Brenda Salenave Santana, Vinicius Woloszyn, Leandro Krug Wives
In this work, we propose an analysis of the presence of gender bias associated with professions in Portuguese word embeddings. The objective of this work is to study gender implications related to stereotyped professions for women and men in the context of the Portuguese language.
Vinicius Woloszyn, Guilherme Medeiros Machado, Leandro Krug Wives et al.
Automatic Text Summarization strategies have been successfully employed to digest text collections and extract its essential content. Usually, summaries are generated using textual corpora that belongs to the same domain area where the summary will be used. Nonetheless, there are special cases where it is not found enough textual sources, and one possible alternative is to generate a summary from a different domain. One manner to summarize texts consists of using a graph model. This model allows giving more importance to words corresponding to the main concepts from the target domain found in the summarized text. This gives the reader an overview of the main text concepts as well as their relationships. However, this kind of summarization presents a significant number of repeated terms when compared to human-generated summaries. In this paper, we present an approach to produce graph-model extractive summaries of texts, meeting the target domain exigences and treating the terms repetition problem. To evaluate the proposition, we performed a series of experiments showing that the proposed approach statistically improves the performance of a model based on Graph Centrality, achieving better coverage, accuracy, and recall.
Augusto Dias Pereira dos Santos, Leandro Krug Wives, Luis Otavio Alvares
The increasing use of social networks generates enormous amounts of data that can be used for many types of analysis. Some of these data have temporal and geographical information, which can be used for comprehensive examination. In this paper, we propose a new method to analyze the massive volume of messages available in Twitter to identify places in the world where topics such as TV shows, climate change, disasters, and sports are emerging. The proposed method is based on a neural network that is used to detect outliers from a time series, which is built upon statistical data from tweets located on different political divisions (i.e., countries, cities). The outliers are used to identify topics within an abnormal behavior in Twitter. The effectiveness of our method is evaluated in an online environment indicating new findings on modeling local people's behavior from different places.