Repositório UFSCar

For more information, visit: RI UFSCar

Instructions for authors: Click here

Explore by document type

Discover UFSCar’s academic and scientific output

Communities

Select a community to browse its collections.

View all

Recent Submissions

  • Thumbnail Image
    Item type:Item,
    Visual Vernacular e os 12 princípios da animação: um estudo sobre a obra bolinha de ping pong
    (Universidade Federal de São Carlos, 2026-08-30)
    Carvalho, Leonardo Cisoto Barbosa de
    ;
    Silva, Anderson Marques da
    ;
    http://lattes.cnpq.br/7112238825025462
    ;
    https://orcid.org/0009-0003-1873-5654
    ;
    Vasconcelos, Stephanie Caroline Alves
    Visuality constitutes a central element in the construction of meaning in both sign languages and animation. From this perspective, this study investigates the connections between the Visual Vernacular (VV) genre, an artistic expression found in sign languages, including Brazilian Sign Language (Libras), and the twelve principles of animation proposed by Thomas and Johnston (1995). The study is based on the hypothesis that both share strategies for constructing movement, narrative, and expressiveness, contributing to the understanding of Libras as a visual, discursive, and artistic language. This is a qualitative, exploratory study developed through a literature review and an interpretative analysis of the work Bolinha de Ping-Pong, by Rimar R. Segala. Initially, a literature review was conducted to identify studies related to the topic and to establish the analytical categories. Subsequently, the video was examined through repeated observations, documenting occurrences of the animation principles in dialogue with the Visual Vernacular categories proposed by Monteiro (2023). The analyses were discussed collaboratively with the research supervisors and organized into three categories: Physical Aspects, Temporal Aspects, and Affective Aspects, established according to semantic similarities between the animation principles and the expressive resources of Libras. The findings demonstrate that elements such as imagery descriptions, non-manual expressions, spatial organization, rhythm, and shifts in perspective correspond to classical animation principles, broadening the understanding of visual expressiveness in Libras.
  • Thumbnail Image
    Item type:Item,
    Aprendizado positivo e não rotulado guiado por densidade
    (Universidade Federal de São Carlos, 2026-10-02)
    Castro Neto, Carlos Pereira de
    ;
    Naldi, Murilo Coelho
    ;
    http://lattes.cnpq.br/0573662728816861
    ;
    https://orcid.org/0000-0002-3107-8236
    ;
    http://lattes.cnpq.br/6844472233094299
    Positive-Unlabeled Learning is a special case of semi-supervised learning in which only a few examples of the positive class are labeled, while the rest remain unlabeled, comprising both positive and negative instances. The two-step strategy, widely adopted in the literature, consists of extracting a subset of reliable negative examples from the unlabeled data and subsequently training a classifier based on them. This work proposes a modular framework based on density-based hierarchical clustering for the extraction of reliable negatives and for a novel transductive semi-supervised classification strategy. The approach is founded on the exploration of the data's topological structure and on density-based probability estimates. To extend the method to data represented as graphs, the strategy is combined with latent representation learning techniques. An adaptation focused on topology preservation is proposed for scenarios where connected nodes tend to be similar (homophily), alongside a novel spatially guided metric learning architecture, developed in this work for networks where connections between distinct classes are predominant (heterophily). The experimental validation encompassed several datasets, contrasting the method with recent state-of-the-art approaches. The findings corroborate the premise that density in the latent space is a strong indicator of class separability in the absence of negative labels, ensuring the proposed approach achieves competitive, and sometimes superior, performance in both negative extraction and classification across tabular data and homophilic graphs. The scenario proves more challenging in strongly heterophilic networks, where the developed model preserves a good capacity for negative extraction but encounters limitations regarding the complexity of semi-supervised propagation during the final classification. This dynamics in complex graphs, coupled with the model's sensitivity to structural parameters, outlines the current application limits of the framework and points to natural directions for future work.
  • Thumbnail Image
    Item type:Item,
    Modelagem da informação tecnológica baseada em patentes para a inovação
    (Universidade Federal de São Carlos, 2026-07-31)
    Zaniro, Dênis Leonardo
    ;
    Faria, Leandro Innocentini Lopes de
    ;
    http://lattes.cnpq.br/0767710394930118
    ;
    https://orcid.org/0000-0002-8369-1315
    ;
    Quoniam, Luc
    Innovation has been researched and practiced in multiple ways in recent years and, despite certain nuances, this concept has evolved beyond the simple idea of introducing something new into the economic context to reflect the complexity involved throughout the entire innovation process. The literature also indicates that innovation processes are affected by different factors and depend on the availability of technological information. The primary source of technological information for innovation comes from patent documents; however, there are two major challenges to consider. The first is that the information describing the inventions behind patents may be difficult for different audiences to understand and does not represent the full complexity of technological information which, by definition, is multidimensional. The second is that information sources that may serve as inputs for innovation are usually disconnected and involve considerable volumes of data, creating obstacles for analytical strategies that are not based on computational approaches. Furthermore, certain patent and inventive databases are paid services, hindering access for communities characterized by scarce resources. Based on this context, the main objective of this research was to model technological information aimed at innovation in such a way that, within the resulting dataset, four informational perspectives coexist inseparably: inventive, encyclopedic, scientific, and industrial. An informational architecture is proposed, guided by a reference model and constituted through the integration of databases, classification schemes, and an inventive methodology. Multiple research methods were adopted and combined, with a predominantly exploratory objective and both quantitative and qualitative approaches. Due to the volume of data, computational techniques grounded in a set of algorithms were applied. For the construction of the informational model, the following components were used: the TRIZ methodology (the Russian acronym for Theory of Inventive Problem Solving) and TRIZ effects databases; the International Patent Classification; Wikipedia and Wikidata; the Lattes Curriculum; the National Classification of Economic Activities; and the Espacenet database. As a result, the resulting dataset provides millions of potential pathways connecting researchers, new technologies, and industrial sectors in Brazil. The use cases discussed provide evidence that, by integrating fragmented data from distinct sources into a single whole, according to the proposed methodological framework, it is possible to arrive at certain inventive insights that could not easily be mapped through the simple combination of the individual parts. Specifically, regarding the scientific perspective, a certified and expanded patent database derived from the Lattes Curriculum was achieved, overcoming the unitary patent-based perspective in order to attain a broad view of inventions necessary for integration into the model. From the STS (Science, Technology, and Society) perspective, this research is part of a context shaped by social demands and market dynamics under a constructivist approach, representing only one part of a larger and unfinished system and, therefore, subject to continuous evolution. Thus, it is considered that the results obtained and their subsequent developments may contribute to strengthening the connections between research, technology, and industry in Brazil.
  • Thumbnail Image
    Item type:Item,
    Agroecologia, autoconsumo e reprodução social: segurança e soberania alimentar em assentamentos rurais do interior paulista
    (Universidade Federal de São Carlos, 2026-10-01)
    Veron, Ana Claudia Olartechea
    ;
    Carvalho, Joelson Gonçalves de
    ;
    http://lattes.cnpq.br/5680333699420527
    ;
    http://lattes.cnpq.br/0141039743607067
    ;
    https://orcid.org/0000-0002-8591-0446
    Este artigo analisa as relações entre agroecologia, segurança alimentar e soberania alimentar nos assentamentos PDS Santa Helena, em São Carlos (SP), e PDS Mário Lago, em Ribeirão Preto (SP). O objetivo é compreender em que medida práticas agroecológicas, articuladas ao acesso à terra, ao autoconsumo, à diversificação produtiva e à comercialização de excedentes, contribuem para fortalecer a reprodução social das famílias assentadas. A pesquisa, de abordagem qualitativa e caráter exploratório, combinou revisão bibliográfica, entrevistas semiestruturadas, roda de conversa, observação participante e aplicação complementar da Ebia. Os resultados indicam que a agroecologia amplia a disponibilidade e a qualidade dos alimentos, favorece maior autonomia alimentar e fortalece circuitos curtos de comercialização. Contudo, essas potencialidades permanecem condicionadas por entraves estruturais, como dificuldades de acesso à água, crédito, assistência técnica, transporte e canais estáveis de comercialização.
  • Thumbnail Image
    Item type:Item,
    Leitura e compreensão do texto videossinalizado no ensino de Libras como língua adicional
    (Universidade Federal de São Carlos, 2026-09-22)
    Martins, Keylliane de Sousa
    ;
    Cruz, Carina Rebello
    ;
    http://lattes.cnpq.br/4701066373682779
    ;
    https://orcid.org/0000-0002-5430-2130
    ;
    Zuin, Poliana Bruno
    Brazilian Sign Language (Libras), due to its visuospatial modality, requires learners to engage their visual and motor skills to understand and produce signs. This study aimed to analyze the behavior of readers who are non-deaf listeners learning Libras as an additional language when engaging with video-recorded signed texts, focusing on whether they used strategies to support their reading and comprehension of the content. To achieve this objective, digital teaching materials were developed and used with undergraduate students enrolled in a Portuguese Language and Literature teacher education program at a public university who were taking a Libras course. The following specific objectives were established: to investigate the reading and comprehension processes of learners studying Libras as an additional language; to determine whether modality effects in video-recorded signed texts interfere with the reading and comprehension process; to analyze whether the fable and poetry genres, in the visuospatial modality, contain elements that facilitate reading and comprehension; and to investigate whether readers use strategies, and which ones they employ when reading a text, in order to understand a video-recorded signed text. The theoretical framework draws on the work of Bakhtin (2010), Leffa (1996, 2016), Kato (2007), Quadros and Karnopp (2011), Solé (1998), and Tomitch (2007, 2008), among other authors. The study adopted a qualitative methodological approach, considering human beings as subjects engaged in dialogue and collaboration. The data generated through participants’ interactions with the video-recorded signed texts were analyzed using content analysis, as proposed by Bardin (2016). The findings show that participants employed different strategies in their attempts to understand the texts, including cognitive and metacognitive strategies. However, the use of these strategies alone did not ensure full comprehension of the texts. Meaning construction was also found to be related to participants’ linguistic knowledge, background knowledge, experience with Libras, level of proficiency, and the characteristics of the texts themselves.