Brazil

O processo de roteamento naInternet of Things(IoT) é um aspectochave quando dispositivos podem mover. O roteamento habilita a co-municação entre os dispositivos, potencializando a funcionalidadedos objetos do dia-a-dia. Contudo, a mobilidade causa frequentesmudanças na topologia tornando o roteamento um desafio. A maio-ria dos protocolos de roteamento da IoT usam um estimador dequalidade de enlace (Link Quality Estimator(LQE)) no processo deescolha de rotas, sendo oExpected Transmission Count(ETX) umdos mais utilizados para ponderar enlaces sem fio entre dispositivosestáticos. Contudo, o ETX não funciona bem quando os dispositivosse movem. Neste artigo, apresentamos oMobility aware ExpectedTransmission Count(MobETX) como estimador para enlaces semfio entre dispositivos móveis. MobETX mantém as característicasdo ETX, mas usa métricas de mobilidade dos dispositivos para darassistência na escolha de rotas. Em nossos experimentos, o pro-tocolo de roteamento usando MobETX apresentou melhora de aomenos5%na taxa de entrega de pacotes quando comparado comETX em cenários de alta mobilidade e&gt;98%de entrega de pacotesem cenários estáticos.

WebMedia 2021

Uma métrica de roteamento para Internet das Coisas Móveis

The 9th Brazilian Conference on Intelligent Systems (BRACIS), the 16h Encontro Nacional de Inteligência Artificial e Computacional (ENIAC), and the 12th Concurso de Teses e Dissertações em Inteligência Artificial e Computacional (CTDIAC) was held from 20 to 23 October, 2020.

BRACIS originated from the combination of the two most important scientific events in Brazil in Artificial Intelligence (AI) and Computational Intelligence (CI): the Brazilian Symposium on Artificial Intelligence - SBIA (22 editions), and the Brazilian Symposium on Neural Networks - SBRN (13 editions). ENIAC is a national forum for researchers, practitioners, educators and students to present and discuss innovations, trends, experiences, developments and work in progress in the fields of Artificial and Computational Intelligence. CTDIAC is a satellite event whose goals are to disseminate and award the best dissertations and theses (MSc and PhDs concluded in Brazilian universities from June 2018 to May 2020) in Artificial and Computational Intelligence areas.

In this year, the Symposium on Knowledge Discovery, Mining and Learning (KDMiLe) will be held in conjunction with BRACIS 2020. This sysposium aims at integrating researchers, practitioners, developers, students and users to present their research results, to discuss ideas, and to exchange techniques, tools, and practical experiences – related to the Data Mining and Machine Learning areas.

BRACIS 2020

O processo de roteamento naInternet of Things(IoT) é um aspectochave quando dispositivos podem mover. O roteamento habilita a co-municação entre os dispositivos, potencializando a funcionalidadedos objetos do dia-a-dia. Contudo, a mobilidade causa frequentesmudanças na topologia tornando o roteamento um desafio. A maio-ria dos protocolos de roteamento da IoT usam um estimador dequalidade de enlace (Link Quality Estimator(LQE)) no processo deescolha de rotas, sendo oExpected Transmission Count(ETX) umdos mais utilizados para ponderar enlaces sem fio entre dispositivosestáticos. Contudo, o ETX não funciona bem quando os dispositivosse movem. Neste artigo, apresentamos oMobility aware ExpectedTransmission Count(MobETX) como estimador para enlaces semfio entre dispositivos móveis. MobETX mantém as característicasdo ETX, mas usa métricas de mobilidade dos dispositivos para darassistência na escolha de rotas. Em nossos experimentos, o pro-tocolo de roteamento usando MobETX apresentou melhora de aomenos5%na taxa de entrega de pacotes quando comparado comETX em cenários de alta mobilidade e>98%de entrega de pacotesem cenários estáticos.

technical paper

### Welcome to the Brazilian Symposium on Multimedia Systems and Web 
Promoted annually by the Brazilian Computer Society (SBC), the Brazilian Symposium on Multimedia Systems and Web (WebMedia) is the main event on the subject in Brazil and an excellent opportunity for scientific and technical exchanges between students, researchers and professionals in the areas of Multimedia, Hypermedia and Web. This year, we will be celebrating the 27th edition of WebMedia, to be held from **November 5th to 12th, 2021**, organized by the Federal University of Minas Gerais and Federal University of São João Del Rei. For the second time in its history, the event will be held online, with interaction between the participants being done through tools and web platforms.
 <iframe width="560" height="315" src="https://www.youtube.com/embed/2GvWKRUDtAY" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe> 
The Webmedia 2021 schedule will feature:
 
* Technical sessions for the presentation of scientific articles.
* National and international lectures
* Panel with Companies
* Workshops
* Short courses
* Project meetings open to the community
 

The topics of interest for paper submission are:
 
* Web and Social Networks
* Multimedia
* Digital TV, Ubiquitous and Mobile Computing
 

In addition to the main track, WebMedia 2021 will feature the Tools and Applications Workshop, the Scientific Initiation Papers Contest, the Theses and Dissertations Contest, the RNP CT-Video Workshop.
 
The organization of WebMedia 2021 is under the responsibility of the [Department of Computer Science at UFMG (DCC/UFMG)](http://dcc.ufmg.br/) and [Department of Computer Science at UFSJ (DCOMP/UFSJ)](https://dcomp.ufsj.edu.br/)
 
**#PartiuWebMedia2021 #WebMedia2021**

The Brazilian Symposium on Multimedia Systems and Web (WebMedia) is the main event on the subject in Brazil and an excellent opportunity for scientific and technical exchanges between students, researchers and professionals in the areas of Multimedia, Hypermedia and Web.

A considerable part of the culture and behavior of societies arederived from the habits and preferences built up over time. One rep-resentative characteristic that a group can present is the preferencefor certain food groups, thus, building the gastronomic identity ofeach region around the world. With the increasingly broad con-nections established by social networks, it is now more feasibleto analyze such preferences on a large scale. This study examinesrecipes from Allrecipes.com network in three continents: Amer-ica, Europe, and Asia. Based on the evaluations made by the users,a score was developed, allowing the separation of the recipes intwo broad groups: well evaluated and poorly evaluated. All theingredients of these recipes were extracted and used to assemble anetwork whose links were made via pointwise mutual information.This measure of association, used in pairs of ingredients, allowedus to find the main ingredients common to the countries. Our studymay help to better understand the success, or otherwise, of a recipe,in a specific locality, based on its main ingredients. Thus, one ofthe main utilities envisioned for this work is to establish betterrecommendations for recipes.

Estudo de Preferências por Receitas do AllRecipes pelo Mundo

Context-aware recommendation systems use contextual data to rec-ommend a fully personalized suggestion to their users, for instance,using the massive workloads produced by the usage of mobile apps.In this paper, we collected and analyzed a dataset from social me-dia related to the Belo Horizonte’s 2020 Carnival to understandhow this event attracts tourists (or Belo Horizonte’s non-resident),analyzing their interactions with a large recommendation system.We built a point-of-view of an event mixing its features in twosocial networks: Twitter and Google Review. Our results show thatcombining traits lead to better information about the given contextusing social networks. It helps both the tourists choosing where totravel and the local establishments to provide better services.

Analyzing A Touristic Event Popularity Using Social Networks

Muitas técnicas têm sido desenvolvidas em prol da vigilância au-tomatizada, desde alarmes inteligentes aos sistemas de câmeras. Aautomatização da análise de imagens em vídeos, principalmente adetecção de objetos tem crescido nos últimos tempos.Este trabalhotem por principal objetivo detectar objetos específicos para auxiliarno alerta sobre porte de armas, incêndios e pichações, em temporeal, aos órgãos da Secretaria de Segurança Pública da Bahia.

Análise experimental para a detecção de objetos em vídeos de câmeras de vigilância - Uma abordagem para porte de arma, incêndio e pichação.

Candidates and job vacancies may remain for long periods withoutreal opportunities in Recommendation Systems (RSs) for onlinerecruitment. We refer to these scenarios as theMatching ScarcityProblem(MaSP). We formalize the MaSP and propose a strategy toidentify candidates and vacancies suffering from it. We also proposefive heuristics, which suggest changes in CVs and jobs descriptions,to mitigate the MaSP. The best heuristic was able to reduce up to50% the number of CVs and jobs suffering from MaSP.

O Problema de Escassez de Matchings em Recomendações nos Domínios de Recrutamento

Wearable devices emerged from the advancement of communica-tion technology and the miniaturization of electronic components.These devices periodically monitor the user’s vital signs and gen-erally have low battery life. This thesis proposes Odin, a modelfor optimized collection of vital signs based on adaptive strategies.The comparison with related works indicates that Odin is the onlyproposal that presents an adaptive collection of vital signs, whichallows the optimized generation of context history. The researchvalidation is divided in three groups. In the first group, simulationswere performed in different scenarios, through requisition control,an increase of 114% in battery life was obtained in the adaptivescenario compared to a collection without adaptivity. In the secondgroup, the Odin model was applied to a real physiological data-base and allowed for a reduction in the number of requests due tothe adaptability of the collection. This reduction optimized batteryconsumption by 66% compared to a collection without adaptability.Finally, in the third group, the model was applied in collecting datathrough an Arduino and also a cardiac monitoring strap availableon the market (Polar H7). During this third part of the evaluation,savings of 18% were observed in the power consumption of themobile devices involved in the data collection.

Odin: Um modelo de coleta adaptativa de sinais vitais

Blockchain has been widely used to store decentralized and securetransactions involving cryptocurrency (for instance, Bitcoin andEthereum solutions). On the other hand, Smart city applicationsare concerned about how data and services can be safely storedand shared. In this regard, this research investigates the use ofBlockchain in the Smart Cities scenario. Based on that, a platformnamed Orthus was proposed to support Blockchain in Smart Cityinitiatives focused on scalability. This work presents a case studyabout how to use the proposed platform in the Natal Smart CityInitiative in Brazil to handle land registration. Moreover, it also com-pares this platform with other implementations that use Blockchainin different domains. Finally, this research confirms that the useof blockchain technology has much to contribute to Smart Citysolutions once it enables the creation of solutions in distributednetworks to meet the demand of the entire population.

Orthus: A Blockchain Platform for Smart Cities

The massive use of online social media networks is a reality nowa-days. Their increasing usage also raises growth in malicious activ-ities in social media, one of which is the use of automated users(bots) that disseminate false information and can insert bias in anal-yses done on gathered social media data. Based on the concept ofmedia literacy, this research presents a method to teach the humanuser to identify a pattern of a text produced by a bot, providinga tool (guide) to analyze social media text. Users who learned toidentify a bot user with the guide had an average of 90% accuracy inthe classification of new messages, against 57% of the participantswho had no contact with the guide. The produced guide received ausefulness rating between 4 and 5 by the participants (scale from 1to 5, with 5 being the highest value)

Pattern Identification of Bot Messages for Media Literacy

We propose tackling theHit Song Predictionproblem through amultimodal form with songs’ features fused together. Specifically,we describe songs from three feature modalities:music,artistandalbum. Initially, we identify collaboration profiles in a success-based musical network, unveiling how professional connectionscan significantly impact their success. Then, we use time seriesand the Granger Causality test for assessing whether there is acausal relationship between collaboration profiles and artists’ pop-ularity. Finally, we model theHit Song Predictionproblem as twodistinct tasks:classificationandplacement. The former is a classicalbinary classification model and directly applies our fusion strate-gies. The latter is a modeling approach that ranks a song relativeto a given chart, predicts hit songs, and provides comparative pop-ularity information of a set of songs. Furthermore, we emphasizecollaboration artists’ profiles as important features when describingtheir songs. Overall, our empirical studies confirm the effectivenessof our method that fuses heterogeneous data for both tasks.

Collaboration-Aware Hit Song Analysis and Prediction

The encoding of 360 videos presents many challenges. These spherical videos require a planar projection stage prior to encoding, and there are no studies on how projection interferes withthe encoder decisions. Also, due to the increased resolution, encoding 360 videos is computationally costlier than conventional videos. In this work, we perform an extensive evaluation of how the texture distortions caused by the projection interfere withthe encoder’s behavior and exploit these interferences to propose multiple fast algorithms to accelerate the encoding. Experimental results show that the proposed algorithms reduce encoding time significantly with minor coding efficiency penalties, while the evaluations expose novel insights into the encoder behavior. The reached results arepublished in important eventsand journals.

Exploring ERP Distortions to Reduce the Encoding Time of 360 Videos

In this work, we propose a framework that automatically extractssemantic topics from scientific publications related to research onCOVID-19. The framework has four main building blocks: (i) pre-processing, (ii) topic modeling, (iii) topic correlation with authorsand institutions, and (iv) summarization interface. The first blockcorresponds to the application of pre-processing strategies in textson the considered articles and the definition of their semantic repre-sentation. The topic modeling block aims to find semantic topics inthe publications used. The third block correlates these topics withthe articles themselves and the authors, institutions, and countriesrelated to each article. The summary interface provides an intuitiveview for all these analyses. Our evaluation shows that our frame-work is capable of automatically extracting relevant characteristicsfrom the articles, identifying the main themes addressed by them,as well as the correlation of researchers, institutions and countriesfor different topics of research on COVID-19.

Downloads

Next from WebMedia 2021

Estudo de Preferências por Receitas do AllRecipes pelo Mundo

Similar lecture

ML-MDLText: um método de classificação de textos multirrótulo de aprendizado incremental

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from WebMedia 2021

Estudo de Preferências por Receitas do AllRecipes pelo Mundo

Similar lecture

ML-MDLText: um método de classificação de textos multirrótulo de aprendizado incremental

Downloads