Brazil

As commodities milho e soja são produtos consumidos em grande escala no mundo. As flutuações nos preços de mercado têm efeitos de longo alcance sobre os consumidores, agricultores e processadores de grãos. Assim, a previsão dos preços desses grãos tem atraído significativa atenção dos pesquisadores. Os modelos de previsão geralmente usam dados quantitativos de série temporal. No entanto, fatores qualitativos externos podem influenciar os dados em séries temporais, como eventos políticos, crises econômicas e o mercado de câmbio estrangeiro. Essas informações não são explícitas nos dados da série temporal e esses fatores podem influenciar a previsão dos valores das variáveis. Os dados textuais extraídos de notícias, fóruns e redes sociais podem ser uma fonte de conhecimento sobre fatores externos e potencialmente úteis para modelos de previsão de séries temporais. Alguns estudos apresentam técnicas de mineração de texto para combinar dados textuais com séries temporais. Porém, as representações existentes apresentam algumas limitações, como a maldição da dimensionalidade e atributos ineficazes. Este trabalho aplica métodos de pré-processamento em séries temporais e usa representações combinadas com dados textuais para prever o preço futuro do milho e da soja. Os resultados indicam que os métodos utilizados podem ser uma alternativa para melhorar o desempenho de previsão em tarefas de regressão.

BRACIS 2020

Forecasting future corn and soybean prices: an analysis of the use of textual information to enrich time-series.

time-series; text mining; forecasting; agricultural commodities

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.

technical paper

One of the main challenges for cell segmentation is to separate overlapping cells, which is also a challenging task for cytologists. Here we propose a method that combines different algorithms for cervical cell segmentation of Pap smear images and searches for the best result underlying the maximization of a similarity coefficient. We carried out experiments with three state-of-the-art segmentation algorithms on images with clumps of cervical cells. We extracted features such as coefficient of variation and overlapping ratios for each cell grouping and selected the most appropriate algorithm to segment each cell clump. For decision criterion, we identified the cell clumps of the training dataset and calculated the mentioned features. We segmented each clump by the algorithms and reckoned the Dice measure from each segmentation. Finally, we used the kNN classifier to predict the best algorithm among neighboring k-clumps by choosing the one with the largest number of wins. We validated our proposal on multifocal cervical cell images and obtained an average Dice around 76.6% without using a threshold value. These results demonstrated that the proposed ensemble of segmentation algorithms is promising and suitable for cervical cell image segmentation.

Ensemble of algorithms for multifocal cervical cell image segmentation

Data augmentation is a widely adopted method for improving model performance in image classification tasks. Although it still not as ubiquitous in Natural Language Processing (NLP) community, some methods have already been proposed to increase the amount of training data using simple text transformations or text generation through language models. However, recent text classification tasks need to deal with domains characterized by a small amount of text and informal writing, e.g., Online Social Networks content, reducing the capabilities of current methods. Facing these challenges by taking advantage of the pre-trained language models, low computational resource consumption, and model compression, we proposed the PRE-trained Data AugmenTOR (PREDATOR) method. Our data augmentation method is composed of two modules: the Generator, which synthesizes new samples grounded on a lightweight model, and the Filter, that selects only the high-quality ones. The experiments comparing Bidirectional Encoder Representations from Transformer (BERT), Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM) and Multinomial Naive Bayes (NB) in three datasets exposed the effective improvement of accuracy. It was obtained 28.5% of accuracy improvement with LSTM on the best scenario and an average improvement of 8% across all scenarios. PREDATOR was able to augment real-world social media datasets and other domains, overcoming the recent text augmentation techniques.

Pre-trained Data Augmentation for Text Classification

In Natural Language Processing, stance detection is the computational task of deciding whether a piece of text expresses a favourable or unfavourable attitude (or stance) towards a given topic. Stance detection may be divided into two subtasks: deciding whether a piece of text conveys any stance towards the target topic and, once we have established that the text does convey a stance, determining its polarity (e.g., favourable or unfavourable) towards the target. Both tasks - hereby called stance recognition and (stance) polarity classification - are the focus of the present work. Taking as a basis a corpus of 13.7k tweets in the Brazilian Portuguese language, and which conveys stances towards five moral issues (abortion legislation, death penalty, drug legalisation, lowering of criminal age, and racial quotas at universities), we compare a number of long short-term memory (LSTM) and bidirectional LSTM (BiLSTM) models for stance recognition and polarity classification. In doing so, the two tasks are addressed both independently and as a joint model. Results suggest that the use of BiLSTM models with attention mechanism outperform the alternatives under consideration, and pave the way for more comprehensive studies in this domain.

Twitter Moral Stance Classification using Long Short-Term Memory Networks

Deep language models, like ELMo, BERT and GPT, have achieved impressive results on several natural language tasks. These models are pretrained on large corpora of unlabeled general domain text and later supervisedly trained on downstream tasks. An optional step consists of finetuning the language model on a large intradomain corpus of unlabeled text, before training it on the final task. This aspect is not well explored in the current literature. In this work, we investigate the impact of this step on named entity recognition (NER) for Portuguese legal documents. We explore different scenarios considering two deep language architectures (ELMo and BERT), four unlabeled corpora and three legal
NER tasks for the Portuguese language. Experimental findings show a significant improvement on performance due to language model finetuning on intradomain text. We also evaluate the finetuned models on two general-domain NER tasks, in order to understand whether the aforementioned improvements were really due to domain similarity or simply due to more training data. The achieved results also indicate that finetuning on a legal domain corpus hurts performance on the general-domain NER tasks. Additionally, our BERT model, finetuned on a legal corpus, significantly improves on the state-of-the-art performance on the LeNER-Br corpus, a Portuguese language NER corpus for the legal domain.

A Study on the Impact of Intradomain Finetuning of Deep Language Models for Legal Named Entity Recognition in Portuguese

The development of treatments based on the patient’s individual characteristics has been an emergent medical approach. The objective is to improve individual responses and overall survival. Thus, there is a need for computational tools able to identify and describe subgroups of patients for which the survival response significantly differs from the overall behaviour. However, there are few algorithms that address this matter. The majority of works of literature aim at building predictive models rather than understanding the characteristics that delineate subgroups with unusual survival. The approaches that provide understanding on factors that interfere in the survival behaviour usually resort to the stratification of the data based on previously known variable’s interactions, lacking the ability to shed light into new, possibly unknown, interactions. In contrast to the existent predictive approaches, we propose the use of supervised descriptive pattern mining in order to discover local patterns able to describe subsets of patients that present unusual survival behaviour. In this presentation, we present the ESM-AM (Exceptional Survival Model Ant Miner) algorithm, an Exceptional Model Mining approach to the discovery of subgroups with exceptional survival functions that explores the use of ant-colony optimization as search heuristic for the pattern mining task.

Exceptional Survival Model Mining

The most recent successful time series prediction models are
a combination of three elements: traditional stochastic models, machine
learning models and signal processing techniques. CEEMDAN-LSTM
models have combined empirical mode decomposition and long shortterm memory neural networks to achieve state-of-the-art results for financial data. In this work, we propose a generalized CEEMDAN-LSTM
architecture for time series forecasting capable of dealing with exogenous features as input, and the consequences of input data growth, such
as convergence difficulties. Our model was applied to time series from
10 of the most liquid Brazilian stocks, and results show that accuracy
is overall improved when compared to the original single feature input
CEEMDAN-LSTM architecture.

Downloads

Next from BRACIS 2020

Ensemble of algorithms for multifocal cervical cell image segmentation

Similar lecture

Previsão do preço futuro do milho e da soja: Uma análise do uso de informação textual para enriquecer séries temporais.

.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 BRACIS 2020

Ensemble of algorithms for multifocal cervical cell image segmentation

Similar lecture

Previsão do preço futuro do milho e da soja: Uma análise do uso de informação textual para enriquecer séries temporais.

Downloads