Marcos Baez
Universidad Nacional del Este, Paraguay, Alto Parana, Faculty Member
Research Interests:
Research Interests:
Research Interests:
Text classification is one of the most common goals of machine learning (ML) projects, and also one of the most frequent human intelligence tasks in crowdsourcing platforms. ML has mixed success in such tasks depending on the nature of... more
Text classification is one of the most common goals of machine learning (ML) projects, and also one of the most frequent human intelligence tasks in crowdsourcing platforms. ML has mixed success in such tasks depending on the nature of the problem, while crowd-based classification has proven to be surprisingly effective, but can be expensive. Recently, hybrid text classification algorithms, combining human computation and machine learning, have been proposed to improve accuracy and reduce costs. One way to do so is to have ML highlight or emphasize portions of text that it believes to be more relevant to the decision. Humans can then rely only on this text or read the entire text if the highlighted information is insufficient. In this paper, we investigate if and under what conditions highlighting selected parts of the text can (or cannot) improve classification cost and/or accuracy, and in general how it affects the process and outcome of the human intelligence tasks. We study this...
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Here we present the datasets derived from our experiments on using crowdsourcing for document classification tasks. These experiments resemble a two-step process that first highlights excerpts from the text and then leverage these to... more
Here we present the datasets derived from our experiments on using crowdsourcing for document classification tasks. These experiments resemble a two-step process that first highlights excerpts from the text and then leverage these to workers for classification. Thus our experiments groups into highlighting generation and classification. For generating highlights, we leverage crowdsourcing and automatic approaches such us extractive summarization and question answering models. For our classification experiments, we consider documents from two different domains: systematic literature reviews and amazon product reviews. Specifically, we study how highlighting text passages could aid workers in judging the relevance of a document given an input question. We spec these datasets to benefit not only to study these particular problem domains but a broader set of classification problems where individual judgments from workers are scarce.<br><br>In a nutshell, the datasets represe...
Research Interests:
Research Interests:
Crowdsourcing is being increasingly adopted as a platform to run studies with human subjects. Running a crowdsourcing experiment involves several choices and strategies to successfully port an experimental design into an otherwise... more
Crowdsourcing is being increasingly adopted as a platform to run studies with human subjects. Running a crowdsourcing experiment involves several choices and strategies to successfully port an experimental design into an otherwise uncontrolled research environment, e.g., sampling crowd workers, mapping experimental conditions to micro-tasks, or ensure quality contributions. While several guidelines inform researchers in these choices, guidance of how and what to report from crowdsourcing experiments has been largely overlooked. If under-reported, implementation choices constitute variability sources that can affect the experiment's reproducibility and prevent a fair assessment of research outcomes. In this paper, we examine the current state of reporting of crowdsourcing experiments and offer guidance to address associated reporting issues. We start by identifying sensible implementation choices, relying on existing literature and interviews with experts, to then extensively ana...
Research Interests:
The genus Sapromyza was recorded for the first time at Madeira by BECKER (1908) who described the species Sapromyza indigena BECKER, 1908 and further recorded the species Sapromyza infumata Becker, 1908 and Sapromyza hyalinata (Meigen,... more
The genus Sapromyza was recorded for the first time at Madeira by BECKER (1908) who described the species Sapromyza indigena BECKER, 1908 and further recorded the species Sapromyza infumata Becker, 1908 and Sapromyza hyalinata (Meigen, 1826). CZERNY (1932) included only those records and FREY (1949) later described two new species for the island: Sapromyza hirtiloba Frey, 1949 (which was previously recorded as S. infumata by BECKER (1908), a species endemic to the Canary Islands and not present in Madeira) and Sapromyza madeirensis Frey, 1949 (it was previously recorded by BECKER (1908) as S. hyalinata). Until now, the fauna of the genus Sapromyza in Madeira was thus considered to comprise three endemic species: S. indigena, S. hirtiloba and S. madeirensis (SHATALKIN 2000). During the last few years the author has been collecting and studying this genus in the island and as results six new species have been found. Thus, the genus Sapromyza is at the present time a major example of r...
Research Interests:
Research Interests:
In this paper we explore the challenges and opportunities of designing information systems in healthcare with an emphasis on informational needs of family caregivers and work practices of professionals. We focus particularly on the... more
In this paper we explore the challenges and opportunities of designing information systems in healthcare with an emphasis on informational needs of family caregivers and work practices of professionals. We focus particularly on the context of Nursing Homes (NH), where family members and care professionals are often faced with challenging situations that can affect their ability to communicate and collaborate effectively, and thus, leading to the episodes of conflicts or mismatch of expectations. We report on two sets of user studies with staff and residents’ family members in four nursing homes, studying current information practices, factors that influence them, and explore design alternatives that could target identified issues.
Research Interests:
Systematic literature reviews (SLRs) are at the heart of evidence-based research, setting the foundation for future research and practice. However, producing good quality timely contributions is a challenging and highly cognitive... more
Systematic literature reviews (SLRs) are at the heart of evidence-based research, setting the foundation for future research and practice. However, producing good quality timely contributions is a challenging and highly cognitive endeavor, which has lately motivated the exploration of automation and support in the SLR process. In this paper we address an often overlooked phase in this process, that of planning literature reviews, and explore under the lenses of cognitive process augmentation how to overcome its most salient challenges. In doing so, we report on the insights from 24 SLR authors on planning practices, its challenges as well as feedback on support strategies inspired by recent advances in cognitive computing. We frame our findings under the cognitive augmentation framework, and report on a prototype implementation and evaluation focusing on further informing the technical feasibility.
Research Interests:
We present CrowdHub, a tool for running systematic evaluations of task designs on top of crowdsourcing platforms. The goal is to support the evaluation process, avoiding potential experimental biases that, according to our empirical... more
We present CrowdHub, a tool for running systematic evaluations of task designs on top of crowdsourcing platforms. The goal is to support the evaluation process, avoiding potential experimental biases that, according to our empirical studies, can amount to 38% loss in the utility of the collected dataset in uncontrolled settings. Using CrowdHub, researchers can map their experimental design and automate the complex process of managing task execution over time while controlling for returning workers and crowd demographics, thus reducing bias, increasing utility of collected data, and making more efficient use of a limited pool of subjects.
Research Interests:
lancehead reptile (Campbell and Lamar,1989; 2004; Caccialli, 2009) distributed in northern Argentina, central and southern Brazil, Paraguay and Uruguay (Giraudo, 2002; Campbell and Lamar, 2004; Cacciali, 2009; Cacciali et al., 2016). This... more
lancehead reptile (Campbell and Lamar,1989; 2004; Caccialli, 2009) distributed in northern Argentina, central and southern Brazil, Paraguay and Uruguay (Giraudo, 2002; Campbell and Lamar, 2004; Cacciali, 2009; Cacciali et al., 2016). This species is associated with swampy, humid and open areas (Giraudo 2002; Campbell & Lamar 2004; Cacciali 2009) and is often found near habitats deteriorated by humans (Cacciali 2009). Bothrops species feed mainly on mammals, frogs and lizards, but also birds, snakes and centipedes can be found in its diet (Martins et al., 2002). Martins et al. 2002 described and correlated Bothrops feeding habits with ecological aspects, and stated that Bothrops alternatus is a mammal specialist, this was also reported by Andrade & Abe (1999), Zanella & Cechin (2009), however the information of the mammals species preyed by this species is not available. Hence we report new data on the diet of an adult of B. alternatus. Our observation took place October 15th, 2016 i...
Research Interests:
Research Interests:
Research Interests:
In this demonstration paper we showcase an extensible and reusable pipeline for automatic paraphrase generation , i.e., reformulating sentences using different words. Capturing the nuances of human language is fundamental to the... more
In this demonstration paper we showcase an extensible and reusable pipeline for automatic paraphrase generation , i.e., reformulating sentences using different words. Capturing the nuances of human language is fundamental to the effectiveness of Conversational AI systems, as it allows them to deal with the different ways users can utter their requests in natural language. Traditional approaches to utterance paraphrasing acquisition, such as hiring experts or crowd-sourcing, involve processes that are often costly or time consuming, and with their own trade-offs in terms of quality. Automatic paraphrasing is emerging as an attractive alternative that promises a fast, scalable and cost-effective process. In this paper we showcase how our extensible and reusable pipeline for automated utterance paraphrasing can support the development of Conversational AI systems by integrating and extending existing techniques under an unified and configurable framework.
Research Interests:
Research Interests:
Research Interests:
Factors such as instructions, payment schemes, platform demographics, along with strategies for mapping studies into crowdsourcing environments, play an important role in the reproducibility of results. However, inferring these details... more
Factors such as instructions, payment schemes, platform demographics, along with strategies for mapping studies into crowdsourcing environments, play an important role in the reproducibility of results. However, inferring these details from scientific articles is often a challenging endeavor, calling for the development of proper reporting guidelines. This paper makes the first steps towards this goal, by describing an initial taxonomy of relevant attributes for crowdsourcing experiments, and providing a glimpse into the state of reporting by analyzing a sample of CSCW papers.
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
La presencia de parásitos en arenas de las plazas y parques pueden poner en riesgo la salud de los niños y de las personas que acuden a ellas. El objetivo del estudio fue la descripción de los tipos de parásitos encontrados en las arenas... more
La presencia de parásitos en arenas de las plazas y parques pueden poner en riesgo la salud de los niños y de las personas que acuden a ellas. El objetivo del estudio fue la descripción de los tipos de parásitos encontrados en las arenas de los parques públicos de Ciudad del Este. Se realizó un estudio observacional, descriptivo, de corte transversal, en 15 parques de Ciudad del Este, durante los meses de octubre del 2019 a febrero del 2020. Se recolectaron 71 muestras, bajo las condiciones existentes en el sitio, en una porción de suelo de 10 cm de largo, 10 cm de ancho y 3 cm de profundidad, se tomó aproximadamente 200 g de arena con una pala de jardín, utilizando guantes y colocándolo en una bolsa de plástico estéril y herméticamente cerrado. Para la búsqueda e identificación de los parásitos, se utilizó la técnica de flotación de Willis y centrifugación de Ritchie. De los 15 parques el 53.3%(n=8) presentaron, una muestra de arena contaminada con parásitos. De las n=71 muestras o...
Research Interests: Humanities and Art
Background. This review studies technology-supported interventions to help older adults, living in situations of reduced mobility, overcome loneliness, and social isolation. The focus is on long-distance interactions, investigating the... more
Background. This review studies technology-supported interventions to help older adults, living in situations of reduced mobility, overcome loneliness, and social isolation. The focus is on long-distance interactions, investigating the (i) challenges addressed and strategies applied; (ii) technology used in interventions; and (iii) social interactions enabled. Methods. We conducted a search on Elsevier’s Scopus database for related work published until January 2020, focusing on (i) intervention studies supported mainly by technology-mediated communication, (ii) aiming at supported virtual social interactions between people, and (iii) evaluating the impact of loneliness or social isolation. Results. Of the 1178 papers screened, 25 met the inclusion criteria. Computer and Internet training was the dominant strategy, allowing access to communication technologies, while in recent years, we see more studies aiming to provide simple, easy-to-use technology. The technology used was mostly ...
Research Interests:
Background Intergenerational relationships are beneficial for both grandparents and grandchildren. A positive grandparent-grandchild relationship can improve the psychological well-being of older adults and be a source of social support,... more
Background Intergenerational relationships are beneficial for both grandparents and grandchildren. A positive grandparent-grandchild relationship can improve the psychological well-being of older adults and be a source of social support, family history, and identity development. Maintaining meaningful interactions can be, however, a challenging endeavor, especially as life events lead to relocating geographically. Grandparents and grandchildren can have different preferences in terms of communication mediums and different assumptions about the real conversational needs of the other. Objective In this study, we will investigate the feasibility and effect of sharing memories of older adults with their grandchildren in social media. This intervention focuses on bringing snippets of the lives of the grandparents into the grandchildren’s social media feed and analyzing the potential effect on relational quality, relational investment, and conversational resources from the perspective of ...
Research Interests:
Objectives Text classification is a recurrent goal in machine learning projects and a typical task in crowdsourcing platforms. Hybrid approaches, leveraging crowdsourcing and machine learning, work better than either in isolation and help... more
Objectives Text classification is a recurrent goal in machine learning projects and a typical task in crowdsourcing platforms. Hybrid approaches, leveraging crowdsourcing and machine learning, work better than either in isolation and help to reduce crowdsourcing costs. One way to mix crowd and machine efforts is to have algorithms highlight passages from texts and feed these to the crowd for classification. In this paper, we present a dataset to study text highlighting generation and its impact on document classification. Data description The dataset was created through two series of experiments where we first asked workers to (i) classify documents according to a relevance question and to highlight parts of the text that supported their decision, and on a second phase, (ii) to assess document relevance but supported by text highlighting of varying quality (six human-generated and six machine-generated highlighting conditions). The dataset features documents from two application dom...
Research Interests:
Research Interests:
ObjectiveIn this paper we study if and under what conditions crowdsourcing can be used as a reliable method for collecting high-quality emotion labels on pictures. To this end, we run a set of crowdsourcing experiments on the widely used... more
ObjectiveIn this paper we study if and under what conditions crowdsourcing can be used as a reliable method for collecting high-quality emotion labels on pictures. To this end, we run a set of crowdsourcing experiments on the widely used IAPS dataset, using the Self-Assessment Manikin (SAM) emotion collection instrument, in order to rate pictures on valence, arousal and dominance, and explore the consistency of crowdsourced results across multiple runs (reliability) and the level of agreement with the gold labels (quality). In doing so, we explored the impact of targeting populations of different level of reputation (and cost) and collecting varying numbers of ratings per picture.ResultsThe results tell us that crowdsourcing can be a reliable method, reaching excellent levels of reliability and agreement with only 3 ratings per picture for valence and 8 per arousal, with only marginal difference between target populations. Results for dominance were very poor, echoing previous studi...
Research Interests:
Research Interests:
Research Interests:
BackgroundIntervention programs to promote physical activity in older adults, either in group or home settings, have shown equivalent health outcomes but different results when considering adherence. Group-based interventions seem to... more
BackgroundIntervention programs to promote physical activity in older adults, either in group or home settings, have shown equivalent health outcomes but different results when considering adherence. Group-based interventions seem to achieve higher participation in the long-term. However, there are many factors that can make of group exercises a challenging setting for older adults. A major one, due to the heterogeneity of this particular population, is the difference in the level of skills. In this paper we report on the physical, psychological and social wellbeing outcomes of a technology-based intervention that enable online group exercises in older adults with different levels of skills.MethodsA total of 37 older adults between 65 and 87 years old followed a personalized exercise program based on the OTAGO program for fall prevention, for a period of eight weeks. Participants could join online group exercises using a tablet-based application. Participants were assigned either to...
Research Interests:
Research Interests:
Stimulation of a physically active lifestyle among older adults is essential to health and well-being. The objective of this study was to evaluate the feasibility and user opinion of a home-based exercise program supported by a sensor and... more
Stimulation of a physically active lifestyle among older adults is essential to health and well-being. The objective of this study was to evaluate the feasibility and user opinion of a home-based exercise program supported by a sensor and tablet application for frail older adults. Community-dwelling older adults (aged ≥70 y) living in The Netherlands were recruited in 2014. Participants exercised 3 months with and 3 months without supervision from a remote coach. Feasibility was operationalized as adherence to exercise (percentage of 5 exercise bouts per week completed) and to wearing the sensor (with 70% defined as sufficient adherence) and the number of problems reported. User opinion was measured with a questionnaire addressing ease of use of the technology and opinion on the program. Twenty-one of 40 enrolled participants completed the trial. Adherence overall was 60.9% (average of 3 bouts per week). Adherence among completers (69.2%) was significantly higher than adherence amon...
Research Interests:
Research Interests:
Research Interests:
Having a sense of purpose is one of the tenets of well-being, at any age. Here, the authors review technologies that could help older adults remain active in society -- in particular, those who can't leave their home regularly or... more
Having a sense of purpose is one of the tenets of well-being, at any age. Here, the authors review technologies that could help older adults remain active in society -- in particular, those who can't leave their home regularly or easily. The authors also discuss areas that current research and practice haven't yet addressed satisfactorily.
Research Interests:
Background.Regular physical activity can substantially improve the physical wellbeing of older adults, preventing several chronic diseases and increasing cognitive performance and mood. However, research has shown that older adults are... more
Background.Regular physical activity can substantially improve the physical wellbeing of older adults, preventing several chronic diseases and increasing cognitive performance and mood. However, research has shown that older adults are the most sedentary segment of society, spending much of their time seated or inactive. A variety of barriers make it difficult for older adults to maintain an active lifestyle, including logistical difficulties in going to a gym (for some adults, leaving home can be challenging), reduced functional abilities, and lack of motivation. In this paper, we report on the design and evaluation of Gymcentral. A training application running on tablet was designed to allow older adults to follow a personalized home-based exercise program while being remotely assisted by a coach. The objective of the study was to assess if a virtual gym that enables virtual presence and social interaction is more motivating for training than the same virtual gym without social in...
Research Interests:
Research Interests:
En Canarias los panfágidos están representados por los géneros endémicos Acrostira (3 spp.) y Purpuraria (1 sp.). Son saltamontes ápteros que viven sobre sus plantas nutricias, muy quietos durante el día, siendo de noche cuando muestran... more
En Canarias los panfágidos están representados por los géneros endémicos Acrostira (3 spp.) y Purpuraria (1 sp.). Son saltamontes ápteros que viven sobre sus plantas nutricias, muy quietos durante el día, siendo de noche cuando muestran mayor actividad. Tienen gran tamaño, pero su coloración críptica y las posturas agazapadas que adoptan sobre los tallos de las plantas hacen de ellos insectos muy difíciles de observar, por lo que el conocimiento de sus poblaciones y biología es escaso. Acrostira euphorbiae es un endemismo de La Palma, incluido en el Catálogo Nacional de especies animales en peligro de extinción. Hasta el presente estudio, sólo se conocía de la localidad típica situada en El Remo (Los Llanos de Aridane), un pequeño tabaibal de aproximadamente una hectárea. A lo largo de un año se ha buscado nuevas poblaciones sistemáticamente, y los ejemplares encontrados fueron marcados y ubicados con GPS. Los muestreos realizados por toda la isla confirman que esta especie, propia ...