Subtopic annotation and automatic segmentation for news texts in Brazilian Portuguese

Corpora ◽  
2017 ◽  
Vol 12 (1) ◽  
pp. 23-54 ◽  
Author(s):  
Paula C.F. Cardoso ◽  
Thiago A.S. Pardo ◽  
Maite Taboada

Subtopic segmentation aims to break documents into subtopical text passages, which develop a main topic in a text. Being capable of automatically detecting subtopics is very useful for several Natural Language Processing applications. For instance, in automatic summarisation, having the subtopics at hand enables the production of summaries with good subtopic coverage. Given the usefulness of subtopic segmentation, it is common to assemble a reference-annotated corpus that supports the study of the envisioned phenomena and the development and evaluation of systems. In this paper, we describe the subtopic annotation process in a corpus of news texts written in Brazilian Portuguese, following a systematic annotation process and answering the main research questions when performing corpus annotation. Based on this corpus, we propose novel methods for subtopic segmentation following patterns of discourse organisation, specifically using Rhetorical Structure Theory. We show that discourse structures mirror the subtopic changes in news texts. An important outcome of this work is the freely available annotated corpus, which, to the best of our knowledge, is the only one for Portuguese. We demonstrate that some discourse knowledge may significantly help to find boundaries automatically in a text. In particular, the relation type and the level of the tree structure are important features.

2012 ◽  
Author(s):  
Felipe Rodrigues ◽  
Richard Semolini ◽  
Norton Trevisan Roman ◽  
Ana Maria Monteiro

This paper describes TSeg – a Java application that allows for both manual and automatic segmentation of a source text into basic units of annotation. TSeg provides a straightforward way to approach this task through a clear point-and-click interface. Once finished the text segmentation, the application outputs an XML file that may be used as input to a more problem specific annotation software. Hence, TSeg moves the identification of basic units of annotation out of the task of annotating these units, making it possible for both problems to be analysed in isolation, thereby reducing the cognitive load on the user and preventing potential damages to the overall outcome of the annotation process.


2017 ◽  
Vol 1 (104) ◽  
pp. 55-64 ◽  
Author(s):  
Kadir Yıldız ◽  
Pınar Güzel ◽  
Fırat Çetinöz ◽  
Tolga Beşikçi

Background. In this research, we aimed to investigate the effects of outdoor camps on orienteering athletes. Methods. The study group consisted of 74 athletes (44 males and 30 females, aged 11.94 ± 1.32 years) who participated in Bolu outdoor camp on the 3 rd –13 th of August, 2015. Interview technique, which is one of the qualitative research methods, was used as data collection tool and content analysis method was used for data analysis. Results. Demographic factors were interpreted after the analysis of the obtained data and three main research questions were discussed under the topics of the views of athletes about the concept of Orienteering which is an outdoor sport, themes and codes regarding the purpose of Orienteering by the students who participated in the outdoor camp, and themes and codes about the outcomes of Orienteering for the students who participated in outdoor camps. Conclusion. It is suggested that a policy must be developed within the Ministry of Youth and Sport and Sport Federations in order to disseminate more deliberate and more comprehensive outdoor education among young people and measures should be taken to provide extensive participation.


2021 ◽  
Vol 1 ◽  
pp. 1373-1382
Author(s):  
Avril Thomson ◽  
Hilary Grierson

AbstractThe paper reports on a study that aims to gain an understanding of how senior engineering design students engage and attain throughout the various stages of the design process during a major design project. Following a literature review it sets out to answer 3 main research questionsQ1. Do students engage more with certain stages of the design process during major project work?;Q2. Do students attain better during certain phases of the design process during major project ?Q3. Is there a difference in this attainment between year groups of the same degree programme ?The methodology adopted employs an analysis of marks and an online questionnaire to collect data. Patterns and trends in how senior BEng and MEng Product Design Engineering students engage and attain within the design process are presented, identified and discussed and in turn used to inform reflection on the research questions set.


2020 ◽  
Vol 12 (24) ◽  
pp. 10360
Author(s):  
Hyun-Do Yun ◽  
Sun-Woong Kim ◽  
Wan-Shin Park ◽  
Sun-Woo Kim

The purpose of this study was to experimentally evaluate the effect of a hinged steel damping system on the shear behavior of a nonductile reinforced concrete frame with an opening. For the experimental test, a total of three full-scale reinforced concrete frame specimens were planned, based on the “no retrofitting” (NR) specimens with non-seismic details. The main research questions were whether the hinged steel damping system is reinforced and whether torsion springs are installed in the hinged steel damping system. From the results of the experiment, the hinged steel damping system (DR specimen) was found to be effective in seismic retrofitting, while isolating the opening of the reinforced concrete (RC) frame, and the torsion spring installed at the hinged connection (DSR specimen) was evaluated to be effective in controlling the amount of deformation of the upper and lower dampers. The strength, stiffness, and energy dissipation capacity of the DSR specimen were slightly improved compared to the DR specimen, and it was confirmed that stress redistribution was induced by the rotational stiffness of the torsion spring installed in the hinge connection between the upper and lower frames.


2019 ◽  
Vol 8 (1) ◽  
pp. 50-67
Author(s):  
Lukasz Damurski ◽  
Jacek Pluta ◽  
Jerzy Ładysz ◽  
Magdalena Mayer-Wydra

Services originally developed as natural concentrations of human activity, reflecting the Christallerian hierarchy of central places. Today, those natural mechanisms are challenged by strong competition from online facilities. More and more services are offered by the internet and this affects the traditional ‘bricks-and-mortar' urban development. In this article, the main research problems of the inter-relatedness of real and virtual environments are defined in the context of urban neighbourhood service centres. The process of conversion from offline services into online ones is treated as a canvas for building a comprehensive research model for studying the development of the contemporary urban services sector in the local scale. Particular research questions and hypotheses are formulated and followed by a set of methods for further empirical research.


Symmetry ◽  
2020 ◽  
Vol 12 (11) ◽  
pp. 1768
Author(s):  
R. Nemec ◽  
A. Jahodova Berkova ◽  
S. Hubalovsky

This article describes the research results aimed at distance education during the Covid-19 pandemic and closing schools and its symmetry with the classical state in terms of time, difficulty, and the mental and physical condition of students. An important aspect is therefore to maintain the symmetry of attitudes to teaching in face-to-face form and distance form. In terms of the eight-year gymnasium in the Czech Republic, students’ attitudes to the teaching subject informatics were investigated. The main research questions in our study dealt with whether students felt equally balanced regarding the amount of tasks and time taken for home preparation during the Covid-19 outbreak compared with the time before the quarantine and their condition (both mental and physical) during the Covid-19 outbreak. The research was conducted using an anonymous questionnaire, which was answered by 110 out of 180 students. According to the results, it is evident that students felt that during the distance education, there are more tasks compared to face-to-face ones. Students also claimed to spend more time learning at distance education than at school. On the other hand, they agreed that the self-education schedule is suitable for them. In terms of the questionnaire, their condition (both mental and physical) was also evaluated, which was slightly above the average.


2021 ◽  
Vol 72 ◽  
pp. 1385-1470
Author(s):  
Alexandra N. Uma ◽  
Tommaso Fornaciari ◽  
Dirk Hovy ◽  
Silviu Paun ◽  
Barbara Plank ◽  
...  

Many tasks in Natural Language Processing (NLP) and Computer Vision (CV) offer evidence that humans disagree, from objective tasks such as part-of-speech tagging to more subjective tasks such as classifying an image or deciding whether a proposition follows from certain premises. While most learning in artificial intelligence (AI) still relies on the assumption that a single (gold) interpretation exists for each item, a growing body of research aims to develop learning methods that do not rely on this assumption. In this survey, we review the evidence for disagreements on NLP and CV tasks, focusing on tasks for which substantial datasets containing this information have been created. We discuss the most popular approaches to training models from datasets containing multiple judgments potentially in disagreement. We systematically compare these different approaches by training them with each of the available datasets, considering several ways to evaluate the resulting models. Finally, we discuss the results in depth, focusing on four key research questions, and assess how the type of evaluation and the characteristics of a dataset determine the answers to these questions. Our results suggest, first of all, that even if we abandon the assumption of a gold standard, it is still essential to reach a consensus on how to evaluate models. This is because the relative performance of the various training methods is critically affected by the chosen form of evaluation. Secondly, we observed a strong dataset effect. With substantial datasets, providing many judgments by high-quality coders for each item, training directly with soft labels achieved better results than training from aggregated or even gold labels. This result holds for both hard and soft evaluation. But when the above conditions do not hold, leveraging both gold and soft labels generally achieved the best results in the hard evaluation. All datasets and models employed in this paper are freely available as supplementary materials.


2021 ◽  
Vol 11 (9) ◽  
pp. 1166
Author(s):  
Magdalena Bury-Kamińska ◽  
Aneta Szudy-Szczyrek ◽  
Aleksandra Nowaczyńska ◽  
Olga Jankowska-Łęcka ◽  
Marek Hus ◽  
...  

The paper presents a study on the changes in cognitive functioning in patients undergoing chemotherapy with diagnosed multiple myeloma (MM). The aim of the study was to answer the following two main research questions: Does the treatment stage differentiate the functioning of cognitive processes in patients with diagnosed MM and to what extent? Is it possible to treat biological factors (TNF-α, IL-6, IL-10, and BDNF) as predictors of patients’ cognitive functioning? The patients were examined twice, before the treatment and after 4–6 cycles of chemotherapy. Selected neuropsychological research methods as well as experimental and clinical trials were employed to diagnose the patients’ general cognitive state, attention, memory, and executive functions. The level of biological factors was assessed with the ELISA test. The results show that the patients’ cognitive functioning was worse before the treatment than during the cytostatic therapy. It was also possible to predict the cognitive state of patients suffering from multiple myeloma based on a selected biological parameter (neurotrophin BDNF).


2018 ◽  
Vol 4 (1) ◽  
pp. 2-12
Author(s):  
Anna Bargagliotti ◽  
Dorothea Herreiner ◽  
Jefrey A. Phillips

The April 2017 National Science Foundation-funded Breaking the Boundaries in STEM Education conference brought together Southern California science, technology, engineering and mathematics (STEM) faculty to explore equity, problem-solving, and computing in an interdisciplinary manner. Two main research questions guided the overall scope of the conference: (1) What are the common threads across disciplines to approach the teaching and learning of skills that are relevant in STEM? (2) What are the challenges and barriers that need to be overcome in order to foster collaboration across disciplines to impact the teaching and learning of skills relevant in STEM? We describe the background of the conference and provide an overview of the questions addressed.


2021 ◽  
Vol 58 (1) ◽  
pp. 129-139
Author(s):  
Otabek Narziev

This paper provides the necessary information and analysis for understanding and considering the main research questions and discussions of the research. Notably, this section outlines the background to capital market formation and development in CIS countries through a brief history of the CIS; considers the necessity of capital market and its regulation in CIS countries; reviews the institutional and legal framework of capital market regulation, and analyzes certain problems of capital market development.


Sign in / Sign up

Export Citation Format

Share Document