Brazil

Automatically dealing with Natural Language User-Generated Con-tent (UGC) is a challenging task of utmost importance, given the amount of information available over the web. We present in this paper an effort on building tokenization and Part of Speech (PoS) tagging systems for tweets in Brazilian Portuguese, following the guidelines of the Universal Dependencies (UD)project. We propose a rule-based tokenizer and the customization of current state-of-the-art UD-based tagging strategies for Portuguese, achieving a 98%f-score for tokenization, and a 95% f-score for PoS tagging. We also introduceDANTEStocks, the corpus of stock market tweets on which we base our work, presenting preliminary evidence of the multi-genre capacity of our PoS tagger.

BRACIS 2021

Universal Dependencies for Tweets in Brazilian Portuguese: Tokenization and Part of Speech Tagging

pos-tag

twitter

tokenization

**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. **STIL** is the bi-annual Language Technology event.
All events are supported by the Brazilian Computer Society – Special Interest Groups: **CE-IA**, **CE-IC** and **CE-PLN**.

To access the 10th Brazilian Conference on Intelligent Systems (**BRACIS**), you need to register for the conference.  Please visit the Conference official site to register

[Register now](http://c4ai.inova.usp.br/bracis2021/)

Register now

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. 
STIL is the bi-annual Language Technology event.
All events are supported by the Brazilian Computer Society – Special Interest Groups: CE-IA, CE-IC and CE-PLN

technical paper

Morphological neural networks represent a class of artificial neural networks whose neurons perform an operation from mathematical morphology followed by the application of an activation function. This paper provides a comparative study of different approaches that use morphological neural networks. Specifically, according to the training rule, we review incremental approaches, approaches based on maximum descent methods, extreme learning machines, and convex-concave optimization procedures. Computational experiments showed that, on average, the reduced dilation-erosion perceptron with bagging and ensemble strategies had better results in several binary classification problems.

A Comparative Study on Morphological Neural Networks for Binary Classification

O objetivo desse trabalho é analisar conjuntos de variáveis que podem ter mais influência na previsão do resultado de uma partida de futebol, utilizando técnicas como cálculos de probabilidade e algoritmos de previsão, com a intenção de obter lucros em apostas. As técnicas utilizadas para o desenvolvimento foram redes bayesianas, com o algoritmo Naïve Bayes, e a de probabilidade, baseada no cálculo de Poisson. Os dados utilizados para treinamento foram do Campeonato Brasileiro, entre 2010 a 2017, sendo considerados os dados dos anos de 2018 e 2019 para testes. Os principais resultados atingidos foram de 53% de acerto do resultado de uma partida e as principais variáveis envolvidas foram força de ataque e força de defesa.

Análise de Variáveis em Partidas de Futebol: Previsão de Resultados com Naïve Bayes e Poisson

The Sustainable Development Goals (SDGs) are part of a global effort to reduce the impacts of climate change, promoting social justice and economic growth. 
The United Nations provides a database with hundreds of indicators to track the SDGs since 2016 for a total of 302 regions. 
This work aims to assess which countries are in a similar situation regarding sustainable development. 
Principal Component Analysis was used to reduce the dimension of the dataset and k-means algorithm was used to cluster countries according to their SDGs indicators. For the years of 2016, 2017 and 2018 were obtained 11, 13 and 11 groups, respectively. This paper also analyses clusters changes throughout the years.

A Comparative Analysis of Countries' Performance According to SDG Indicators based on Machine Learning

Companies created for money-laundering or as a means for tax-evasion are harmful to the country’s economy and society. This problem isusually tackled by governmental agencies by having officials to pore over com-panies’ financial data and to single out those that exhibit fraudulent behavior. Such work tends to be slow-paced and tedious. This paper proposes a machinelearning-based system capable of classifying whether a company is likely to beinvolved in fraud or not. Based on financial and tax data from various compa-nies, four different classifiers –k-Nearest Neighbors, Random Forest, SupportVector Machine (SVM), and a Neural Network – were trained and then used toindicate fraud. The best-performing model achieved a macro-averaged F1-scoreof 92.98% with the Random Forest.

A Machine Learning-based System for Financial Fraud Detection

DevOps practices have been increasingly used by software engineering teams in order to improve development stages. In processes that involve machine learning (ML), DevOps can also be applied to deploy machine learning models into production - a practice also known as MLOps. This systematic mapping aims to understand how DevOps has been applied to machine learning processes and what challenges are faced. Most of the 15 articles selected make use of CI/CD practices and propose architectures for the implementation of ML models. The biggest challenges are the characteristics inherent to the ML models and resistance to change.

Applying DevOps to Machine Learning Processes: A Systematic Mapping

Collaborative Filtering (CF) can be understood as the process of predicting the preferences of users and deriving useful patterns by studying their activities. In the survey context, it can be used to predict answers to questions as combinations of other available answers. In this paper, we aim to test five CF-based algorithms (item-item, iterative matrix factorization, neural collaborative filtering, logistic matrix factorization, and an ensemble of them) to estimate scores in four survey applications (checkpoints) composed of 700,000 employee’s ratings. These data have been collected from 2019 to 2020 by a large Brazilian tech company with more than 10,000 employees. The results show that collaborative filtering approaches provide relevant alternatives to score questions of surveys. They provided good quality estimates. This result can be further explored to eventually reduce the size of questionnaires, avoiding burden phenomena faced by respondents when dealing with large surveys.

Comparative Analysis of Collaborative Filtering-Based Predictors of Scores in Surveys of a Large Company

In Brazil, several minority languages suffer a serious risk of extinction. The appropriate documentation of such languages is a fundamental step to avoid that. However, for some of those languages, only a small amount of text corpora is digitally accessible.
Meanwhile, there are many issues related to the identification of indigenous languages, which may help to identify key similarities among them, as well as to connect related languages and dialects.
Therefore, this paper proposes to study and automatically classify 26
neglected Brazilian native languages, considering a small amount of training data, under a supervised and unsupervised setting. Our findings indicate that the use of machine learning models in the analysis of Brazilian Indigenous corpora is very promising, and we hope this work encourage more research on this topic in the next years.

Analysis of a Brazilian Indigenous corpus using machine learning methods

Over the past few years, information technology has been transforming the legal world, automating processes and, consequently, reducing the time
needed to create and analyse digital legal documents. One of the most studied
problems in this area is the recognition of named entities (NER) in unstructured
texts. Previous work has not addressed the detection of legal entities through the
application of models based on neural networks available in natural language
processing libraries. In this article, the use of the libraries Spacy and FLAIR
were analyzed in the context of NER in initial petitions. The models were trained with pre-defined architectures and evaluated in two corpora, one of them
developed in the scope of this work. The results obtained with these experiments
demonstrated good results with both platforms Spacy and FLAIR, with superior
performance when adopting BiLSTM-CRF with FLAIR embeddings.


Named entity recognition in Portuguese legal documents using deep learning

The initial petition is one of the most important components of a civil litigation process. Automating the analysis of these documents might reduce the time necessary for the postulatory phase's conclusion. The parties qualification body is the section in which are exposed the informations about the entities involved in the process. This paper suggests the employment of named entity extraction techniques on the problem of information extraction and recognition on initial petitions. With this in mind, was created a part description body corpora extracted from Brazilian courts. Seven BiLSTM-CRF models with distinct combinations of vector representations of words were trained, evaluated, and compared to investigate their effects on the performance of an algorithm with that architecture and, in this way, improve the recognition of legal entities in legal texts. Unlike other works based on BiLSTM-CRF for NER tasks in the legal domain, this research emphasizes not the architectures employed, but rather the text representation methods used. The experiments performed with the developed corpus show that the stacking of character, word, and pooled FLAIR embeddings is the preferred combination to extract the best possible performance from BiLSTM-CRF hybrid models.

A comparative analysis of text embedding approach to extract named entities in Portuguese legal documents

Textual information, despite being digital, are not computationally structured, requiring the use of techniques to structure and extract information. This work aims to evaluate machine learning NER (Named entity recognition) tools for the Brazilian and European Portuguese language variants. The Apache OpenNLP, Stanford CoreNLP and spaCy tools were selected for the evaluation. The HAREM corpus was used to train and evaluate the generated machine learning model; a tool was developed to preprocess the HAREM corpus. Two comparisons were performed: a general comparison and between the Portuguese variants. It was found that variants can confuse the training and evaluation process of NER models.

Premium content

Universal Dependencies for Tweets in Brazilian Portuguese: Tokenization and Part of Speech Tagging

Downloads

Next from BRACIS 2021

A Comparative Study on Morphological Neural Networks for Binary Classification

Similar lecture

Identificando sintomas de depressão em postagens do Twitter em português do Brasil