NovaIntell - projecto de text-Mining para a língua portuguesa numa empresa de Gestão de Informação e Conhecimento

Dissertação apresentada como requisito parcial para obtenção do grau de Mestre em Estatística e Gestão de Informação

Antigenic and genetic characterization of the first rabies virus isolated from the bat Eumops perotis in Brazil

Although the main transmitters of rabies in Brazil are dogs and vampire bats, the role of other species such as insectivorous and frugivorous bats deserves special attention, as the rabies virus has been isolated from 36 bat species. This study describes the first isolation of the rabies virus from the insectivorous bat Eumops perotis. The infected animal was found in the city of Ribeirão Preto, São Paulo. The virus was identified by immunofluorescence antibody test (FAT) in central nervous system (CNS) samples, and the isolation was carried out in N2A cell culture and adult mice. The sample was submitted to antigenic typing using a panel of monoclonal antibodies (CDC/Atlanta/USA). The DNA sequence of the nucleoprotein gene located between nucleotides 102 and 1385 was aligned with homologous sequences from GenBank using the CLUSTAL/W method, and the alignment was used to build a neighbor-joining distance-based phylogenetic tree with the K-2-P model. CNS was negative by FAT, and only one mouse died after inoculation with a suspension from the bat's CNS. Antigenic typing gave a result that was not compatible with the patterns defined by the panel. Phylogenetic analysis showed that the virus isolated segregated into the same cluster related to other viruses isolated from insectivorous bats belonging to genus Nyctinomops ssp. (98.8% nucleotide identity with each other).

A real-time quantitative assay for hepatitis B DNA virus (HBV) developed to detect all HBV genotypes

Hepatitis B virus (HBV) is a major cause of chronic liver disease worldwide. Besides genotype, quantitative analysis of HBV infection is extensively used for monitoring disease progression and treatment. Affordable viral load monitoring is desirable in resource-limited settings and it has been already shown to be useful in developing countries for other viruses such as Hepatitis C virus (HCV) and HIV. In this paper, we describe the validation of a real-time PCR assay for HBV DNA quantification with TaqMan chemistry and MGB probes. Primers and probes were designed using an alignment of sequences from all HBV genotypes in order to equally amplify all of them. The assay is internally controlled and was standardized with an international HBV panel. Its efficacy was evaluated comparing the results with two other methods: Versant HBV DNA Assay 3.0 (bDNA, Siemens, NY, USA) and another real-time PCR from a reference laboratory. Intra-assay and inter-assay reproducibilities were determined and the mean of CV values obtained were 0.12 and 0.09, respectively. The assay was validated with a broad dynamic range and is efficient for amplifying all HBV genotypes, providing a good option to quantify HBV DNA as a routine procedure, with a cheap and reliable protocol.

Automatic syllabification for danish text-to-speech systems

In this paper, a rule-based automatic syllabifier for Danish is described using the Maximal Onset Principle. Prior success rates of rule-based methods applied to Portuguese and Catalan syllabification modules were on the basis of this work. The system was implemented and tested using a very small set of rules. The results gave rise to 96.9% and 98.7% of word accuracy rate, contrary to our initial expectations, being Danish a language with a complex syllabic structure and thus difficult to be rule-driven. Comparison with data-driven syllabification system using artificial neural networks showed a higher accuracy rate of the former system.

TESTING A SUBTYPE-SPECIFIC GP41 AMPLIFICATION METHOD FOR GENOTYPING INDIVIDUALS INFECTED BY HUMAN IMMUNODEFICIENCY VIRUS TYPE-1 IN THE BRAZILIAN POPULATION OF ITAJAÍ, SOUTH BRAZIL

The method used by YAGYU et al. for the subtype-specific polymerase chain reaction (PCR) amplification of the gp41 transmembrane region of the human immunodeficiency virus type-1 (HIV-1) env gene, was tested. HIV-1 proviral DNA from 100 infected individuals in Itajaí, South Brazil was used to analyze this method. Seventy individuals were determined according to this method as having PCR products at the expected size for subtypes B, C, D and F. Of these individuals, 26 (37.1%) were observed as having the expected amplification for subtype C, and 42 (60%) were observed as having the expected products for subtypes B and D. Of the subtype B and D amplicons, 16 (22.9%) were classified as subtype D, and 26 (37.1%) were classified as subtype B. Two individuals (2.9%) had amplicons that were observed after subtype F-specific amplification was performed. Sequencing and comparing the patient sequences to reference sequences confirmed the classification of sequences of subtypes C and B. However, sequences that were falsely determined as being D and F in the PCR assay were determined as being subtypes C and B, respectively, by sequence analysis. For those individuals from whom no amplified products were obtained, a low viral load that was indicated in their patient history may explain the difficulty in subtyping by PCR methods. This issue was demonstrated by the results of ANOVA when testing the effect of viral load on the success of PCR amplification. The alignment of the obtained sequences with HIV-1 reference sequences demonstrated that there is high intra-subtype diversity. This indicates that the subtype-specific primer binding sites were not conserved or representative of the subtypes that are observed in the Brazilian populations, and that they did not allow the correct classification of HIV-1 subtypes. Therefore, the proposed method by YAGYU et al. is not applicable for the classification of Brazilian HIV-1 subtypes.

Text Mining: análise de sentimentos na classificação de notícias

Trabalho de Projeto apresentado como requisito parcial para obtenção do grau de Mestre em Estatística e Gestão de Informação

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

Framework for the semantic alignment of enterprise’s domain knowledge

Nowadays, the consumption of goods and services on the Internet are increasing in a constant motion. Small and Medium Enterprises (SMEs) mostly from the traditional industry sectors are usually make business in weak and fragile market sectors, where customized products and services prevail. To survive and compete in the actual markets they have to readjust their business strategies by creating new manufacturing processes and establishing new business networks through new technological approaches. In order to compete with big enterprises, these partnerships aim the sharing of resources, knowledge and strategies to boost the sector’s business consolidation through the creation of dynamic manufacturing networks. To facilitate such demand, it is proposed the development of a centralized information system, which allows enterprises to select and create dynamic manufacturing networks that would have the capability to monitor all the manufacturing process, including the assembly, packaging and distribution phases. Even the networking partners that come from the same area have multi and heterogeneous representations of the same knowledge, denoting their own view of the domain. Thus, different conceptual, semantic, and consequently, diverse lexically knowledge representations may occur in the network, causing non-transparent sharing of information and interoperability inconsistencies. The creation of a framework supported by a tool that in a flexible way would enable the identification, classification and resolution of such semantic heterogeneities is required. This tool will support the network in the semantic mapping establishments, to facilitate the various enterprises information systems integration.

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.

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.

Text mining na análise de sentimentos em contextos de big data

Dissertação de mestrado integrado em Engenharia e Gestão de Sistemas de Informação