Improved mask detection based on YOLOv5s

COVID-19 epidemic is not over. The correct wearing of masks can effectively prevent the spread of the virus. Aiming at a series of problems of existing mask-wearing detection algorithms, such as only detecting whether to wear or not, being unable to detect whether to wear correctly, difficulty in detecting small targets in dense scenes, and low detection accuracy, It is suggested to use a better algorithm based on YOLOv5s. It improves the generalization and transmission performance of the model by changing the ACON activation function. Then Bifpn is used to replace PAN to effectively integrate the target features of different sizes extracted by the network. Finally, To enable the network to pay attention to a wide area, CA is introduced to the backbone. This embeds the location information into the channel attention. © 2023 SPIE.

A Cross-Corpus Deep Learning Approach to Social Media Emotion Classification

Emotion Detection refers to the identification of emotions from contextual data in the form of written text, such as comments, posts, reviews, publications, articles, recommendations, conversations, and so on. Because of the Internet's exponential uptake and the recent coronavirus outbreak, social media platforms have become a crucial means of sharing thoughts and ideas throughout the entire globe, creating rapid data growth through users' contributions on various platforms. The necessity to acquire knowledge of their behaviors is a matter of great concern for both internet safety and privacy. In this study, we categorize emotional sentiments using deep learning models along with hybrid approaches such as LSTM, Bi-LSTM, and CNN+LSTM. When compared to existing state-of-the-art methods, the experiments showed that the suggested strategy is more robust and achieves an expressively higher quality of emotion detection with an accuracy rate of 94.16%, including strong F1-scores on complex and difficult emotion categories such as Fear (93.85%) and Anger (94.66%) through CNN+LSTM. © 2022 IEEE.

Rate Insight: A Comparative Study on Different Machine Learning and Deep Learning Approaches for Product Review Rating Prediction in Bengali Language

In this contemporary era of digital marketing, ecommerce has emerged as one of the most preferred methods for day-to-day shopping. Ever since the COVID-19 pandemic, online shopping behavior has forever changed to less or no human-to-human interaction. As a result, it is getting more difficult for e-commerce enterprises to observe and evaluate market trends, particularly when done through consumer behavior analysis. To identify behavioral patterns and customer review-rating discrepancies, extensive analysis of product reviews is a substantial research field. Lack of benchmark corpora and language processing techniques, predicting review ratings in Bengali has become increasingly problematic. This paper thoroughly analyzes the approach to product review rating prediction for Bengali text reviews exploiting our own constructed dataset that was collected from an e-commerce website called DarazBD1. We acquired product reviews with labels known as ratings of five sentiment classes, from "1"to "5". It is noteworthy that we established a well-balanced dataset using our automated scraping system and a significant amount of time and effort is spent to maintain quality standards through the human annotation process. Exploration of multiple approaches to machine learning models such as logistic regression, random forest, multinomial naïve Bayes, and support vector machine, the best classification accuracy score of 78.63% is achieved by SVM. Subsequently, using Word2Vec, FastText, and GloVe embeddings with three deep neural network(DNN) architectures: CNN, Bi-LSTM, and a combination of CNN and Bi-LSTM, CNN+Bi-LSTM gave the highest accuracy score of 75.25% among the DNN architectures. © 2022 IEEE.

Sentiment Analysis of Government Policy Regarding PPKM on Twitter Using LSTM

The Policy of PPKM Covid from the government has become a popular topic to be discussed among the public, especially on Twitter. Due to the many responses or opinions about the PPKM that has been implemented by the government in Indonesia. Sentiment Analysis is the basis for research on the issue of Indonesian PPKM by using a deep learning model, namely LSTM. The data collection of tweets is obtained through crawling the data of Twitter API using the 'snscrape' module with the keyword 'PPKM COVID' and the target data is 15,001 tweets. The data is processed and divided into two parts become 80% training data, 20% testing data and using the GRU, BiLSTM and RNN comparison models. Accuracy performance obtained from the four models include LSTM 90%, GRU 89%, BiLSTM 90% and RNN 85%. The comparison of the best accuracy results is obtained from the LSTM and BilSTM models. Furthermore, the result of sentiment obtained a high percentage for negative sentiment with a total percentage of 54.6%, while the positive sentiment had a percentage of 37.0% and neutral sentiment is 8.5%. © 2022 IEEE.

COVID-19 Cases Prediction Using Different LSTM Models and Comparison of Effectiveness of Different Models

Large infectivity and transmissibility of COVID-19 caused severe damage to the economy, education and health of many countries. Due to the increasing number of COVID-19 cases in the world, some predictive methods are therefore needed to forecast the number of cases of COVID-19 in the future. Long short-term memory (LSTM) predicts the correlation between confirmed cases and predicts COVID-19 spread over time. The system shall be trained using training data containing confirmed cases. Various parameters considered are the no of positive cases, the number of recovered cases and the no of deaths every day. LSTM models in different types are evaluated for the time series forecasting confirmed cases, deaths and recovery and the accuracy of the prediction is compared. Different LSTM models like bidirectional LSTM, Gated Recurrent unit, W-LSTM and simple LSTM are helps to predict the no of cases in each country. Model performance is measured using the root mean square error, mean absolute percentage error and r2-score indices. Proposed method can be used to predict other types of pandemics for improved planning. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Bi(III) I-Complexes of Porphyrins for Biomedicine: Synthesis and Spectral-Optical Properties

Bismuth complexes of porphyrins are of interest for IR luminescence diagnostics of cancer, since rather intense emission bands in the range of 800–920 nm have been found. In connection with the COVID-19 pandemic, bismuth compounds are also of interest in the treatment of coronavirus infection. Bismuth complexes of porphyrins of various spatial configurations have been synthesized, and several spectral-optical properties have been investigated. The influence of various substituents on the spectral characteristics was evaluated by methods of studying electronic absorption spectra, luminescence spectra, IR-, and 1H NMR spectroscopy.