- Sentiment Analysis and Opinion Mining
- Advanced Text Analysis Techniques
- Digital Communication and Language
- Spam and Phishing Detection
- Topic Modeling
- Multimedia Communication and Technology
- Tactile and Sensory Interactions
- Mental Health via Writing
- Natural Language Processing Techniques
- Hate Speech and Cyberbullying Detection
- Emotion and Mood Recognition
- Power Line Communications and Noise
- Complex Network Analysis Techniques
- Social Media and Politics
- E-Learning and Knowledge Management
- Multisensory perception and integration
- Gender, Sexuality, and Education
- Online Learning and Analytics
- Text and Document Classification Technologies
- Personality Traits and Psychology
- Interactive and Immersive Displays
- Misinformation and Its Impacts
- ICT in Developing Communities
- Advanced Clustering Algorithms Research
- Imbalanced Data Classification Techniques
Federal Center for Technological Education Celso Suckow da Fonseca
2015-2024
Federal Center for Technological Education of Minas Gerais
2020-2024
Universidade Presidente Antônio Carlos
2024
Meta (Israel)
2020
Instituto Federal do Maranhão
2019
Universidade Federal do Rio de Janeiro
2015-2016
Happiness at Work is considered the Holy Grail of organizational sciences. The belief that happier workers are more productive leads to a win-win situation for both individuals and organizations. Nevertheless, years research have not brought convergent conclusion about topic, mainly due lack widely accepted measure. Usually, questionnaires self-report surveys used; however, these methods embed shortcomings allow studies’ results be questioned. In order overcome shortcomings, present study...
With the changes in human interaction prompted by development of communications platforms over internet, hate speech and offensive language emerged as a contemporary problem. Social networks allow users with different opinions backgrounds to interact without direct eye-to-eye contact. It brings sense safety promote speech, which is even more significant anonymous environments. There are sites called imageboards, composed boards aggregating topics. On some boards, widely speech. However, only...
LIWC is a text analysis program that categorizes words into gram- matical and psychologically derived categories. The currently available lexicon for Brazilian Portuguese (LIWC 2007pt) based on the 2007 version of program. As several studies indicated, 2007pt shows perfor- mance categorization problems. In this scenario, work highlights new 2015pt), 2015 This compares performance 2015pt in classification tasks. Three experiments were conducted results indi- cate outperforms all three
The Moral Foundations Theory defines foundations to explain human moral reasoning and its role in the decision-making process, including how information is perceived interpreted. A problem related aspects of values that currently gaining notoriety spread false known as "Fake News". Natural language processing techniques are being used social sciences studies deal with Fake News detection task. This work introduces brings details from development MFD-BR, a Brazilian Portuguese lexicon based...
Segundo dados do IBGE, as mulheres correspondem a 51,03% da população brasileira. Apesar disso, apenas 23% total de pesquisadores na área Computação são gênero feminino. Diversos fatores sociais parecem ser responsáveis por essa lacuna, porém algumas ações que visam atrair para informática vêm surgindo fim mitigar questão. Diante dessa disparidade, este estudo tem como objetivo comparar representatividade autoria artigos publicados em eventos Congresso Sociedade Brasileira (CSBC) nos últimos...
Emotion detection is crucial in Human-Computer Interaction (HCI). Computers can never respond to individuals affective state if it cannot identify its emotion. Categorical approaches are commonly used for emotion texts. It considers training models based on a set of states. often dictionary that contains words associated with emotional categories. In this scenario, LIWC dictionaries have been widely psychology and linguistics. There many languages, including Brazilian Portuguese...
Time series prediction has been attracting interest of researchers due to its increasing importance in decision-making activities many fields knowledge. The demand for better accuracy time furthered the arising machine learning methods (MLM). Choosing a suitable method particular dataset is challenge and demands established benchmark (BM) performance assessment. Suppose BM selected, an experimental comparison made with MLM. If latter does not provide results same dataset, this indicates that...
Electronic books (e-books) are an alternative to traditional and have some advantages over it. For example, they present dictionary function, allow searching for words in texts, ecologically correct, among others. Moreover, research indicates e-books can improve the learning process. The process is also improved with use of enhanced e-books, which includes multimedia resources (e.g., images, videos, sounds) stimulate adapted interactive learning. Similarly, multisensorial media (MulSeMedia)...
Abstract Objectives Malaria is an infectious disease that annually presents around 200,000 cases in Brazil. The availability of data on malaria crucial for enabling and supporting studies can promote actions to prevent it. Therefore, the goal this paper contribute such by offering integrated dataset containing reported suspected Brazilian Legal Amazon comprising period from years 2009 2019. Data description This a with all medical records patients who were tested has 40 attributes 22,923,977...
For recent or planned deep astronomical surveys, it is important to tell stars and galaxies apart, a task known as Star/Galaxy Separation Problem (SGSP). At faint magnitudes, the separation between pointy extended sources fuzzy, which makes SGSP hard task. This problem even harder for large surveys like Dark Energy Survey (DES) and, in near future, Large Synoptic Telescope (LSST) due their data volume. Hence, search classification methods that are both accurate efficient highly relevant. In...
Augmented e-books provide the reader with a better quality of experience (QoE) by enriching multimedia content and interactive elements. In field, an improve in QoE has been performed use sensory effects, creating so-called mulsemedia applications. this paper, it is considered that augmented influence perception history environment, besides improving its QoE. It presented setup light, wind audio which are synchronized written history. Among other discoveries, collected data suggest such...
Sentiment Analysis is a branch of Affective Computing usually considered binary classification task. In this line reasoning, can be applied in several contexts to classify the attitude expressed text samples, for example, movie reviews, sarcasm, among others. A common approach represent samples use Vector Space Model compute numerical feature vectors consisting weight terms. The most popular term weighting scheme TF-IDF (Term Frequency - Inverse Document Frequency). It an Unsupervised...
Existem mais de 300 milhões indivíduos que sofrem depressão no mundo. Dentre as diversas consequências provenientes desse transtorno, a catastrófica é o suicídio. Nesse aspecto, diversos manifestam em redes sociais os sinais depressivos antes chegar ao extremo do cenário, presente estudo tem objetiva construir um conjunto dados palavras relacionadas à para língua portuguesa. Experimentos preliminares indicam criado auxilia na classificação emoções textos uma rede social brasileira.
Affective computing devices have been used in e-learning, game, and movie areas to acquire individuals' affective states. Thus, states like fun, boredom, frustration, excitement can be detected. Based on these states, many tasks executed. For example, if an individual practicing exercise e-learning system is bored, some action executed by the system; for it could present more challenging tasks. Inspired above scenario, work proposes creation of a model based e-books reading, multisensorial...
Multisensorial books enrich a story with either traditional multimedia content or sensorial effects. The main idea is to increase children’s interest in reading by enhancing their QoE while reading. Studies on enriched and/or augmented e-books also propose synchronizing additional text. However, they usually focus audio, vídeo, images, haptic feedback. In this work, we present MBook , tool for presenting multisensorial books. It decouples the book’s textual from content, as well its...
Studies in human-computer interaction literature concentrate on the problem of extracting affective states from text. However, there is a lack lexicons for Brazilian Portuguese. This study presents an algorithm to create hybrid lexicon The result based LIWC and ANEW-Br, which are available literature. experimental results this work indicate potential produced lexicon.
De acordo com a UNESCO, ao direcionarmos o foco para âmbito da computação, torna-se ainda mais evidente sub-representação das mulheres, tanto no cenário acadêmico, quanto profissional: apenas 5% dos profissionais de desenvolvimento software são mulheres e em 61 115 países as graduadas Ciência Computação representam menos 30% formandos. Assim, desigualdade gênero nas áreas Tecnologia é perceptível eventos ou competições do ramo. Nesse contexto, este trabalho objetiva realizar uma análise...
Este trabalho propõe uma metodologia para detecção e intervenção nos casos de manifestações depressivas em redes sociais, utilizando Computação Afetiva técnicas Aprendizagem Máquina. Baseado nessa metodologia, desenvolveu-se o protótipo SaveMeNow, que pode monitorar plataformas como Facebook Twitter, classificando mensagens potencialmente direcionando-as profissional. O sistema é dividido dois módulos principais: Monitoramento Atendimento, detalhados com suas respectivas tecnologias...
Transtornos psiquiátricos frequentemente apresentam manifestações cutâneas, uma vez que a conexão entre mente e pele é amplamente reconhecida na literatura científica. Este estudo teve como objetivo identificar analisar os transtornos mais prevalentes associados cutâneas compreender suas inter-relações clínicas. Trata-se de revisão integrativa, conduzida nas bases dados Scientific Electronic Library Online (SCIELO) Literatura Latino-Americana do Caribe em Ciências da Saúde (LILACS),...
This work aims at creating a tool for analyzing the psychological and linguistic changes of texts translated from English into Brazilian Portuguese. The aim is to analyze differences between by automatic translators human translators. For this purpose, named LIWC used in its Portuguese version. that distributes lexical words categories with characteristics. Through accounting word categories, seeks evaluate percentage psycholinguistic automatically documents comparison reliable translation...
Changes in our society have been contributing to the raising number of people who will be under effects night sleep deprivation. This can harms health physical, mental and social aspects. research work is intended evaluate if there are differences between affective characteristics entries posted two different time intervals, analyzing psycholinguistic processes dominant emotions associated network users, using LIWC portuguese. In this preliminary study, it was possible show a small...
A depressão é uma doença que vem tomando proporções preocupan- tes em todo o mundo. Segundo a Organização Mundial de Saúde, mais 300 milhões pessoas sofrem deste mal. Com popularização e facilidades acesso à Internet nos dias atuais, comum encontrar indivı́duos depressivos que, na tentativa aliviar seus sintomas, acabam expondo sentimentos redes sociais antes chegar ao extremo do suicı́dio. Este trabalho propõe uso um léxico para aprimorar identificação sintomas brasileiras. Experimentos...