Legal scholars have been trying to predict the outcomes of trials for a long time. In recent years, researchers have been harnessing advancements in machine learning to predict the behavior of natural and social processes. At the same time, the Brazilian judiciary faces a challenging number of new cases every year, which generates the need to improve the throughput of the justice system. Based on those premises, we trained three deep learning architectures, ULMFiT, BERT, and Big Bird, on 612,961 Federal Small Claims Courts appeals within the Brazilian 5th Regional Federal Court to predict their outcomes. We compare the predictive performance of the models to the predictions of 22 highly skilled experts. All models outperform human experts, with the best one achieving a Matthews Correlation Coefficient of 0.3688 compared to 0.1253 from the human experts. Our results demonstrate that natural language processing and machine learning techniques provide a promising approach for predicting legal outcomes. We also release the Brazilian Courts Appeal Dataset for the 5th Regional Federal Court (BrCAD-5), containing data from 765,602 appeals to promote further developments in this area.
Food biodiversity is essential for improving nutrition and reducing hunger in populations worldwide. However, in middle and low-income countries, the biodiversity of food production does not necessarily represent food consumption patterns by population. We used Brazil, one of the world's megabiodiverse countries, as a case study to investigate the following questions: what is the prevalence of consumption of biodiverse foods in Brazil, and what are the socioeconomic factors that influence their consumption throughout the country? We used data from a Brazilian representative national dietary survey to estimate the frequency of food consumption of unconventional food plants, edible mushrooms, and wild meat, in according to socioeconomic variables. Thus, we investigated the socioeconomic predictors of Unconventional Food Plants consumption using methods of Machine Learning (ML) and multiple zero-inflated Poisson (ZIP) regression. We showed that biodiverse food consumption in Brazil is low, just related by 1.3% of the population, varying in according to area, ethnicity, age, food insecurity, sex, and educational level. Our findings of low utilization of biodiversity suggest an important mismatch between the rich biodiversity of the country and its representation in the human diet.
O objetivo deste artigo é analisar o surgimento dos mecanismos de análises preditivas associados ao big data. Primeiro, apresentamos uma introdução ao conceito de “sociedade dos sensores”, que viabiliza a coleta de grandes quantidades de dados constantemente. Estes dados são o fundamento dos algoritmos de análises preditivas, sendo problemáticos para o direito, pois tais análises possibilitam que seres humanos sejam categorizados com base nos resultados de um conjunto complexo de algoritmos que não podem ser explicados de modo tradicional. Por fim, retomamos ao conceito de biopoder de Foucault, argumentando que se trata de um modelo útil para pensar o poder no cenário exposto, desde que devidamente atualizado. Concluímos que o biopoder agora é híbrido, vez que congrega uma diversidade enorme de tecnologias, sempre com a finalidade de identificar e rastrear indivíduos e grupos, bem como criar modelos preditivos de comportamento e risco.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.