A hybrid model for named entity recognition using unstructured medical text

Named entity recognition (NER) is an essential step in the process of information extraction within text mining. This paper proposes a technique to extract drug named entities from unstructured and informal medical text using a hybrid model of lexicon-based and rule-based techniques. In the proposed model, a lexicon is first used as the initial step to detect drug named entities. Inference rules are then deployed to further extract undetected drug names. The designed rules employ part of speech tags and morphological features for drug name detection. The proposed hybrid model is evaluated using a benchmark data set from the i2b2 2009 medication challenge, and is able to achieve an f-score of 66.97%.

Mining streams of short text for analysis of world-wide event evolutions

Streams of short text, such as news titles, enable us to effectively and efficiently learn the real world events that occur anywhere and anytime. Short text messages that are companied by timestamps and generally brief events using only a few words differ from other longer text documents, such as web pages, news stories, blogs, technical papers and books. For example, few words repeat in the same news titles, thus frequency of the term (i.e., TF) is not as important in short text corpus as in longer text corpus. Therefore, analysis of short text faces new challenges. Also, detecting and tracking events through short text analysis need to reliably identify events from constant topic clusters; however, existing methods, such as Latent Dirichlet Allocation (LDA), generates different topic results for a corpus at different executions. In this paper, we provide a Finding Topic Clusters using Co-occurring Terms (FTCCT) algorithm to automatically generate topics from a short text corpus, and develop an Event Evolution Mining (EEM) algorithm to discover hot events and their evolutions (i.e., the popularity degrees of events changing over time). In FTCCT, a term (i.e., a single word or a multiple-words phrase) belongs to only one topic in a corpus. Experiments on news titles of 157 countries within 4 months (from July to October, 2013) demonstrate that our FTCCT-based method (combining FTCCT and EEM) achieves far higher quality of the event's content and description words than LDA-based method (combining LDA and EEM) for analysis of streams of short text. Our method also visualizes the evolutions of the hot events. The discovered world-wide event evolutions have explored some interesting correlations of the world-wide events; for example, successive extreme weather phenomenon occur in different locations - typhoon in Hong Kong and Philippines followed hurricane and storm flood in Mexico in September 2013. © 2014 Springer Science+Business Media New York.

Using data mining for digital ink recognition: dividing text and shapes in sketched diagrams

The low accuracy rates of textshape dividers for digital ink diagrams are hindering their use in real world applications. While recognition of handwriting is well advanced and there have been many recognition approaches proposed for hand drawn sketches, there has been less attention on the division of text and drawing ink. Feature based recognition is a common approach for textshape division. However, the choice of features and algorithms are critical to the success of the recognition. We propose the use of data mining techniques to build more accurate textshape dividers. A comparative study is used to systematically identify the algorithms best suited for the specific problem. We have generated dividers using data mining with diagrams from three domains and a comprehensive ink feature library. The extensive evaluation on diagrams from six different domains has shown that our resulting dividers, using LADTree and LogitBoost, are significantly more accurate than three existing dividers.

Enhancing English language learners’ text understanding through metaphor awareness

 The study examined awareness of metaphor as a tool to enhance English language learners’ understanding of texts with embedded metaphors. Findings revealed that an enhanced awareness of metaphor, as indicated by greater use of the metalanguage of metaphor, longer turns conversation and reflective journals, helped them get deeper text meaning.

Sacredness of the constitutional text and interpretative heresy: the Brazilian Supreme Court decision on same-sex civil unions

The paper provides a close lecture of the arguments and methods of legal construction, employed in the extensive individual opinions written by the Justices of the Brazilian Supreme Court in the case which authorized the same sex civil union. After tracing an outline of the legal problem and his possible solutions, we analyze the individual opinions, showing their methodological syncretism, the use of legal methods and arguments in a contradictory way as well the deficiencies in the reasoning. The Justices use legal arguments, but do not meet the requirements of rationality in the decision-making. We have a rhetorical attempt that aims to satisfy the public opinion than to offer a comprehensive and coherent solution according the normative elements of the Brazilian Federal Constitution of 1988.

Neue medien und text: privatbrief und private E-Mail im vergleich

Die vorliegende Dissertation setzt sich mit dem Phänomen ›Text im Rahmen der Neuen Medien‹ auseinander, indem sie theoretisch und empirisch der in der einschlägigen Forschung aufgeworfenen (und kontrovers diskutierten) Frage nachgeht, ob die Sprache in den Textsorten der neuen Kommunikationsformen als neue Schriftlichkeit bzw. schriftliche Mündlichkeit zu verstehen sei. Dabei konzentriert sie sich exemplarisch auf die Analyse eigens für diesen Zweck erstellter aktueller Textkorpora der Kommunikationsformen E-Mail und Brief und untersucht sie unter den Gesichtspunkten Mündlichkeit / Schriftlichkeit bzw. Nähe / Distanz. ›Kapitel 1‹ umreißt den Forschungsstand zum Thema ›Text und Textsorte‹ sowohl in der Textlinguistik als auch in aktuellen Studien zur Sprache in den Neuen Medien. ›Kapitel 2‹ ist dem kritischen Referat verschiedener bekannter Modelle gewidmet, die sich mit dem Aspekt der Mündlichkeit und Schriftlichkeit bzw. Nähe und Distanz beschäftigen (Hugo Stegers Freiburger Gruppe, Ludwig Söll, Koch/Oesterreicher und Ágel/Hennig) und eine Verortung von Textsorten im Kontinuum Nähe und Distanz vornehmen (Koch/Oesterreicher und Ágel/Hennig).›Kapitel 3‹ schlägt Korrekturen und Ergänzungen betreffend Punktgebung, Modellglossar und Makroanalyse im Modell von Ágel/Hennig vor, das in dieser Arbeit den Analysen der Nähesprachlichkeit in Privatbriefen und privaten E-Mails zugrunde liegt. ›Kapitel 4‹ setzt sich mit Aspekten der beiden Textkorpora auseinander, auf die sich diese Arbeit stützt, und beschreibt und erläutert den Fragebogen, der im Rahmen dieser Arbeit erstellt und Probanden zur Beantwortung vorgelegt wurde. ›Kapitel 5‹ nimmt einen ausführlichen Vergleich der Textsorten ›Privatbrief und private E-Mail‹ vor und beleuchtet abrissartig die Geschichte beider Kommunikationsformen, wobei es sich gleichzeitig auch kritisch mit den in der Forschung in solchen Vergleichsfragen vertretenen Positionen auseinander setzt. ›Kapitel 6‹ interpretiert die Ergebnisse der in der Arbeit durchgeführten Näheanalysen und zieht daraus die Schlussfolgerungen.

Aircraft interior failure pattern recognition utilizing text mining and neural networks

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Discrepancies between abstracts presented at international association for dental research annual sessions from 2004 to 2005 and full-text publication

Purpose. The purpose of this study was to evaluate the discrepancies between abstracts presented at the IADR meeting (2004-2005) and their full-text publication. Material and Methods. Abstracts from the Prosthodontic Section of IADR meeting were obtained. The following information was collected: abstract title, number of authors, study design, statistical analysis, outcome, and funding source. PubMed was used to identify the full-text publication of the abstracts. The discrepancies between the abstract and the full-text publication were examined, categorized as major and minor discrepancies, and quantified. The data were collected and analyzed using descriptive analysis. Frequency and percentage of major and minor discrepancies were calculated. Results. A total of 109 (95.6%) articles showed changes from their abstracts. Seventy-four (65.0%) and 105 (92.0%) publications had at least one major and one minor discrepancies, respectively. Minor discrepancies were more prevalent (92.0%) than major discrepancies (65.0%). The most common minor discrepancy was observed in the title (80.7%), and most common major discrepancies were seen in results (48.2%). Conclusion. Minor discrepancies were more prevalent than major discrepancies. The data presented in this study may be useful to establish a more comprehensive structured abstract requirement for future meetings. © 2012 Soni Prasad et al.

Draft revised text of the constituent declaration and functions and rules of procedure of the Caribbean Development and Cooperation Committee as adopted by the 19th Session of the CDCC, 13-14 March 2002. Note by the Secretariat

Includes bibliography