Sentiment analysis in tweets: an assessment study from classical to modern word representation models.

With the exponential growth of social media networks, such as Twitter, plenty of user-generated data emerge daily. The short texts published on Twitter - the tweets - have earned significant attention as a rich source of information to guide many decision-making processes. However, their inherent characteristics, such as the informal, and noisy linguistic style, remain challenging to many natural language processing (NLP) tasks, including sentiment analysis. Sentiment classification is tackled mainly by machine learning-based classifiers. The literature has adopted different types of word representation models to transform tweets to vector-based inputs to feed sentiment classifiers. The representations come from simple count-based methods, such as bag-of-words, to more sophisticated ones, such as BERTweet, built upon the trendy BERT architecture. Nevertheless, most studies mainly focus on evaluating those models using only a small number of datasets. Despite the progress made in recent years in language modeling, there is still a gap regarding a robust evaluation of induced embeddings applied to sentiment analysis on tweets. Furthermore, while fine-tuning the model from downstream tasks is prominent nowadays, less attention has been given to adjustments based on the specific linguistic style of the data. In this context, this study fulfills an assessment of existing neural language models in distinguishing the sentiment expressed in tweets, by using a rich collection of 22 datasets from distinct domains and five classification algorithms. The evaluation includes static and contextualized representations. Contexts are assembled from Transformer-based autoencoder models that are also adapted based on the masked language model task, using a plethora of strategies.

Predicting Popularity of Video Streaming Services with Representation Learning: A Survey and a Real-World Case Study.

The Internet's popularization has increased the amount of content produced and consumed on the web. To take advantage of this new market, major content producers such as Netflix and Amazon Prime have emerged, focusing on video streaming services. However, despite the large number and diversity of videos made available by these content providers, few of them attract the attention of most users. For example, in the data explored in this article, only 6% of the most popular videos account for 85% of total views. Finding out in advance which videos will be popular is not trivial, especially given many influencing variables. Nevertheless, a tool with this ability would be of great value to help dimension network infrastructure and properly recommend new content to users. In this way, this manuscript examines the machine learning-based approaches that have been proposed to solve the prediction of web content popularity. To this end, we first survey the literature and elaborate a taxonomy that classifies models according to predictive features and describes state-of-the-art features and techniques used to solve this task. While analyzing previous works, we saw an opportunity to use textual features for video prediction. Thus, additionally, we propose a case study that combines features acquired through attribute engineering and word embedding to predict the popularity of a video. The first approach is based on predictive attributes defined by resource engineering. The second takes advantage of word embeddings from video descriptions and titles. We experimented with the proposed techniques in a set of videos from GloboPlay, the largest provider of video streaming services in Latin America. A combination of engineering features and embeddings using the Random Forest algorithm achieved the best result, with an accuracy of 87%.

Provenance-and machine learning-based recommendation of parameter values in scientific workflows.

Scientific Workflows (SWfs) have revolutionized how scientists in various domains of science conduct their experiments. The management of SWfs is performed by complex tools that provide support for workflow composition, monitoring, execution, capturing, and storage of the data generated during execution. In some cases, they also provide components to ease the visualization and analysis of the generated data. During the workflow's composition phase, programs must be selected to perform the activities defined in the workflow specification. These programs often require additional parameters that serve to adjust the program's behavior according to the experiment's goals. Consequently, workflows commonly have many parameters to be manually configured, encompassing even more than one hundred in many cases. Wrongly parameters' values choosing can lead to crash workflows executions or provide undesired results. As the execution of data- and compute-intensive workflows is commonly performed in a high-performance computing environment e.g., (a cluster, a supercomputer, or a public cloud), an unsuccessful execution configures a waste of time and resources. In this article, we present FReeP-Feature Recommender from Preferences, a parameter value recommendation method that is designed to suggest values for workflow parameters, taking into account past user preferences. FReeP is based on Machine Learning techniques, particularly in Preference Learning. FReeP is composed of three algorithms, where two of them aim at recommending the value for one parameter at a time, and the third makes recommendations for n parameters at once. The experimental results obtained with provenance data from two broadly used workflows showed FReeP usefulness in the recommendation of values for one parameter. Furthermore, the results indicate the potential of FReeP to recommend values for n parameters in scientific workflows.

[Prevalence of HIV-associated lipodystrophy in Brazilian outpatients: relation with metabolic syndrome and cardiovascular risk factors]. / Prevalência da lipodistrofia associada ao HIV em pacientes ambulatoriais brasileiros: relação com síndrome metabólica e fatores de risco cardiovascular.

Lipodystrophy in HIV-infected patients (LDHIV) affects 40-50% of HIV-infected patients, but there are no data on its prevalence in Brazil. The aim of this study was to assess the LDHIV prevalence among HIV-infected adult Brazilian individuals, as well as to evaluate LDHIV association with cardiovascular risk factors and the metabolic syndrome (MS). It was included 180 adult HIV-infected outpatients consecutively seen in the Infectology Clinic of Universidade Estadual de Londrina. Anthropometric and clinical data (blood pressure, family and personal comorbidities, duration of HIV infection/AIDS, antiretroviral drugs used, CD4+ cells, viral load, fasting glycemia and plasma lipids) were obtained both from a clinical interview as well as from medical charts. LDHIV was defined as the presence of body changes self-reported by the patients and confirmed by clinical exam. MS was defined using the NCEP-ATPIII criteria, reviewed and modified by AHA/NHLBI. A 55% prevalence of LDHIV was found. Individuals with LDHIV presented a longer infected period since HIV infection, longer AIDS duration and longer use of antiretroviral drugs. In multivariate analysis, women (p=0.006) and AIDS duration >8 years (p<0.001) were independently associated with LDHIV. Concerning MS diagnostic criteria, high blood pressure was found in 32%, low HDL-cholesterol in 68%, hypertriglyceridemia in 55%, altered waist circumference in 17% and altered glycemia and/or diabetes in 23% of individuals. Abnormal waist and hypertriglyceridemia were more common in LDHIV-affected individuals. MS was diagnosed in 36%. In multivariate analysis, the factors associated with MS were: BMI >25 kg/m(2) (p<0.001), family history of obesity (p=0.01), indinavir (p=0.001) and age >40 years on HIV first detection (p=0.002). There was a trend to higher frequency of LDHIV among patients with MS (65% versus 50%, p=0.051). LDHIV prevalence among our patients (55%) was similar to previous reports from other countries. MS prevalence in these HIV-infected individuals seems to be similar to the prevalence reported on Brazilian non-HIV-infected adults.

Prevalência da lipodistrofia associada ao HIV em pacientes ambulatoriais brasileiros: relação com síndrome metabólica e fatores de risco cardiovascular / Prevalence of HIV-associated lipodystrophy in Brazilian outpatients: relation with metabolic syndrome and cardiovascular risk factors

A lipodistrofia associada ao HIV (LAHIV) acomete 40 por cento a 50 por cento dos pacientes infectados pelo vírus, mas sua prevalência no Brasil é desconhecida. O objetivo deste trabalho foi avaliar a prevalência de LAHIV entre adultos brasileiros infectados, bem como sua relação com fatores de risco cardiovascular e síndrome metabólica (SM). Foram avaliados 180 pacientes maiores de 18 anos, infectados por HIV, atendidos no Ambulatório de Infectologia da Universidade Estadual de Londrina. Por meio de entrevista e revisão de prontuário, foram avaliados dados antropométricos, pressão arterial, antecedentes mórbidos pessoais e familiares, duração da infecção por HIV e da aids, drogas anti-retrovirais utilizadas, CD4+, carga viral, glicemia e perfil lipídico. A LAHIV foi definida como a presença de alterações corporais percebidas pelo próprio paciente e confirmadas ao exame clínico. A SM foi diagnosticada usando os critérios do Adult Treatment Panel III (NCEP-ATPIII), revistos e atualizados pela American Heart Association (AHA/NHLBI). A prevalência observada de LAHIV foi de 55 por cento. Os pacientes com LAHIV apresentaram maior duração da infecção por HIV, da aids e do uso de anti-retrovirais. Na análise multivariada, estiveram independentemente associados ao risco de LAHIV: sexo feminino (p = 0,006) e duração da aids > 8 anos (p < 0,001). Quanto aos critérios para SM, hipertensão foi detectada em 32 por cento, baixo HDL-colesterol em 68 por cento, hipertrigliceridemia em 55 por cento, cintura aumentada em 17 por cento e glicemia aumentada e/ou diabetes em 23 por cento dos indivíduos. A cintura aumentada e a hipertrigliceridemia foram mais comuns em portadores de LAHIV. A SM foi identificada em 36 por cento dos pacientes. Na análise multivariada, estiveram associados à SM: IMC > 25 kg/m² (p < 0,001), história familiar de obesidade (p = 0,01), uso de indinavir (p = 0,001) e idade > 40 anos no diagnóstico do HIV (p = 0,002). A LAHIV apresentou...

Lipodystrophy in HIV-infected patients (LDHIV) affects 40-50 percent of HIV-infected patients, but there are no data on its prevalence in Brazil. The aim of this study was to assess the LDHIV prevalence among HIV-infected adult Brazilian individuals, as well as to evaluate LDHIV association with cardiovascular risk factors and the metabolic syndrome (MS). It was included 180 adult HIV-infected outpatients consecutively seen in the Infectology Clinic of Universidade Estadual de Londrina. Anthropometric and clinical data (blood pressure, family and personal comorbidities, duration of HIV infection/AIDS, antiretroviral drugs used, CD4+ cells, viral load, fasting glycemia and plasma lipids) were obtained both from a clinical interview as well as from medical charts. LDHIV was defined as the presence of body changes self-reported by the patients and confirmed by clinical exam. MS was defined using the NCEP-ATPIII criteria, reviewed and modified by AHA/NHLBI. A 55 percent prevalence of LDHIV was found. Individuals with LDHIV presented a longer infected period since HIV infection, longer AIDS duration and longer use of antiretroviral drugs. In multivariate analysis, women (p=0.006) and AIDS duration >8 years (p<0.001) were independently associated with LDHIV. Concerning MS diagnostic criteria, high blood pressure was found in 32 percent, low HDL-cholesterol in 68 percent, hypertriglyceridemia in 55 percent, altered waist circumference in 17 percent and altered glycemia and/or diabetes in 23 percent of individuals. Abnormal waist and hypertriglyceridemia were more common in LDHIV-affected individuals. MS was diagnosed in 36 percent. In multivariate analysis, the factors associated with MS were: BMI >25Kg/m² (p<0.001), family history of obesity (p=0.01), indinavir (p=0.001) and age >40 years on HIV first detection (p=0.002). There was a trend to higher frequency of LDHIV among patients with MS (65 percent versus 50 percent, p=0.051). LDHIV prevalence...