A5075685976- Opinion Dynamics and Social Influence
- Online Learning and Analytics
- Human Pose and Action Recognition
- Smart Agriculture and AI
- Complex Network Analysis Techniques
- Evolutionary Game Theory and Cooperation
- Gait Recognition and Analysis
- Robot Manipulation and Learning
- Advanced Neural Network Applications
- Video Surveillance and Tracking Methods
- Digital Mental Health Interventions
- Recommender Systems and Techniques
- Open Education and E-Learning
- Experimental Behavioral Economics Studies
- Impact of Technology on Adolescents
- Domain Adaptation and Few-Shot Learning
- Social Media and Politics
- Manufacturing Process and Optimization
- Software Engineering Techniques and Practices
- Soft Robotics and Applications
- Date Palm Research Studies
- Language and cultural evolution
- Mobile Crowdsensing and Crowdsourcing
- Suicide and Self-Harm Studies
- Auction Theory and Applications
Universidade Federal de Pelotas
2014-2024
Centro Universitário de Caratinga
2024
Instituto Federal Sul-rio-grandense
2019
Polytechnic Institute of Viana do Castelo
2018
University of Coimbra
2007-2013
Universidade Federal de Campina Grande
2010
Universidade Federal do Rio Grande do Sul
2008
This work provides a study on how Convolutional Neural Networks, trained to identify objects primarily in photos, perform when applied more abstract representations of the same objects. Our main goal is better understand generalization abilities these networks and their learned inner representations. We show that both GoogLeNet AlexNet are largely unable recognize sketches easily recognizable by humans. Moreover, we measured efficacy vary considerably across different classes discuss...
This paper studies the use of Convolutional Neural Networks to automatically detect and classify diseases, nutritional deficiencies damage by herbicides on apple trees from images their leaves. task is fundamental guarantee a high quality resulting yields currently largely performed experts in field, which can severely limit scale add costs. By using novel data set containing labeled examples consisting 2539 6 known disorders, we show that trained are able match or outperform this task,...
Uniquely identifying individuals using anthropometric and gait data allows for passive biometric systems, where cooperation from the subjects being identified is not required. In this paper, we report on experiments a novel set composed of 140 walking in front Microsoft Kinect sensor. We provide methodology to extract features show results applying different machine learning algorithms subject identification tasks. Focusing KNN classifiers, discuss how accuracy varies settings, including...
In Twitter, a retweet happens when user copies another user's tweet and indicates that the content is originally from someone else, common way to share information. This paper explores user’s perceptions of social capital influences practices. We discuss how Twitter users perceive values they create benefits for network themselves three points view: referrals, information access timing. do so through qualitative analysis questionnaires answered by four quantitative case studies retweets....
Contemporary education is a vast field that concerned with the performance of systems. In formal e-learning context, student dropout considered one main problems and has received much attention from learning analytics research community, which reported several approaches to development models for early prediction at-risk students. However, maximizing results obtained by predictions considerable challenge. this work, we developed solution using only students’ interactions virtual environment...
In this paper, we consider the viability of using Microsoft Kinect sensor to extract skeleton points from walking subjects and use these for biometric identification. We do so by capturing several sensor, calculating length body parts inferred extracted training a model later classification lengths labels identifying as examples. cases where one wants discriminate each subject individually only recognizing single is enough, showing that in both Nearest Neighbor algorithm able achieve high...
In this article, we discuss the communicative functions of hashtags during a period major social protests in Brazil. Drawing from theoretical background use Twitter and language, extracted sample 46,090 2,321,249 tweets related to Brazilian June 2013. We analyzed through content analysis, focusing on functions, co-occurrences. also qualitatively group 500 most retweeted understand users’ tagging behavior. Our results show how users appropriate tags accomplish different effects narrative protests.
Purpose The purpose of this paper is to present a CAD‐based human‐robot interface that allows non‐expert users teach robot in manner similar used by human beings each other. Design/methodology/approach Intuitive programming achieved using CAD drawings generate programs off‐line. Sensory feedback minimization the effects uncertainty, providing information adjust paths during operation. Findings It was found it possible program from common drawing and run without any major concerns about...
Purpose The purpose of this paper is to report a collection developments that enable users program industrial robots using speech, several device interfaces, force control and code generation techniques. Design/methodology/approach reported system explained in detail few practical examples are given demonstrate its usefulness for small medium‐sized enterprises (SMEs), where humans need cooperate achieve common goal (coworker scenario). also explores the user interface software adapted use by...
Mental health status assessment is mostly limited to clinical or research settings, but recent technological advances provide new opportunities for measurement using more ecological approaches. Leveraging apps already in use by individuals on their smartphones, such as chatbots, could be a useful approach capture subjective reports of mood the moment.
This work investigates the use of 3D skeleton data retrieved from Microsoft Kinect sensor to obtain gait kinematic parameters human walk for biometric identification. Several subjects were captured walking in front resulting a set containing several points joints each person. These used extract anthropometric information and calculate angles described by lower (hips, knees ankles) during order parameters. Statistical descriptors curves define attributes train machine learning models...
Force/torque sensing is very important for several automatic and industrial robotic applications. Basically, if precise control of the forces that arise from contact between tools parts required to successfully complete task, then a force/torque sensor needed along with some technique. In this paper we focus on aspects applied tasks. Concentrating particular type sensor, demonstrate how use them integrate into applications using robots. Finally, an application presented where force was...
We provide a study of crowdsourced design based on the popular service 99designs, website where users create contests for other to submit their ideas evaluation, competing monetary amount offered. By quantitatively analyzing large number contests, we report dynamics interactions between and designers. show results effects different financial incentives contests' properties quality submissions, as well how rewards are distributed across find that higher do not translate more effort by...
The new ways of doing science, rooted on the unprecedented processing, communication and storage infrastructures that became available to scientists, are collectively called e-Science. Many research labs now need non-trivial computational power run e-Science applications. Grid voluntary computing well-established solutions cater this need, but not accessible for all institutions. Besides, there is an uncertainty about future amount resources will be in such infrastructures, which prevents...
Identifying individuals using biometric data is an important task in surveillance, authentication and even entertainment. This more challenging when required to be performed without physical contact at a distance. Analyzing video footages from for patterns active area of research aiming fulfilling this goal. We describe results on classifiers trained identify collected 140 subjects walking front Microsoft Kinect sensor, which allows tracking 3D points representing subject's skeleton. From we...
To present the protocol and methods for prospective longitudinal assessments-including clinical digital phenotyping approaches-of Identifying Depression Early in Adolescence Risk Stratified Cohort (IDEA-RiSCo) study, which comprises Brazilian adolescents stratified at baseline by risk of developing depression or presence depression.
A cirurgia, como uma jornada épica, exige preparo meticuloso para enfrentar os desafios que se apresentam. avaliação pré-operatória, nesse contexto, assume o papel de bússola crucial, guiando a equipe médica na tomada decisões precisas e prevenção complicações pós-operatórias, garantindo um retorno tranquilo seguro à vida cotidiana. pré-operatória configura luz ilumina riscos potenciais cada paciente enfrenta antes, durante após cirurgia. Através análise abrangente do histórico médico,...
The ability to identify urban locations with a high risk of diseases infections is central aspect public policies aiming at controlling these diseases. In this paper, we investigate the use street-level images, such as those from Google Street View, along Convolutional Neural Networks predict Dengue Fever (DF) and Hemorrhagic (DHF) rates in locations. We conduct case study city Rio De Janeiro, Brazil, using proposed methodology DF/DHF data between years 2010 2014. compare two Siamesebased...
Context: Climate variability plays a fundamental role in shaping agricultural outcomes, influencing crop growth, yield, and resource management strategies. Problem: Identifying meaningful patterns meteorological datasets to optimize practices is challenging complex task. Solution: This paper explores the use of clustering techniques on SIMAGRO-RS weather data uncover for applications. IS Theory: study addresses challenges Information Systems applied Agriculture, especially related General...
Purpose – Interaction with robot systems for the specification of manufacturing tasks needs to be simple since paper targets widespread use robots in small and medium enterprises (SMEs). In best case, existing practices from manual work could used, ensure current employees a smooth introduction technology as natural part their work. The aim is simplify robot‐programming task by allowing user simply make technical drawings on sheet paper. Craftsmen raw sketches several situations: share...