Detecting COVID-19 vaccine hesitancy in India: a multimodal transformer based approach

COVID-19 vaccine hesitancy among adults in India: A primary study based on health behavior theories and 5C psychological antecedents model

Despite the significant success of India's COVID-19 vaccination program, a sizeable proportion of the adult population remains unvaccinated or has received a single dose of the vaccine. Despite the recommendations of the Government of India for the two doses of the COVID-19 vaccine and the precautionary booster dose, many people were still hesitant towards the COVID-19 full vaccination. Hence, this study aimed to identify the primary behavioral and psychological factors contributing to vaccine hesitancy. Cross-sectional data was collected via a multi-stage sampling design by using a scheduled sample survey in the Gorakhpur district of Uttar Pradesh, India, between 15 July 2022 to 30 September 2022. This study has utilized three health behavior models-the Health Belief Model (HBM), the Theory of Planned Behavior (TPB), and the 5C Psychological Antecedents of vaccination, and employed bivariate and multivariable binary logistic regression model to assess the level of vaccine hesitancy and predictive health behavior of the respondents. Results indicate that among the constructs of the HBM and 5C Antecedents models, "perceived benefits", "confidence" and "collective responsibility" showed a lesser likelihood of COVID-19 vaccine hesitancy. However, in the TPB model constructs, a 'negative attitude towards the vaccine' showed a four times higher likelihood of COVID-19 vaccine hesitancy. From the future policy perspective, this study suggested that addressing the issue of 'negative attitudes towards the vaccine' and increasing the trust or confidence for the vaccine through increasing awareness about the benefits of the vaccination in India may reduce vaccine hesitancy.

Severe acute respiratory syndrome Coronavirus-2 GenoAnalyzer and mutagenic anomaly detector using FCMFI and NSCE

In order to deepen our understanding of the virus and help guide the creation of efficient therapies, this study uses artificial intelligence tools to thoroughly explore the genetic sequences of the SARS-CoV-2 virus. The process starts by using the Fuzzy Closure Miner for Frequent Itemsets (FCMFI) on a large corpus of SARS-CoV-2 genomic sequences to reveal hidden patterns, including nucleotides base sequences, repeating motifs, and corresponding interchanges. Then, using the Nucleotide Sequence Comprehension Engine (NSCE) technique, we were able to precisely define the genomic areas for mutation analysis. Structured and unstructured proteins are both strongly impacted by virus mutations, with spike proteins that are linked to the severity of COVID-19 pneumonia being particularly affected. Notably, the Mutagenic Anomaly Detector shows a 65 % efficiency boost in computing genome mutation rates compared to conventional point mutation analysis, while GenoAnalyzer offers a remarkable 93.33 % improvement over existing approaches in recognizing common genomic sequence patterns. These results highlight the potential of FCMFI to reveal complex genomic patterns and significant insights in COVID-19 genetic sequences when combined with mutation analysis. The Mutagenic Anomaly Detector and GenoAnalyzer show promise for revealing hidden genomic patterns and precisely estimating the SARS-CoV-2 mutation rate.