State of São Paulo, southeast Brazil, and their relationship with social vulnerability: a Bayesian analysis

INTRODUCTION: The purpose of this ecological study was to evaluate the urban spatial and temporal distribution of tuberculosis (TB) in Ribeirão Preto, State of São Paulo, southeast Brazil, between 2006 and 2009 and to evaluate its relationship with factors of social vulnerability such as income and education level. METHODS: We evaluated data from TBWeb, an electronic notification system for TB cases. Measures of social vulnerability were obtained from the SEADE Foundation, and information about the number of inhabitants, education and income of the households were obtained from Brazilian Institute of Geography and Statistics. Statistical analyses were conducted by a Bayesian regression model assuming a Poisson distribution for the observed new cases of TB in each area. A conditional autoregressive structure was used for the spatial covariance structure. RESULTS: The Bayesian model confirmed the spatial heterogeneity of TB distribution in Ribeirão Preto, identifying areas with elevated risk and the effects of social vulnerability on the disease. We demonstrated that the rate of TB was correlated with the measures of income, education and social vulnerability. However, we observed areas with low vulnerability and high education and income, but with high estimated TB rates. CONCLUSIONS: The study identified areas with different risks for TB, given that the public health system deals with the characteristics of each region individually and prioritizes those that present a higher propensity to risk of TB. Complex relationships may exist between TB incidence and a wide range of environmental and intrinsic factors, which need to be studied in future research.

Health-related quality of life in patients with Chagas disease: a review of the evidence

Chagas disease (ChD), a neglected tropical disease caused by infection with the parasite Trypanosoma cruzi (T. cruzi), remains a serious public health issue in Latin America and is an emerging disease in several non-endemic countries, where knowledge of the condition and experience with its clinical management are limited. Regionally, the disease is the major cause of disability secondary to tropical diseases in young adults. Health-related quality of life (HRQoL) impairment is common in patients with ChD, especially in those with Chagas dilated cardiomyopathy, the most severe manifestation of the disease, which frequently leads to heart failure. The aim of this review was to conduct a literature search for studies that have evaluated the determining factors of HRQoL in ChD patients. We included cross-sectional, case-control, cohort, and experimental studies, as well as clinical trials that evaluated the HRQoL in ChD patients aged 18 to 60 years and are presenting an explicit description of statistical analysis. Using a combination of keywords based on Descriptors in Health Sciences (DeCS) and Medical Subject Headings (MeSH) for searches in PubMed and the Scientific Electronic Library Online (SciELO), 148 studies were found. After exclusions, 12 studies were selected for analysis. Three main findings were extracted from these studies: 1) cardiac involvement is associated with a worse HRQoL in ChD patients; 2) HRQoL is associated with the patients' functional capacity; and 3) simple and inexpensive therapeutic measures are effective for improving HRQoL in ChD patients. Hence, ChD patients' functional capacity, the effectiveness of non-surgical conservative treatment, and cardiac involvement are important determining factors for the HRQoL in ChD patients.

Effects of continuous versus bolus infusion of enteral nutrition in critical patients

PURPOSE: Enteral alimentation is the preferred modality of support in critical patients who have acceptable digestive function and are unable to eat orally, but the advantages of continuous versus intermittent administration are surrounded by controversy. With the purpose of identifying the benefits and complications of each technique, a prospective controlled study with matched subjects was conducted. PATIENTS AND METHODS: Twenty-eight consecutive candidates for enteral feeding were divided into 2 groups (n = 14 each) that were matched for diagnosis and APACHE II score. A commercial immune-stimulating polymeric diet was administered via nasogastric tube by electronic pump in the proportion of 25 kcal/kg/day, either as a 1-hour bolus every 3 hours (Group I), or continuously for 24 hours (Group II), over a 3-day period. Anthropometrics, biochemical measurements, recording of administered drugs and other therapies, thorax X-ray, measurement of abdominal circumference, monitoring of gastric residue, and clinical and nutritional assessments were performed at least once daily. The principal measured outcomes of this protocol were frequency of abdominal distention and pulmonary aspiration, and efficacy in supplying the desired amount of nutrients. RESULTS: Nearly half of the total population (46.4%) exhibited high gastric residues on at least 1 occasion, but only 1 confirmed episode of pulmonary aspiration occurred (3.6%). Both groups displayed a moderate number of complications, without differences. Food input during the first day was greater in Group II (approximately 20% difference), but by the third day, both groups displayed similarly small deficits in total furnished volume of about 10%, when compared with the prescribed diet. CONCLUSIONS: Both administration modalities permitted practical and effective administration of the diet with frequent registered abnormalities but few clinically significant problems. The two groups were similar in this regard, without statistical differences, probably because of meticulous technique, careful monitoring, strict patient matching, and conservative amounts of diet employed in both situations. Further studies with additional populations, diagnostic groups, and dietetic prescriptions should be performed in order to elucidate the differences between these commonly used feeding modalities.

Ethical aspects in the management of the terminally ill patient in the pediatric intensive care unit

OBJECTIVE: To identify the prevalence of management plans and decision-making processes for terminal care patients in pediatric intensive care units. METHODOLOGY: Evidence-based medicine was done by a systematic review using an electronic data base (LILACS, 1982 through 2000) and (MEDLINE, 1966 through 2000). The key words used are listed and age limits (0 to 18 years) were used. RESULTS: One hundred and eighty two articles were found and after selection according to the exclusion/inclusion criteria and objectives 17 relevant papers were identified. The most common decisions found were do-not-resuscitation orders and withdrawal or withholding life support care. The justifications for these were "imminent death" and "unsatisfatory quality of life". CONCLUSION: Care management was based on ethical principles aiming at improving benefits, avoiding harm, and when possible, respecting the autonomy of the terminally ill patient.

Text mining e twitter : o poder das redes sociais num mercado competitivo

Actualmente, com a massificação da utilização das redes sociais, as empresas passam a sua mensagem nos seus canais de comunicação, mas os consumidores dão a sua opinião sobre ela. Argumentam, opinam, criticam (Nardi, Schiano, Gumbrecht, & Swartz, 2004). Positiva ou negativamente. Neste contexto o Text Mining surge como uma abordagem interessante para a resposta à necessidade de obter conhecimento a partir dos dados existentes. Neste trabalho utilizámos um algoritmo de Clustering hierárquico com o objectivo de descobrir temas distintos num conjunto de tweets obtidos ao longo de um determinado período de tempo para as empresas Burger King e McDonald’s. Com o intuito de compreender o sentimento associado a estes temas foi feita uma análise de sentimentos a cada tema encontrado, utilizando um algoritmo Bag-of-Words. Concluiu-se que o algoritmo de Clustering foi capaz de encontrar temas através do tweets obtidos, essencialmente ligados a produtos e serviços comercializados pelas empresas. O algoritmo de Sentiment Analysis atribuiu um sentimento a esses temas, permitindo compreender de entre os produtos/serviços identificados quais os que obtiveram uma polaridade positiva ou negativa, e deste modo sinalizar potencias situações problemáticas na estratégia das empresas, e situações positivas passíveis de identificação de decisões operacionais bem-sucedidas.

Business intelligence in banking: A literature analysis from 2002 to 2013 using Text Mining and latent Dirichlet allocation

telligence applications for the banking industry. Searches were performed in relevant journals resulting in 219 articles published between 2002 and 2013. To analyze such a large number of manuscripts, text mining techniques were used in pursuit for relevant terms on both business intelligence and banking domains. Moreover, the latent Dirichlet allocation modeling was used in or- der to group articles in several relevant topics. The analysis was conducted using a dictionary of terms belonging to both banking and business intelli- gence domains. Such procedure allowed for the identification of relationships between terms and topics grouping articles, enabling to emerge hypotheses regarding research directions. To confirm such hypotheses, relevant articles were collected and scrutinized, allowing to validate the text mining proce- dure. The results show that credit in banking is clearly the main application trend, particularly predicting risk and thus supporting credit approval or de- nial. There is also a relevant interest in bankruptcy and fraud prediction. Customer retention seems to be associated, although weakly, with targeting, justifying bank offers to reduce churn. In addition, a large number of ar- ticles focused more on business intelligence techniques and its applications, using the banking industry just for evaluation, thus, not clearly acclaiming for benefits in the banking business. By identifying these current research topics, this study also highlights opportunities for future research.

A novel approach for honey pollen profile assessment using an electronic tongue and chemometric tools

Nowadays the main honey producing countries require accurate labeling of honey before commercialization, including floral classification. Traditionally, this classification is made by melissopalynology analysis, an accurate but time-consuming task requiring laborious sample pre-treatment and high-skilled technicians. In this work the potential use of a potentiometric electronic tongue for pollinic assessment is evaluated, using monofloral and polyfloral honeys. The results showed that after splitting honeys according to color (white, amber and dark), the novel methodology enabled quantifying the relative percentage of the main pollens (Castanea sp., Echium sp., Erica sp., Eucaliptus sp., Lavandula sp., Prunus sp., Rubus sp. and Trifolium sp.). Multiple linear regression models were established for each type of pollen, based on the best sensors sub-sets selected using the simulated annealing algorithm. To minimize the overfitting risk, a repeated K-fold cross-validation procedure was implemented, ensuring that at least 10-20% of the honeys were used for internal validation. With this approach, a minimum average determination coefficient of 0.91 ± 0.15 was obtained. Also, the proposed technique enabled the correct classification of 92% and 100% of monofloral and polyfloral honeys, respectively. The quite satisfactory performance of the novel procedure for quantifying the relative pollen frequency may envisage its applicability for honey labeling and geographical origin identification. Nevertheless, this approach is not a full alternative to the traditional melissopalynologic analysis; it may be seen as a practical complementary tool for preliminary honey floral classification, leaving only problematic cases for pollinic evaluation.

Reconstructing transcriptional regulatory networks using data integration and text mining

Transcriptional Regulatory Networks (TRNs) are powerful tool for representing several interactions that occur within a cell. Recent studies have provided information to help researchers in the tasks of building and understanding these networks. One of the major sources of information to build TRNs is biomedical literature. However, due to the rapidly increasing number of scientific papers, it is quite difficult to analyse the large amount of papers that have been published about this subject. This fact has heightened the importance of Biomedical Text Mining approaches in this task. Also, owing to the lack of adequate standards, as the number of databases increases, several inconsistencies concerning gene and protein names and identifiers are common. In this work, we developed an integrated approach for the reconstruction of TRNs that retrieve the relevant information from important biological databases and insert it into a unique repository, named KREN. Also, we applied text mining techniques over this integrated repository to build TRNs. However, was necessary to create a dictionary of names and synonyms associated with these entities and also develop an approach that retrieves all the abstracts from the related scientific papers stored on PubMed, in order to create a corpora of data about genes. Furthermore, these tasks were integrated into @Note, a software system that allows to use some methods from the Biomedical Text Mining field, including an algorithms for Named Entity Recognition (NER), extraction of all relevant terms from publication abstracts, extraction relationships between biological entities (genes, proteins and transcription factors). And finally, extended this tool to allow the reconstruction Transcriptional Regulatory Networks through using scientific literature.

Electronic transport across linear defects in graphene

We investigate the low-energy electronic transport across grain boundaries in graphene ribbons and infinite flakes. Using the recursive Green’s function method, we calculate the electronic transmission across different types of grain boundaries in graphene ribbons. We show results for the charge density distribution and the current flow along the ribbon. We study linear defects at various angles with the ribbon direction, as well as overlaps of two monolayer ribbon domains forming a bilayer region. For a class of extended defect lines with periodicity 3, an analytic approach is developed to study transport in infinite flakes. This class of extended grain boundaries is particularly interesting, since the K and K0 Dirac points are superposed.

Actinella species (Bacillariophyta) from an Amazon black water floodplain lake (Amazonas - Brazil)

The genus Actinella Lewis was studied using planktonic samples from a black water floodplain lake in Central Amazon region. For species identification the taxa were morphological and morphometricaly analyzed on base in light microscope (LM) and scanning electronic microscope (SEM). Five species were registered: Actinella brasiliensis Grunow, A. guianensis Grunow, A. gracile Kociolek, A. mirabilis (Eulenstein ex Grunow) Grunow and A. robusta Hustedt. A. gracile is reported for the first time for Amazon State and black water systems and it is firstly documented with SEM. In addition, a review of geographic distribution of Actinella species in Brazilian Amazon region is given.

Sensory intensity assessment of olive oils using an electronic tongue

Olive oils may be commercialized as intense, medium or light, according to the intensity perception of fruitiness, bitterness and pungency attributes, assessed by a sensory panel. In this work, the capability of an electronic tongue to correctly classify olive oils according to the sensory intensity perception levels was evaluated. Cross-sensitivity and non-specific lipid polymeric membranes were used as sensors. The sensor device was firstly tested using quinine monohydrochloride standard solutions. Mean sensitivities of 14±2 to 25±6 mV/decade, depending on the type of plasticizer used in the lipid membranes, were obtained showing the device capability for evaluating bitterness. Then, linear discriminant models based on sub-sets of sensors, selected by a meta-heuristic simulated annealing algorithm, were established enabling to correctly classify 91% of olive oils according to their intensity sensory grade (leave-one-out cross-validation procedure). This capability was further evaluated using a repeated K-fold cross-validation procedure, showing that the electronic tongue allowed an average correct classification of 80% of the olive oils used for internal-validation. So, the electronic tongue can be seen as a taste sensor, allowing differentiating olive oils with different sensory intensities, and could be used as a preliminary, complementary and practical tool for panelists during olive oil sensory analysis.

Monovarietal extra-virgin olive oil classification: a fusion of human sensory attributes and an electronic tongue

Olive oil quality grading is traditionally assessed by human sensory evaluation of positive and negative attributes (olfactory, gustatory, and final olfactorygustatory sensations). However, it is not guaranteed that trained panelist can correctly classify monovarietal extra-virgin olive oils according to olive cultivar. In this work, the potential application of human (sensory panelists) and artificial (electronic tongue) sensory evaluation of olive oils was studied aiming to discriminate eight single-cultivar extra-virgin olive oils. Linear discriminant, partial least square discriminant, and sparse partial least square discriminant analyses were evaluated. The best predictive classification was obtained using linear discriminant analysis with simulated annealing selection algorithm. A low-level data fusion approach (18 electronic tongue signals and nine sensory attributes) enabled 100 % leave-one-out cross-validation correct classification, improving the discrimination capability of the individual use of sensor profiles or sensory attributes (70 and 57 % leave-one-out correct classifications, respectively). So, human sensory evaluation and electronic tongue analysis may be used as complementary tools allowing successful monovarietal olive oil discrimination.