Although Vietnamese is the 17 th most popular native-speaker language a in the world, there are not many research studies on Vietnamese machine reading comprehension (MRC), the task of understanding a text and answering questions about it. One of the reasons is because of the lack of high-quality benchmark datasets for this task. In this work, we construct a dataset which consists of 2,783 pairs of multiple-choice questions and answers based on 417 Vietnamese texts which are commonly used for teaching reading comprehension for elementary school pupils. In addition, we propose a lexicalbased MRC method that utilizes semantic similarity measures and external knowledge sources to analyze questions and extract answers from the given text. We compare the performance of the proposed model with several baseline lexical-based and neural network-based models. Our proposed method achieves 61.81% by accuracy, which is 5.51% higher than the best baseline model. We also measure human performance on our dataset and find that there is a big gap between machine-model and human performances. This indicates that significant progress can be made on this task. The dataset is freely available on our website b for research purposes.
Vietnamese is a tonal and isolated language. Its highly ambiguity makes the designing of methods for sentiment analysis being difficult. For getting the most effectiveness, the designed method has to analyze sentiment of sentences based on combining the grammar and syllable structures of Vietnamese. In this paper, a method to build a Vietnamese dataset of product reviews with many sentiment levels, including very negative, negative, neutral, positive and very positive, is proposed. This method can be scaled to a large dataset using for analyzing sentiment of product reviews. Moreover, a solution to add more grammar rules of Vietnamese into the pre-processing of sentiment analysis is also constructed. Those rules simulate the sentiment recognition of humans and help to increase the accuracy of sentiment determination. The combination of grammar rules and some methods for sentiment analysis are experimented on the Vietnamese dataset of product reviews to classify them into sentiment-levels. The testing results show that their accuracy and F-measure are improved and suitable to apply in the practical business analyzing of customer behaviors.
Recently, COVID-19 has affected a variety of real-life aspects of the world and has led to dreadful consequences. More and more tweets about COVID-19 has been shared publicly on Twitter. However, the plurality of those Tweets are uninformative, which is challenging to build automatic systems to detect the informative ones for useful AI applications. In this paper, we present our results at the W-NUT 2020 Shared Task 2: Identification of Informative COVID-19 English Tweets. In particular, we propose our simple but effective approach using the transformer-based models based on COVID-Twitter-BERT (CT-BERT) with different fine-tuning techniques. As a result, we achieve the F1-Score of 90.94% with the third place on the leaderboard of this task which attracted 56 submitted teams in total.
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.