Lexicon-based sentiment analysis to detect opinions and attitude towards COVID-19 vaccines on Twitter in Italy.

The paper proposes a methodology based on Natural Language Processing (NLP) and Sentiment Analysis (SA) to get insights into sentiments and opinions toward COVID-19 vaccination in Italy. The studied dataset consists of vaccine-related tweets published in Italy from January 2021 to February 2022. In the considered period, 353,217 tweets have been analyzed, obtained after filtering 1,602,940 tweets with the word "vaccin". A main novelty of the approach is the categorization of opinion holders in four classes, Common users, Media, Medicine, Politics, obtained by applying NLP tools, enhanced with large-scale domain-specific lexicons, on the short bios published by users themselves. Feature-based sentiment analysis is enriched with an Italian sentiment lexicon containing polarized words, expressing semantic orientation, and intensive words which give cues to identify the tone of voice of each user category. The results of the analysis highlighted an overall negative sentiment along all the considered periods, especially for the Common users, and a different attitude of opinion holders towards specific important events, such as deaths after vaccination, occurring in some days of the examined 14 months.

Real Time Web-based Facemask Detection

The widespread spread of the Covid-19 virus in 2020-2021 is very worrying for all people around the world, coupled with the spread of a new variant of the Covid-19 virus, which is more aggressive and easily transmitted, causing public unrest about when this pandemic will end. The policy of using masks to reduce the spread of the virus has been made to minimize the spread. But even if there is a policy, there are still people who don't want to wear masks. Therefore, a mask detection system is needed to help differentiate whether someone uses a mask or not by displaying alerts in a form of web application. This research was conducted using several data augmentation techniques to increase the variation of the data to be used before training the algorithm model using the Convolutional Neural Network (CNN) algorithm with MobileNetV2 and VGG19 architectures. Both models are then evaluated where the architecture with the best performance will be implemented in the form of a web application. The accuracy of both models was compared, with the result of MobileNetV2 being 99% accurate and VGG19 being 98%. MobileNetV2 as the model that has the best accuracy value will be implemented in the form of a web application using the Haar Feature-Based Cascade to detect masks. The web application will be publicly accessed local at Universitas Multimedia Nusantara. © 2022 IEEE.

A hand rubbing classification model based on image sequence enhanced by feature-based confidence metric.

Hand hygiene is critical for declining the spread of viruses and diseases. Over recent years, it has been globally known as one of the most effective ways against COVID-19 outbreak. The World Health Organization (WHO) has suggested a 12-step guideline for hand rubbing. Due to the importance of this guideline, several studies have been conducted to measure compliance with it using Computer Vision. However, almost all of them are based on processing single images as input, referred to as baseline models in this paper. This study proposes a sequence model in order to process sequences of consecutive images as input. The model is a mixture of Inception-ResNet architecture for spatial feature extraction and LSTM for detecting time-series information. After training the model on a comprehensive dataset, an accuracy of 98.99% was achieved on the test set. Compared to the best baseline models, the proposed sequence model is correspondingly about 1% and 4% better in terms of accuracy and confidence, though 3 times slower in inference time. Furthermore, this study demonstrates that the accuracy metric is not necessarily adequate to compare different models and optimize their hyperparameters. Accordingly, the Feature-Based Confidence Metric was utilized in order to provide a more pleasing comparison to discriminate the proposed sequence model with the best baseline model and optimize its hyperparameters.

Sentimental Analysis of Streaming COVID-19 Twitter Data on Spark-Based Framework

Social media has become an inevitable part of human’s daily life enabling people to express their opinion, sentiments, and ideologies. During this COVID-19 pandemic when the whole world went into a lockdown situation, Twitter served as an outlet for people to express their emotions. This work proposes streaming the real-time Twitter data on COVID-19 using Twitter API and handling the streaming big data using the Apache Spark framework. Here the fake account detection to detect the non-legit accounts present in the streamed data was accomplished by the proposed feature-based algorithm which attain overall accuracy of 98.74%. This constructed fake account detection model filters out the genuine accounts from the API streamed Twitter data. Sentimental analysis on these genuine Twitter accounts is performed by modifying the Natural Language Processing (NLP) state-of-art algorithm called Bidirectional Encoder Representations from Transformers (BERT). The proposed method achieved 88.30% of classification accuracy rate by concatenation of the pooled NN layer with the influential feature. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

How to Churn Deep Contextual Models?

This paper searches for optimal ways of employing deep contextual models to solve practical natural language processing tasks. It addresses the diversity in the problem space by utilizing a variety of techniques that are based on the deep contextual BERT (Bidirectional Encoder Representation from Transformer) model. A collection of datasets on COVID-19 social media misinformation is used to capture the challenge in the misinformation detection task that arises from small labeled data, noisy labels, out-of-distribution (OOD) data, fine-grained & nuanced categories, and heavily-skewed class distribution. To address this diversity, both domain-agnostic (DA) and domain-specific (DS) BERT pretrained models (PTMs) for transfer learning are examined via two methods, i.e., fine-tuning (FT) and extracted feature-based (FB) learning. The FB is implemented using two approaches: non-hierarchical (features extracted from a single hidden layer) and hierarchical (features extracted from a subset of hidden layers are first aggregated, then passed to a neural network for further extraction). Results obtained from an extensive set of experiments show that FB is more effective than FT and that hierarchical FB is more generalizable. However, on the OOD data, the deep contextual models are less generalizable. It identifies the condition under which DS PTM is beneficial. Finally, bigger models may only add an incremental benefit and sometimes degrade the performance. © 2021 ACM.

A Bag-of-Features (BoF) Based Novel Framework for the Detection of COVID-19

Novel coronavirus (COVID-19) is a hazardous virus. Initially, detected in China and spread worldwide, causing several deaths. Over time, there have been several variants of COVID-19, we have grouped all of them into two major categories. The categories are known to be variants of concern and variants of interest. Talking about the first of these two, it is very dangerous, and we need a system that can not only detect the disease but also classify it without physical interaction with a patient suffering from COVID-19. This paper proposes a Bag-of-Features (BoF) based deep learning framework that can detect as well as classify COVID-19 and all of its variants as well. Initially, the spatial features are extracted with deep convolutional models, while hand-crafted features have been extracted from several hand-crafted descriptors. Both spatial and hand-crafted features are combined to make a feature vector. This feature vector feeds the classifier to classify different variants in respective categories. The experimental results show that the proposed methodology outperforms all the existing methods. © 2021 IEEE.

Feature Based Statistical Model of Employee Productivity with Real Time Checked Data

The COVID-19 pandemic has led to a decentralization of the workforce in many industries. Due to the stay-at-home orders to control the spread of the virus, many are working from home. Even though modern technological advancements have helped some companies adapt to this new norm, many others are still scrambling to find the best way to remotely manage employees and accommodate their needs. Our research shows that the current challenges organizations face in managing their human capital are like the ones they face due to workplace demographic changes. This study focuses on analyzing those challenges and how human competency can be unlocked and developed to encourage sustainable autonomous working in an office, at home, or during frequent traveling. This study investigates the challenges faced by both organizations and employees, and presents a new business model that helps with the sustainable use of human resources and improves employee efficiency. © 2021 IEEE.

COVID-19 Lung CT Images Recognition: A Feature-Based Approach

The SARS-CoV-2 is quickly spreading worldwide resulting in millions of infection and death cases. As a consequence, it is increasingly important to diagnose the presence of COVID-19 infection regardless of the technique applied. To this end, this work deals with the problem of COVID-19 classification using Computed Tomography (CT) images. Precisely, a new feature-based approach is proposed by exploiting axial CT lung acquisitions in order to differentiate COVID-19 versus healthy Computed Tomography (CT) images. In particular, first-order statistical measures as well as numerical quantities extracted from the autocorrelation function are investigated with the aim to provide an efficient classification process ensuring satisfactory performance results. © 2021, Springer Nature Switzerland AG.

Understanding Art with AI: Our Research Experience

Artificial Intelligence solutions are empowering many fields of knowledge, including art. Indeed, the growing availability of large collections of digitized artworks, coupled with recent advances in Pattern Recognition and Computer Vision, offer new opportunities for researchers in these fields to help the art community with automatic and intelligent support tools. In this discussion paper, we outline some research directions that we are exploring to contribute to the challenge of understanding art with AI. Specifically, our current research is primarily concerned with visual link retrieval, artwork clustering, integrating new features based on contextual information encoded in a knowledge graph, and implementing these methods on social robots to provide new engaging user interfaces. The application of Information Technology to fine arts has countless applications, the most important of which concerns the preservation and fruition of our cultural heritage, which has been severely penalized, along with other sectors, by the ongoing COVID pandemic. On the other hand, the artistic domain poses entirely new challenges to the traditional ones, which, if addressed, can push the limits of current methods to achieve better semantic scene understanding. © 2021 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0). CEUR Workshop Proceedings (CEUR-WS.org)

Analysis of Social Trends Related to COVID-19 Pandemic Utilizing Social Media Data

In this paper, we propose a method for analyzing social trends related to the coronavirus disease (COVID-19) pandemic by using social media data. The proposed method reveals that there is a correlation between tweets posted by users in Twitter and the number of infected people in a certain period. Specifically, the proposed method extracts tweet features based on the relationship between the contents and keywords of tweets. Compared to the previous approaches which focus only on the number of tweets, the proposed method can capture more richer information. Therefore, high correlation between the tweet features and the number of infected people can be obtained. For analyzing the tweets related to COVID-19, the proposed method consider not the number of tweets but the contents of the tweets. This is the main contribution of this paper. We verify the effectiveness of the proposed method through experiments on real-world datasets. © 2021 IEEE.